Home        The Boat        Photo Essay        Contact

Failure Fiasco

Equipment problems onboard Eva and the comedy of failure mode analysis.

Raw Water Pump
Navigation Lights
Inverters and Chargers
Consumer Electronics
Rig and Gear
Engine and Electrical
Propeller and Shaft
Marine Electronics
LFP House Bank
Truck and Trailer

Windlass - Manual to Electric
The old Simpson Laurence two speed manual windlass was reliable and very powerful. What it was not was efficient, and an enormous amount of one's effort went into overcoming friction. When we got the boat the windlass was full of grease, which made the situation even worse. Cleaning the grease out and filling the windlass with oil helped considerably to reduce friction, but the manual windlass was still so inefficient that it was easier to simply pull the anchor up by hand in less than about 50 feet of water. The replacement was a very nice Lewmar Pro700 horizontal electric windlass that is quite efficient despite it's simple brush type permanent magnet motor and no control circuitry. The maximum pulling force of the little Lewmar is sometimes disappointingly low, but for all normal use the anchor comes zipping right up. The only problem with the Lewmar is that the cylindrical keys that keep the gypsy from spinning on the shaft corroded quickly. Despite my having disassembled the gypsy several times to lubricate the clutch with lithium grease the keys continued to corrode away. Within two years the keys had corroded and worn away to the point that the clutch no longer worked smoothly. Replacing the cylindrical keys frequently would mitigate long term problems.

Autopilot - Drive Failure
It might be expected that cheap above decks autopilot drives would fail quickly, but the rate of failure has been alarmingly high. The tiller pilots with integrated control circuitry often failed when overloaded. The failure mode was that the plastic case would break in a way that allowed water to enter the housing. Once water entered the housing the control circuitry failed quite quickly. These broken tillerpilot drives did however still work with external control circuitry, and most of them went for many thousands of sea miles being used simply as drive units. On one of these cheap Raymarine ST2000 units the high torque reduction belt broke. A replacement belt got it going again, but the aftermarket belt began to slip after a while. Replacing the aftermarket belt with another identical one helped only for a short period of time. The belt tension is not adjustable on the ST2000, and the high torque belt always appears to be far too loose. The more robust Raymarine SPX5 units solve the problem of the above decks control circuitry failing, and are much more robust units in every way. The drive units on the SPX5 did however fail on us unbelievably frequently. A good drive unit lasts for a really long time, but most of the ball lead screws we got were defective and failed prematurely. We however never had any trouble with the motors or reduction gears failing. The SPX5 control circuitry was however sometimes very poor. Adjusting the settings on the unit required disengaging the autopilot, which made changes while underway annoyingly difficult. In addition to this difficulty with adjusting settings it often seemed that changing the settings did nothing other than make the autopilot work more poorly. We bought two complete SPX5 units. When the control circuitry on our first unit inexplicably failed at the same time that our first ball lead screw failed they were both repaired by the NZ Raymarine distributor. The ball lead screw was delivered back to us with no grease on it, and began to fail almost immediately. The repaired control circuitry also seemed nearly as bad as it had originally been in terms of the changeable settings not doing any good for getting the unit to work well. Years later though this repaired control circuitry inexplicably began to work nearly perfectly with all stock settings. When the control circuitry began to do a good job of steering the boat under sail though a new failure mode cropped up. The unit would sometimes put the helm hard over when motoring in flat calm conditions and send the boat 90 degrees off course. Sometimes it would then recover and go back on course, but sometimes it would just kick off onto standby mode. The second SPX5 unit we purchased had a horrible control unit that could not be set to work at all. Under rough sailing conditions the performance of this unit was absolutely abysmal, and no amount of trying different settings could get it to work acceptably. The drive unit on this second SPX5 also failed after only a rather short period of time. The point of this story is that the cheap autopilots certainly can work, but their failure modes tend to look a whole lot more like design sabotage than simple inadequacies.

Raw Water Pump - Poor Service
The cooling pump on the Yanmar SB8 was way too big for that small engine and also gave trouble in some infuriating ways. When we first put the boat in the water we found that the wrong size seal had been installed on the drive shaft for the pump. There are two seals on the shaft, one to keep the oil in the engine and one to keep the water in the pump. They are for the same size shaft, but they go in different size bores in the pump housing. The seal to keep the water in the pump has a smaller bore than the seal to keep the oil in the engine. Whoever had had the pump apart the last time had put the same size seals in both locations. The smaller seal did not seal against the housing, and oil leaked out if the engine was operated at higher engine speeds. This was an easy and inexpensive repair, but it made me wonder how much frustration this little error might have caused the last owner. The raw water pump also seemed to go through impellers amazingly quickly. It was not that they wore down fast, but rather that the pump did not work well unless it had a brand new impeller in it. It was not that the pump would not pump at all, but rather that unless a brand new impeller was installed it would easily loose it's prime in rough conditions. After I bought some cheap aftermarket impellers in Singapore the problem became much worse. The aftermarket impellers hardly worked at all in rough conditions. After much frustration I happened to notice that the back of the housing was slightly bent away from the impeller. This then was the problem, someone had bent the housing badly during a botched service procedure. Likely the same person who had installed the wrong oil seal.

Navigation Lights - Defective Bulb
We had quite a bit of trouble with the 30 year old wiring where it went up through the compression post. Part of the problem was with deteriorated floor boards that flexed and pushed on the wiring. The wood on these old plywood floorboards was still in fairly good condition, but as with much of the plywood on the boat the glue had broken down and the plywood came delaminated. Some of the old 30 year old wiring that I had reused when I rewired the boat also eventually gave trouble both in the bilge area and where it enters the mast above deck. The worst lighting failure though was when our fancy $100 LED replacement tri-color bulb failed. This was a particularly infuriating bulb failure as it did not fully stop working, but instead would intermittently cut out. Sometimes it would come on, and sometimes it would not. When I removed the bulb and tested it, it appeared to work. What I found though was that there was some circuitry in the bulb that would cause it to shut off if the slightest amount of instability in current supply was detected. For a while we used the old incandescent bulb for the tri-color, but a new LED replacement bulb got the current consumption back down to a reasonable level. The incandescent bi-color on the bow pulpit had never gotten much use, but we used it for a while when the tri-color was out of commission. In rough conditions though dipping the bow pulpit into the water with the bi-color lit caused the incandescent bulb to burn out. Replacing this bulb with an LED replacement unit got the bi-color to work reliably under all conditions.

Inverters and Chargers - "Kill Mode"
The first inverter we had on board was a Colman 800W unit that was very powerful and able to start multiple large loads without difficulty. The no load power consumption of this Colman inverter was rather high at 0.6A, but then after a few years the no load power consumption inexplicably went up to 0.9A. The replacement was a cheap $100 Pro X One inverter from an autoparts store in Honolulu. The 1000W Pro X One was also very powerful, and able to run all manner of large power tools. What it would not do though was start a second load when one load was running. The no load power consumption was however much lower at about 120mA. The Pro X One inverter died when it got wet on the heavy smash down to French Polynesia. The replacement was an even cheaper $80 PowerBright brand 1100W inverter that had an even lower no load power consumption of about 80mA. Again this cheap inverter would run very large loads, but would not start a second load. The inability of the inverter to start a second load was so bad that it would not even power two laptop computers at the same time unless they were plugged in in the correct order so that the one with the power supply with the higher inrush current was started first. This was very annoying, but even worse was the confusing stray current on the case. This stray current measured 50V D.C., but was somehow able to provide a small but apparently not dangerous shock if the case was touched. Normally 70V is required to be felt through dry skin, so I am not exactly sure how this signal that measured 50V D.C. could so reliably provide a noticeable shock. Ultimately though it was not the shocking case but the 1100W inverter's other "kill mode" that was the biggest problem. When the inverter kicked off on it's inability to start a second load it did not simply shut down, but rather went into "kill mode" where it pulsed a high voltage signal through the output that had the ability to kill less robust electronics that were plugged into it. The laptop power supplies were immune to this "kill mode", but cheaper battery chargers died easily. I would have been better off to have thrown the PowerBright brand inverter away and bought a new one as the cost of replacing battery chargers became quite large. Amazingly it was usually possible to buy 12V battery chargers just about anywhere in the world that would work on 115V power. Many chargers were specifically labeled as working on any voltage from 100-240V. Twice though I just guessed by looking at the type of charger that it would work on 115V even though it was labeled as working only on 240V. Once I did buy a charger that would not work on 115V, but that was a result of having gone shopping on a Caribbean Island before having ascertained what the electrical standard there was. I was somehow under the impression that Santa Lucia used 115V as their domestic standard, but I then found that the 6A 12V charger I bought would only work on 240V.

Consumer Electronics - Dry but Dead
I was able to keep my laptop computers and digital cameras dry throughout our worst passages by placing them all in the only totally dry location on the whole boat. That one dry location was, quite ironically, the wet locker. The small hanging locker forward of the enclosed head always stayed perfectly dry, even in the most severe storms at sea, and this was a major advantage for keeping electronics working. My first two laptop computers that I left California with, a new Toshiba and an ancient Dell, both continued to work across the Pacific and through Southeast Asia over a period of three years. In the Arabian Sea though the Toshiba died for no reason. Sometimes it would boot up, but most of the time it would just lock on bootup. I wiped the hard drive and reinstalled Windows Vista, but the problem persisted. The old Dell that I had carried as a spare and taken out periodically to test also was non-functional when finally needed. It would not power up at all. I was later told by a computer shop that the Toshiba had suffered an irreparable hardware failure that only Toshiba could fix, for more than a new laptop cost. It was probably the Bios chip that failed. The replacement Sony and Accer laptop computers I bought in Egypt and Greece continued to work, but suffered from some software problems. The Sony came with Windows 7 pre-installed but was said to be able to create it's own re-installation disks. When Windows 7 crashed not long after I got the computer the re-installation disks did not work. I installed Windows XP, but drivers were not available for much of the hardware and functionality was very poor. Eventually I got a re-installation copy of Windows 7 but it keeps harassing me that the Product Key off the sticker on the laptop is not valid. When I contacted Microsoft they told me that my Product Key is indeed valid, but that they could not help with the software itself. The Acer worked pretty well until I left it out in the rain one day, now many of the keys on the keyboard no longer work and sometimes the computer acts like the enter key is stuck down. My Sony digital cameras also have survived onboard. The old A200 did get dropped once which broke the sensor shift image stabilization system and jammed the sensor off center. The A55 that I bought in Australia failed while shooting video with a 70mm lens. Sony fixed the camera for free, and the repaired sensor shift mechanism has stood up to a great deal of heavy use. The autofocus mechanisms in three lenses failed, all in the same way. When the plastic drive gear broke it jammed the autofocus mechanism so that the lens could not be focused manually or automatically. Repairing the lenses involved removing the broken plastic drive gear so that they could only be used as manual focus lenses. The problem was that on each of these lenses the aperture drive mechanism also failed shortly after I disasembled the lens to fix the focus mechanism. On the first lens I did this to the aperture drive failed just a few days later, and the lens was so crappy wide open that I threw it overboard. The next lens that this happened to lasted only a few weeks after I removed the autofocus gear, but it was somewhat better wide open and I continued to use it for some time. The third lens whose autofocus mechanism failed worked for several years after I removed the autofocus drive gear. When the aperture drive mechanism on this third lens finally failed I found that it was good enough wide open that I kept using it. I did of course also have other lenses that had functional autofocus and aperture drive mechanisms. Lenses with built in autofocus drive motors never suffered drive failures, but they did not always continue to work well under autofocus either.

Rig and Gear - Normal Wear
The running rigging of a sailing vessel wears and fails, but the rates of wear and modes of failure depend hugely on both the quality and design of the materials as well as how they are used and what conditions the vessel encounters. Onboard Eva we had relatively little trouble with the running rigging, as it was made of high quality materials, well oversized and expertly used. Still though things did wear out, and sometimes things unexpectedly failed. The sails slowly deteriorated in the sunlight and the weaker they got the more easily they failed in tough conditions. The mainsail once got caught on the standing rigging and was torn by an overzealous winch operator. I stitched the sail back together and it was good as new. Many times the stitching chafed through and a sail came apart. The older the stitching the more easily it chafed through and failed. None of my hand stitching ever failed, but it did not have a chance to get nearly as old as the original stitching on the 20 year old sails we were using. The polyester double braid running rigging that came with the boat was already quite old, and we continued to use much of it. Right from the beginning we began to replace the running rigging, and I always made sure to have a substantial length of new half inch and seven sixteenths inch polyester double braid on board for upgrades and replacements. The worst running rigging failure that can occur typically is when the halyards wear through at the mast head, so I installed two spare halyards and made sure that they stayed in decent condition. I tried a few pieces of parallel core polyester line also, and the lower stretch and slightly higher strength seems to be a significant advantage for some applications. The feel of the parallel core line is however not quite as nice as regular polyester double braid. The aluminum teeth on line stoppers wore out, and I sharpened some sets two times on the course of the circumnavigation. Steel teeth did not wear much, and never required sharpening. I did install one line stopper with plastic teeth. Likely owing to the radical oversize of the teeth on this line stopper it never gave any trouble over a period of quite a few years. The 1/4" G40 anchor chain we bought in California seemed to wear out very quickly, and we threw half of it away in Hawaii and replaced it with some strange long link 1/4" high test chain that we had delivered to the boat sight unseen. This long link chain was cheap and very robust, but it did not work with the windlass even though the Gypsy on the windlass was for 5/16" chain. The G40 1/4" high test chain had worked fine on the 5/16" Gypsy after I cut the corners off of each tooth to prevent jamming. When we bought the new Lewmar electric windlass it came with a Gypsy for 1/4" G40 chain, but we also special ordered a gypsy for 6mm chain which was available in Australia. The Australian 6mm galvanized anchor chain was built to an Australian standard that was dimensionally equivalent to the European standard for 6mm chain. The 6mm chain was inexpensive as well as much lighter than 1/4" chain. Despite the significantly smaller size compared to 1/4" chain the Australian 6mm chain proved to be plenty strong in a number of severe anchoring situations as well as very robust as it held up well to many years of heavy use. Still though the galvanizing does come off chain, and the small heavily loaded 6mm chain has worn a bit over the years. Swapping the chain end for end once effectively doubles the life of the chain since it is the first 50 feet or so that wears most severely. A key piece of equipment for using lightweight anchor chain is the nylon anchor snubber. At first I had considerable difficulty with these nylon snubbers chaffing through, but the advantage for a smooth and quiet ride as well as lighter load on the chain was worth the effort. When I switched to the small 6mm chain I decided that snubber failure could be catastrophic in a storm, so I devised a better snubber system. This improved snubber system involves a piece of low stretch line spliced onto a longer piece of stretchy three strand nylon line. This piece of slippery and low stretch polypropylene line going over the bow roller hardly chaffs at all, and I never had a failure in three years of severe use of the same piece of line.

Engine and Electrical - Long Lived but Cantankerous
The Yanmar SB8 held up well to 5300 hours of varied use. Most of those hours were either cruising along at 2000 to 2400RPM or charging batteries out of gear at 70A at 2000RPM. We did also run the engine at 3000RPM quite a bit, and sometimes all the way up to the maximum rated speed of 3200RPM. We often ran the engine under a heavier load at 2000RPM motoring and charging batteries at the same time. This was normally done with the charging output reduced to 45A, but the engine was well capable of motoring at 2000RPM while the alternator charged at full 70A current and we did run it this way quite a bit as well. At 3200RPM the engine burned 0.45GPH under a bollard thrust test. At 3200RPM and 5.5 knots the engine burned 0.38GPH. The engine was able to pull the propeller and a full 80A charging load all the way up to about 2800RPM with no smoke, and up to 2900RPM in a cloud of black smoke. I had quite a bit of frustration with the adjustment of the regulator valve on the Yanmar SB8 until I realized that the adjustment procedure in the Factory Service Manual was in error. Adjustment of the regulator linkage is actually quite simple and easy, it is just a matter of assuring that the regulator valve fully closes when the regulator linkage is moved away from the stop position when the engine is not running. The regulator valve and regulator linkage do wear, and adjustment becomes necessary after a thousand hours or so of operation. As far as actual failures the Yanmar SB8 was mostly pretty kind to us. The mixing elbow on the wet exhaust system developed a crack and had to be replaced while we were in Hawaii, but this may have been a direct result of the problems with the cooling water pump that we always had. Getting the old mixing elbow off was a herculean task that required several lengths of pipe to use as cheater bars on our large pipe wrenches. Replacing the forward clutch disk in French Polynesia was the biggest "heavy line" job we had to do on the engine, but it was actually quite an easy job once we got the propeller shaft out of the way. While we had the transmission out we saw that one of the motor mounts was broken as well. We replaced it with the only motor mount of a similar size that was available in Tahiti. The replacement had a much larger threaded shaft on it, and required filing out the flange on the transmission to get it to fit. Almost immediately this new larger motor mount also broke. We decided that the SB8 really only wanted three engine mounts, and we had no further trouble with the remaining three engine mounts all the way around the world. I had a bit of trouble with the custom mount I built for the Bosch 90A alternator. After 4000 hours of operation the mount loosened up and began to rattle around. The problem was that I had never torqued down the pivot bolt, and this oversight eventually caught up with me. The repair was as simple as adding a few more washers and torquing the pivot bolt down tightly. The big 90A alternator on a single belt was also hard on belts, and I had some difficulty at first getting the correct belt installed. The problem was that the pulley on the engine does not take precisely the same type of belt that the Bosch alternator takes. Since the alternator pulley is smaller than the engine pulley it is the fit of the belt to the alternator pulley that is most critical, and with the correct belt installed I was able to get over a thousand hours on one belt. Heavy charging at 70A was however hard on the belt, and just a short period of operation with a poorly adjusted belt easily prematurely toasted it. Fuel contamination was only very rarely a problem, but we had a large Racor brand pre-filter installed that was able to deal with substantial contamination. We got water in the tank once back in 2006 because the vent line was not routed high enough, and it took months for the last bit of water on the bottom of the tank to work it's way through. In Indonesia we once got some contaminated fuel that clogged the fuel filter, but luckily the quantity of contamination was low enough that a few filter changes was all that was required to clean it up. In the Mediterranean we once again had a bit of trouble with water in the fuel, but this time it was the result of repeatedly getting fuel that was saturated with water. Diesel fuel has the ability to absorb a small amount of water, and once it is saturated with water it will not absorb any more. If fuel is supplied already saturated with water then any small amount of water that finds it's way into the fuel system builds up on the bottom of the tank and will not come out. Adding two liters of extremely expensive drying solvent imported from Germany got the small amount of water off the bottom of the tank, and occasional use of Startron brand enzyme fuel treatment kept the problem from coming back. The fuel gauge sending units failed three times. They were of the standard automotive type, even though we did buy them all at chandleries. The first one failed in just one month, the second lasted for five years and the last one failed after about eight months. The original gauge in the cockpit which I installed in 2005 and was often submerged in water lasted the entire time. After the third sending unit failed we switched to the more expensive type of sending unit which has a slide instead of a pivot. I believe that it was mostly the motion of the boat that killed the pivot type sending units, as the one that lasted for five years was a much shorter model for a 12" deep tank. Also exposed in the cockpit were the "ignition" switch and the current selector switch, both of which I had to replace after four years due to corrosion. The headlight switch I used to select lower charging amperages on the alternator was actually repairable, and I have since installed it in a Jeep as a headlight switch.

Propeller and Shaft - Plating versus Corrosion
The original Michigan Propeller 12x8 prop was already pretty badly worn when we got the boat, but it continued to work quite well. We had installed a new cutlass bearing in 2005, but after traversing the muddy waters of South East Asia the cutlass bearing was once again worn quite a bit. We ran with the worn out cutlass bearing all the way to Europe even though sometimes there was trouble with vibration at high engine speed. For quite a while in the Gulf of Aden and the Red Sea we had a huge amount of trouble with vibration at higher engine speeds, but eventually I found that a small piece of plastic fish net was jammed around the shaft behind the prop. Removing this piece of plastic got the prop spinning fairly smoothly again, although the worn cutlass bearing really did cause some high speed vibration problems all by itself. In Greece we hauled out to tackle the cutlass bearing as well as paint the bottom for the first time since we hauled out in New Zealand nearly two years earlier. Getting the prop shaft out turned out to be a huge challenge. It just would not come out. We purchased a hydraulic jack at a local tool store and jacked the shaft out through the cutlass bearing. What the problem was, was that a thick layer of copper material had been plated onto the shaft by the action of the zinc on the propeller shaft. When I put the boat back together with the new cutlass bearing I did not put any zinc on the propeller shaft. Leaving the zinc off stopped the plating action of copper from the stern tube onto the propeller shaft, but it also seemed to contribute to the final demise of the 35 year old propeller. In the blustery conditions of the Mediterranean in the winter I often ran the engine at 3200RPM to smash a short distance up into a protected cove or harbor, and it seemed that this was too fast for the propeller to spin through the 2.1:1 reduction box. The after sides of the tips of the blades developed severe cavitation damage, and eventually the propeller began to perform poorly. We had carried a spare 13x6 propeller all the way around the world with us, and on Barbados in the Caribbean we installed it on Eva. The spare propeller turned out to have a slightly different diameter shaft tapper on it, although the angle of the tapper was an exact match. To install the spare propeller I had to build a spacer to go between the propeller and the propeller nut. The installation went well, and the new propeller ran smoothly at all speeds. The new propeller was however considerably different than the old one in that it used less power and seemed to provide somewhat more driving force. At 2000RPM the smaller pitch meant that the amount of power put to the water was significantly less, but fuel consumption also dropped off. At higher engine speeds the larger diameter meant that driving force went up, and the boat was considerably faster at 3000RPM than it had been before. In severe conditions though the new propeller turned out not to be quite as good at punching into strong wind, and this was a major dissapointment. It is however a bit hard to sort out just what was going on with the new propeller, as the engine had begun to smoke more under a heavy load even before we installed the new propeller. It was some very heavy fuel that we got in the Cape Verde Islands that first made the heavy load smoking problem crop up. Even on the old propeller the engine would not rev all the way up to 3200RPM. With the new propeller this heavy Cape Verde fuel caused smoking already at about 2700RPM motoring into a stiff head wind. The next time we refueled the heavy load problem mostly went away, and the new prop would spin all the way up to 3200RPM. What I eventually determined was that although the new propeller was a more desirable diameter and pitch for the SB8, both using less fuel and allowing higher maximum speeds, it was not as well designed a propeller. The old Michigan Prop had more cup to the blades, and was better at developing high bollard thrust. The new propeller had much flatter blades that worked well as long as the boat speed remained high. The bollard thrust of the new propeller was however not nearly as high, and this caused disappointing performance in the most challenging of conditions.

Marine Electronics - More Robust
When we bought the boat the VHF radio was a good receiver, but would not transmit at all. I hooked a test antenna up to the unit, but it still would not transmit at all. I replaced the VHF radio with an ICOM M302 that we already had, and I gingerly tried to transmit on 1W. The antenna would not transmit, even though it worked well for receiving. I replaced the 3 foot mast top whip antenna and the VHF radio system worked perfectly. The 7/16" coaxial cable and the 35 foot mast made for an amazingly competent 25W VHF system, and communications with other sailboats was often possible out to 30 or even 40 miles. Years later though it seemed like the wiring or the antenna was failing again, as both receiving and transmitting mostly stopped working. It was the transmission capability that failed first, and then later I noticed that the unit was not receiving as well either. The other ICOM M302 that I installed in the aft cabin with a short piece of 3/16" coaxial cable to a 3 foot whip antenna on the stern rail always worked perfectly. Long rang communications with the lower antenna height of course did not work as well, and five to ten miles was the usual maximum. The five watt hand held VHF unit seemed to work quite well when we tested it, but then for many years we had little or no use for it. Much later when we tried to use it the handheld unit only just barely worked, and appeared to be transmitting at significantly less than one watt. What did work quite well were the GMRS 3.5 watt handheld units. The NiMH batteries did not however hold up well as it appeared that the charger was radically overcharging them. I was also disappointed to find that my FCC GMRS license was not recognized by many other countries, and use of the GMRS radios was not strictly permitted in most places. The Iridium 9555 satellite phone itself appeared to always work well, but the system that it connected to was disapointingly non-functional. We bought the phone in NZ, and voice communications were possible whenever we tested it in NZ. Data communications did not however work as none of the software provided with the phone or by the phone dealer did a lick of good to get the thing working as a modem. Once we left NZ the phone no longer worked well for voice communications either. To get a call through a perfectly clear day was required, and it was necessary to walk around the deck until a location was found where the rig was not between the phone and the satellite. Once these conditions were met a call could often be placed, but the call was nearly always dropped within 30 seconds. The combination of choppy low bit rate audio and the call frequently being dropped tended to infuriate who ever we were trying to communicate with, and the phone was of rather little use. Several times I downloaded every piece of software available on the Iridium.com website, but the situation was the same with no data communication capability what so ever. Then in 2011 a new piece of software became available on the Iridium website. The documentation provided with this new software indicated that it was not in fact new, but had in the past only been available through the Iridium phone dealers. Yes, it was the software required to use the phone as a data modem. By this time we were in the Mediterranean, but voice communications on the Iridium phone still worked extremely poorly. Data communications however were a bit easier. Even on cloudy days I could send and receive email, although the system still almost always dropped the call within 30 seconds. At first the GRIB files were formatted by saildocs.org to be several times larger than I calculated should be necessary. It was taking over a minute to download just one small forecast, and the phone usually cut out before the download was complete. Having to try two or three times to get the email meant that the cost of the weather forecast was astronomically high, but we could for the first time on our voyage reliably get a good weather forecast. In the Mediterranean I also nearly always bought access to the cell phone network, which provided a more convenient and much lower cost way to get all the weather forecasts we could possibly want. In many countries access to the cell phone network cost only fifteen or twenty dollars a month, but in some countries it was as much as 40 dollars a month. When we first arrived in Greece data access was sold at the very high price of thirty dollars every two weeks for English speakers and at lower rates for speakers of Greek. When the phone system was changed so that the lower price plans could be selected in English I was able to get limited data communications and essentially unlimited local voice communications for about ten dollars a month. When we set off across the Atlantic our ability to receive weather forecasts over the Iridium phone once again was cut off. This time it was a small change by my webhosting company that they did not bother to tell me about. This change in the way that the email servers worked meant that I could receive emails, but I could not send them. Once I got access to a more substantial internet connection in the Caribbean I was able to download an article about the new settings and it was a simple matter to get my email working again. The Iridium phone network also worked a whole lot better in the Caribbean than elsewhere in the world, and both voice and data communications were possible under essentially all conditions without dropped calls. The small $40 full band AM/FM receiver I direct ordered from China in 2008 worked perfectly all around the world, and I was able to pull in all sorts of extremely weak signals. The biggest problem with the cheap Chinese receiver was that it had essentially no ability to filter out local interference. I had to shut down everything onboard in order to pull in weak signals. The refrigerator was so loud that I could hear nothing over it, and likewise the alternator on the engine precluded any sort of reception. The inverter too made receiving difficult, and even the autopilot groaned and clicked so loudly that it interfered with most reception. The worst thing about the cheap little radio was that even the wire for a perfectly flat D.C. current supply interfered drastically with the reception of quite strong signals. It only worked on the internal batteries, which I had to charge periodically. As the radio had no means of keeping track of the state of charge of the batteries, and the batteries usually lasted for months, I often got cut off right in the middle of the most important weather forecast I had ever tried to receive. I often listened in on HAM nets and SSB cruisers nets, although this was not usually much of a way to get weather information. In Australia the met office had a fairly good HF broadcast that I was usually able to pick up even over in Indonesia. The USCG also had worldwide coverage with booming strong transmissions, but the format of the forecasts is of only very limited use for small vessels. Oh great, it is going to blow less than 20 knots like it always does! The USCG HF forecasts would be good for storm avoidance, but onboard Eva it was better to look for perfect 5 to 8 knot (windspeed not boat speed) sailing conditions even if it meant occasionally having to ride out a gale. The only GPS receiver that ever failed was the old hand held Magellan unit that we had had for ages. It had worked for many years, but one day it just went dead and would not turn on. The Garmin chartplotters never failed, but the tranceiver for the sonar unit did give quite a bit of trouble. Just months after installing the unit the tranceiver went dead. Instead of waiting around for a warranty replacement we just bought a new sonar tranceiver for $100 and the unit worked perfectly again. Many years later something went wrong with the sonar again. The 178C unit would still ping, but either the transmission or reception of the signal was not working well and the depth of water that we could find the bottom in dropped from 600 feet to about 70 feet. The 178C also developed a bad habit of sometimes turning on with such a faint screen image that it could only be seen by shining a bright flashlight into the screen. Very strangely I found that there was an optical sensor in the lower corner of the screen that would turn the screen back onto normal mode if a bright light was shined on it. An LED flashlight became an indispensible tool for turning the 178C on. The Garmin 545s that I installed in the aft cabin always worked perfectly, but it's software was harder to use in every way and the unit used more than twice as much power as the old 178C. My newer Sony cameras also had GPS receivers, but I was glad that they only ever got used for navigation while ashore. The GPS system itself worked quite well, and even without the WAAS system precision from day to day was usually well within 10 feet. The amount of time it took to acquire a position fix at power on was however usually more like 20-30 seconds without the WAAS system, where with the WAAS system a fix was normally found in less than five seconds. The fastest non-WAAS receiver I ever had was the Sony a55 camera, which could often get a fix in the amount of time required to carefully manual focus the lens. The Sony cameras have no WAAS receivers, but WAAS data for anywhere in the world can be downloaded from the Sony website if an internet connection is available.

Refrigeration - Functional but Power Hungry
The original Waeco ice box conversion system worked flawlessly for 5 years, but the coating on the aluminum refrigerant line broke down where it went through the insulated partition. In early 2012 the aluminum line corroded through and the refrigerator suddenly stopped working. After a few weeks of poking around southern Italy for a replacement Waeco unit with no luck we settled on an Italian made unit that uses the same Danfoss BD35 compressor. The new Vitrifrigo refrigeration kit has a larger "O" shaped evaporator, a larger capacity fin over tube condenser and is rated for a much larger box. The new system keeps the box much colder, can freeze more meat, and can even keep ice cream hard but it does use more electrical power as well. In cool weather the draw is 20 amp hours per day, and in sweltering tropical heat the thing seems to be able to suck down close to 50 Ahr/day. This despite the fact that we added even more insulation around the box when we installed the new unit. The electrical consumption of the refrigerator always seemed a bit on the high side, but particularly in colder climates the 15Ahr/day was easily supported. The new refrigeration system understandably used more power since it kept the refrigerated space much colder, but in hot weather the electrical consumption really seemed a lot higher than it needed to be. A three and a half cubic foot box has a surface area of 14 square feet, and the three inches of high performance polystyrene foam insulation would have an R value of 11. With a 35 degree Fahrenheit refrigerated space temperature and an outside ambient temperature of 90 degrees Fahrenheit that would be a heat loss of 1680 BTU per day. At a condensation temperature of 131 degrees Fahrenheit and an evaporation temperature of 23 degrees Fahrenheit the Danfoss BD35 compressor is rated to run at a coefficient of performance (COP) of 2.06, which means that just over twice as much energy (thermal energy) is removed from the refrigerated space as the compressor consumes (electrical energy). That works out to just 20Ahr/day of 12V electricity at an ambient temperature of 90 degrees Fahrenheit. There is of course some considerable additional heat loss around the door area, but even in the tropics night time temperatures are normally considerably less than 90 degrees. For some reason it just does not seem like the Vitrifrigo refrigeration system is operating as efficiently as the Danfoss spec sheet indicates is possible.

Television - Remote Failure
The television equipment onboard Eva proved to be robust and reliable, but changing standards caused total failure. Back in 2006 we bought a flatscreen TV to go on the bulkhead. It was a nice little Sharp brand unit that was advertised as having a resolution of 1080 by 1440 even though the built in tuner was only of the old analog type. When I opened the box I found that the instruction manual listed the screen resolution as only 833 by 1106, but intending to use it on the analog broadcast system I merely noted this as a severe advertising problem. The picture was perfect, better than had been attainable with CRT televisions from the past. The only annoying thing about the little Sharp TV was that the adjustment of the brightness was critical for both power consumption and easy viewing. At night the screen looked really good and used hardly any power with a low brightness setting. In the daytime though it was necessary to manually turn the brightness up quite a lot, and the TV used twice as much power with this higher brightness setting. We also bought an amplified omnidirectional Shakespeare brand antenna system. This antenea system has an amplifier built into the antenna assembly which is powered by a D.C. signal that travels on the same coax cable as the picture and sound signals. With a long length of 1/4" coaxial cable we were able to pull the omnidirectional antenna far up the mast on a halyard and get good long rang reception even when the boat was spinning around on a single hook. This system worked well for many years, but then the broadcasts were banned by the FCC. By this time we were on our way out far away from NTSC standard TV broadcasts anyway, and we bought a converter box in Australia to tune in the new digital signals. This worked quite well, but the weaker digital signal was not able to be picked up over as long of a distance. Over shorter distances the digital signal worked much better, and a rock solid picture was only interrupted by severe weather or large objects between the transmitter and the antenna. The European standard TV tuner did not however work in much of the world. It was not until we got to Europe that we once again could get TV, and then it was in Greek, Turkish and Italian.

LFP House Bank - Robust but Deteriorating
The Lithium Ferro Phosphate house bank certainly has been a highly successful piece of equipment onboard Eva, but almost five years after installing the battery premature deterioration has cropped it's ugly head. After three and a half years of continuous daily use the 180Ahr rated cells still worked just fine, but were showing severe signs of degradation. Capacity had dropped to 80% of the rated capacity, and longer taper charging periods were being required to attain a full state of charge. In this three and a half years the total number of cycles on the batteries was not more than 1000, although upwards of 600 of those cycles were very deep. Particularly since the refrigeration system failed and was replaced with the more power hungry unit the house bank wracked up more cycles with the engine being run every other day as opposed to every third day. In the last 6 months of use in the tropics the diminished capacity of the LFP house bank and the high 50Ahr per day consumption of the refrigerator meant that often the engine had to be run every day to keep up with other electrical loads. After a year in storage the LFP cells still work, but capacity has dropped even a bit more. The cells now hold 70% of their rated capacity, but this level of degradation is excessive. The manufacturer rates these big lithium ferro phosphate cells to do more than 2000 cycles to an 80% depth of discharge, and they also claim that it is possible to retain 60 or even 70% capacity after 8000 cycles. So the question is why did the cells fail prematurely. One thing is that these cells that were guaranteed by the distributor to be 180Ahr cells are in fact in smaller cases that are normally rated at 160Ahr. This however seems like a minor issue considering that the size and weight of the cells is large for their capacity compared to what other lithium ferro phosphate batteries are capable of doing. The energy density of even the 180Ahr cells is only about 100 Whr/kg (at 1C), which is considerably less than the 120 or even 130 Whr/kg (at 1C) that smaller lithium ferro phosphate cells usually attain. The question remains, why did the cells deteriorate prematurely. In order to determine the cause of the deterioration it is necessary to consider the possible limitations of the batteries. Lithium ferro phosphate cells normally are said to be fully resistant to damage from being deeply discharged, and this certainly seems to be true. What does however rapidly damage lithium ferro phosphate cells is reversing. If individual cell low voltage cutoffs are not used there is the possibility that deep discharges will cause one or more of the series of cells to experience a reverse voltage. An out of balance condition and cell reversing appears to have been what killed our 26Ahr dingy battery. Most of the cells still worked and held 80% capacity, but one of each of the strings of four cells was totally dead. When I recombined some of the good cells into smaller battery packs they worked just fine as long as they were not deeply discharged. Because the cells were mismatched the batteries were out of balance, and deep discharging would cause one cell to experience a reverse voltage at the end of the discharge. Reversing the cells appeared to totally kill them within just one to three cycles. The big 180Ahr cells however always stayed perfectly in balance, and even when the battery was accidentally discharged to 8.5V no cell dropped below 2.0V. About four times over the years the LFP house bank was acidentally fully discharged (to the 10.5V low voltage cut offs on the refrigerator and the inverter), but each time I checked the individual cell voltages and they were all within a few tenths of a volt of each other (under a discharge test load similar to the load that had brought them down). Now that the LFP house bank has lost considerable capacity the balance of the cells is not quite as perfect as it always was before, but the cells still remain within just a few tenths of a volt of each other at full discharge (while still discharging) or full charge (while still charging at 14.4V). Another possibility for the cause of premature failure of the cells would be excessive high voltage charging. The four cell bank really only needed 14.2V to fully charge, but I set the charge voltage up at 14.4V so that the taper charge period would be extremely short. The recommended charge voltage for lithium ferro phosphate cells is normally listed at 3.65V per cell, which would be 14.6V for a four cell battery. One manufacturer claimed that their cylindrical lithium ferro phosphate cells could withstand a year (9000 hours) of continuous charging at 3.65V and still deliver 80% capacity at the end of this abuse. Certainly high voltage charging does degrade lithium ferro phosphate batteries, but the approximately 200 or 300 hours that we held our 180Ahr LFP house bank at 14.4V seems like a lot less than 9000 hours at 14.6V. The LFP house bank appeared to get sluggish if it was not fully charged for a week or two, and this raises the possibility that time spent in a partial state of charge may in fact somehow degrade lithium ferro phosphate cells. Lithium ferro phosphate cells are however said to be totally resistant to damage from being stored in a partial state of charge. Ultimately the only thing that can be concluded is that at least one of the claims about the performance of lithium ferro phosphate batteries appears to be false. Is it full discharges that lithium ferro phosphate batteries are not able to withstand? Is it hundreds of hours of high voltage charging that lithium ferro phosphate batteries are not able to withstand? Is it time spent in a partial state of charge that lithium ferro phosphate batteries are not able to withstand? Then there is also the possibility that our 180Ahr cells failed for some other reason having to do with defective materials or design. All four of the cells deteriorated in exactly the same way, which is considerably different than one cell failing simply because it was different than all the other cells. The most likely cause of this excessive loss of capacity and increased taper charging is a depletion of one of the reagents in the cells. Perhaps they were built with an insufficient reserve of electrolyte. What is the electrolyte? The manufacturer very vaguely mentions only that the electrolyte is a mixture of organic solvents. Little seems to be known about the chemistry of lithium ferro phosphate batteries, but the manufacturer of our cells did publish a document that says that lithium hexafloro phosphate is present in the electrolyte. What they did not bother to mention is whether the lithium hexafloro phosphate is present in the electrolyte when the cell is fully charged or only when the cell is in a partial state of charge. For more on the lithium chemistry see  Lithium Ion Batteries.

Truck and Trailer - Standard and Easy
The 1995 Dodge pickup truck we bought specifically to haul Eva has been an extremely useful vehicle although the automatic transmission and two wheel drive have often been slightly problematic. We were aiming for a manual transmission four wheel drive truck preferably with some sort of a distributor type injection pump, but the used four wheel drive trucks were selling for quite a premium at the time. The inline injected '95 we found was undesirable to most people because the transmission was kaput and it had a whole lot of miles on it. Since the engine still ran and we knew we could easily rebuild the automatic transmission the only big down side was that it was two wheel drive. The four speed automatic is very similar to the old Chrysler 727 TorqueFlite transmissions, but it is smaller and lighter and has both an overdrive unit on the back and a lock up torque converter clutch. After rebuilding the transmission in 2005 it has given no further trouble. After a total of 30,000 miles, nearly 10,000 of which towing Eva, the transmission pan was very clean with only a small amount of clutch material around the magnet and a slightly darkened but mostly clear filter element. Even with the automatic transmission continuing to work perfectly there are two rather severe problems. First gear is not low enough, and we have several times found ourselves stuck with insufficient power to pull Eva up a hill. We once had to back down a steep section on the highway and give it another run being careful not to let the speed drop off too much in first gear. The main problem with the automatic transmission though is that it uses a whole lot of fuel when the torque converter lockup clutch is unlocked. As soon as the hill is too steep to pull locked up in third gear the power at the rear wheels drops off dramatically and the coolant temperature climbs rapidly. Running empty the high gearing allows 22mpg at a sustained 70mph cruise even with a rather big camper cap sticking up above the cab. At 55mph the mileage improves very slightly to 24mpg. At lower speeds the big engine and transmission don't do so well, and fuel mileage often drops off to 20mpg when the torque converter clutch is unlocked or even all the way down to 18mpg if city traffic is encountered. (For more on fuel mileage see Fuel Hogs). The governor in the P7100 series Bosch inline pump on our 1995 Dodge Cummins is inconsistent, and does not allow the full rated power output of the engine. The original Chrysler specification was for 400 or 420 foot pounds of torque at 1500RPM and 160 or 175hp at 2500RPM for the 1995 Dodge Cummins engine. The reality is that the governor on the pump we have is quite different than this. In high gear with the torque converter clutch locked up the engine will just barely pull Eva at 55 to 60mph (1500-1600RPM) with only the slightest amount of reserve power to accelerate very slowly on flat ground and even the smallest grade requires a downshift. Once up to 62mph (1700RPM) though the engine makes more torque and Eva can be pulled up slight grades. The maximum speed with the peddle to the metal was usually arround 67 to 71mph, and even driving all the time with the peddle to the metal the fuel mileage did not drop off to less than 13mpg. At 67mph the fuel mileage of 13mpg would be a fuel consumption of 5.1GPH. Even assuming the maximum normally attainable efficiency for a diesel engine of 145 grams per horsepower hour that works out to only 112hp on standard seven pound per gallon number two diesel fuel oil, much less than the engine is rated to produce. The 400 foot pounds of torque at 1500RPM would be 114hp, and a normaly aspirated 5.9 liter Cummins 6B is rated to do 100hp continiously at 1800RPM. Down at 55mph the 13.5mpg would be a fuel consumption of 4.1GPH, which at 145g/hphr would be 89hp. Again that is hardly more than the 85hp continious rating of the normally aspirated Cummins 6B at 1500RPM. The reality is that 1500RPM is a bit too low of a piston speed for a four and three quarter inch stroke engine, and even a Cummins 6B setup to run as a constant speed 1500RPM generator set engine attains a maximum efficiency of only 155 g/hphr. The 4.1GPH fuel consumption at 55mph works out to just 83hp assuming 155 g/hphr, which is actually less than the 1500RPM continious rating of the normally aspirated engine. The Cummins 6BT engines have been available with a variety of injection pumps and power output ratings anywhere from the 160hp of the 1995 Dodge to the 360hp at 2800RPM one hour rating of the high output marine engine using the same Bosch P7100 series injection pump. Peak operational efficiencies of all of the various Cummins 6B engines are in the 155 to 170 g/hphr range, with substantial ranges of speeds and loads falling bellow 180g/hphr. Assuming 155 g/hphr for the 5.1GPH fuel consumption towing Eva at 67mph the power output of the engine is only 105hp. That is really hardly more than than the normally aspirated 1800RPM continious rating of 100hp, which is itself a rather low power output rating considering that a 5.9 liter normally aspirated engine running at 1800RPM might make as much as 150hp maximum output. The inconsistency in the governor mechansim has occured only at higher engine speeds. Right from the first time we towed Eva with the 95 Dodge it would only just barely pull the load locked up in high gear at 55mph, and this is a big part of the reason we initially chose to always tow in third gear. For many years though the P7100 pump would deliver a very long injection duration up at higher engine speeds, and it seemed like the engine made it's full rated power up at 2000 to 2500RPM. At a stop in first gear the engine would spin the torque converter up to 2500RPM, and while driving along 2700RPM was where the engine put maximum power to the torque converter. Up at 2700RPM though it was obvious that the injection end timing was quite a bit too late as the engine would sputter and run rough at maximum power output, although it never blew visible black smoke. Even down at 2500RPM under a full load it seemed like the injection end timing was too late, and we always tried to avoid full power operation at those elevated engine speeds. The inconsistency crops up in the fact that sometimes the engine will not rev above 2300RPM at a stop and the power output at high engine speed is obviously considerably less than at other times. We normally don't need to rev the engine above 2300RPM to tow Eva, but not having maximum pulling force at low speed in first gear has several times caused problems. Since a torque converter is a hydraulic drive the power curve is roughly cubic, if 160hp were put to the torque converter at 2500RPM then only about 125hp would be put to the torque converter at 2300RPM. The main problems we have had with the trailer have to do with the surge brake system. When we bought the used trailer that had been sitting for almost two decades all we had to do was add brake fluid to get the brake system to sort of work. As it turned out though only two of the wheels had functional brakes, and they overheated and caught fire coming down the north side of the Grapevine out of L.A. We towed the rest of the way home with no trailer brakes, but I then rebuilt the entire braking system with new wheel cylinders, new lines, new shoes and new springs. As soon as we set out with the boat the next time we were amazed at the stopping power of the brake system, but we did not yet know how to use surge brakes. Eventually I overheated the brakes again, and this time so severely that a tire blew out. Fixing the brake system by the side of the road required new shoes and wheel cylinders for two wheels. The only cylinders available were expensive aluminum castings, but we installed them and they worked just fine. I also devised a procedure for using the trailer brakes to prevent them from overheating. The main problem was that once the surge brakes engaged they often would not fully disengage. It was running along with the surge brakes partially engaged that caused the overheating problems. The only way to prevent the brakes from overheating was to assure that they fully disengaged after they were used. What this meant was usually trying to break gently with the truck so that the surge brakes did not engage, but if the surge brakes were used it was necessary to pull over and then accelerate hard in first gear to fully disengage the surge brakes. With this procedure we had no further trouble with the brakes until after we came back from the circumnavigation. After sitting for five years the brakes on the trailer appeared to work, but on the empty haul across to Florida one of the wheel cylinders began to leak. It was this time very difficult to get a replacement cylinder. Then when we headed off with Eva on the trailer another wheel cylinder failed. Again it was very difficult to find a replacement cylinder. Both the cylinders that failed after sitting for five years were the aluminum castings that had cost twice as much as the traditional steel castings. Other than the one that melted from the flaming overheated brakes we only had one tire failure. It was the one bias ply tire that we were running because one of our set of radials had succumbed to the flaming trailer brakes. The bias ply tire failed when I hit a big bump coming into Bakersfield, but the real cause of the failure was undoubtably low air pressure. The radials worked reliably all the way down to as little as 35psi, but I usually ran them up at more like 50 to 65psi. The bias ply tire had just 60psi in it when it failed, and I am fairly certain that it would not have failed if it was run with the 80psi maximum. I had often noticed that the one bias ply tire was hotter to the touch after prolonged high speed cruising than the other tires. We also had some trouble with the truck bouncing. At first this was a rhythmic bucking that only occurred at certain speeds on concrete roads. Replacing all four shock absorbers on the truck totally solved this problem, but the new front shocks were harsher and gave a worse ride when empty. The cheap Chinese shocks I put on the back actually failed. When I took them off the rebound circuits were totally blown out. The replacement set of shocks were extremely harsh giving an uncomfortable ride over small bumps at anything less than 70mph. They also had insufficient rebound damping, and even when empty the truck bounced out of control over uneven highways. We towed the empty trailer all the way out to florida and towed Eva back from Florida with these defective shocks. In both directions the truck often would bounce so much that it felt like the front wheels were coming off of the road. We never had the original bucking problem come back, but I did also modify the hitch so that the ball was about six inches farther forward. Eventually I replaced the defective shocks, and the new ones provide both a smooth ride over small bumps while also keeping the truck in control when hitting the big rollers at speed. What is going on here is that there are actually two different types of damping that can be used either on compression or rebound for a total of four different damping parameters for suspension systems. These would be called always on compression damping, volume dependant compression damping, always on rebound damping and volume dependant rebound damping. The most important type of damping is the volume dependant rebound damping, which allows the wheel to bounce back from a small bump but firmly resists bouncing back from large bumps. The reason that this is important is that too much always on rebound damping causes the suspension to "pack-up" over repeated small bumps and is in a large part responsible for the washboard pattern that shows up on heavily used dirt and gravel roads. Excessive always on rebound damping also severely interferes with drive traction in off-road vehicles. A fairly large amount of volume dependant compression damping is important for off-road racing type suspension as it allows the big hits coming off of a jump to be absorbed without excessively stiff spring rates. Normally any suspension system will benefit from some volume dependant compression damping, but it certainly can be done without on road vehicles that are normally not driven aggressively over bumps. Some amount of always on rebound damping is always used, and it causes no problems as long as it is not excessive to the point of causing unwanted packing up of the suspension or other traction problems. Particularly two wheel drive pickup trucks suffer greatly when too much always on rebound damping is used. It is the always on compression damping that causes a harsh ride, and the best ridding suspension has essentially zero always on compression damping. For absolute highest performance though many applications benefit from some small amount of always on compression damping. It is the always on type of damping that is required to stop the original bucking problem we had when towing Eva. As it turns out though it takes an amazingly small amount of always on damping to prevent this type of problem. Both always on rebound damping and always on compression damping act to prevent bucking, but too much of either of them cause radically different problems. Most people would much rather have more always on rebound damping because the always on compression damping causes such a harsh ride. Even for off-road racing there is only a certain amount of always on compression damping that is desirable. If the always on compression damping is excessive then braking and cornering traction over very rough terrain will be compromised. For a two wheel drive truck the front shocks can tolerate somewhat more always on rebound damping and the rear shocks can tolerate considerably more always on compression damping. On most vehicles both ends tend to require a fairly similar amount of volume dependant rebound damping and volume dependant compression damping. Often talked about is "velocity sensitive valving", or "speed sensitive damping", and in some cases these are marketing terms for volume dependant damping. True speed sensitive damping (suspension travel speed not vehicle speed) is in fact often used. In general the goal of speed sensitive rebound damping would be to attain a high level of damping at low suspension travel speeds while not exceeding the desired amount of damping at higher suspension travel speeds. Said another way speed sensitive rebound damping could help to provide enough rebound damping to prevent out of control bouncing over the big rollers while not providing so much rebound damping over chatter bumps that packing-up of the suspension occurs. It is easy here to see how speed sensitive rebound damping and volume dependant rebound damping can be confused since they both act towards the same goal: Enough rebound damping to prevent out of control situations while not having so much rebound damping that packing-up and loss of traction occurs over repeated small bumps. Speed sensitive rebound damping and volume dependant rebound damping may act towards the same goals, but the valving is quite different and the results are also quite different. Volume dependant rebound damping has an enormous capability to provide control over the big rollers without causing any packing-up or traction problems. Speed sensitive rebound damping on the always-on circuit on the other hand simply helps out a bit by beginning to control the bouncing the very instant it begins. The volume dependant rebound damping cirucit generaly would benefit from speed sensitive damping as well to provide a large amount of damping at low suspension travel speeds while still allowing the suspension to extend quickly enough from big hits to be ready for another big hit immediately following. Always on compression damping might also benefit from speed sensitive valving in order to deliver enough damping at low suspension travel speeds to prevent bucking while not providing so much always on compression damping at higher suspension compression speeds to interfere with good handling and a smooth ride over rough terrain. The main problem with speed sensitive valving on the always on compression damping is that it can lead to a vehicle that feels harsh when it is driven slowly but then feels smooth and in control when it is driven fast. The volume dependant compression damping circuit is a unique situation, and would generally require a rather small amount of damping at low suspension compression speeds. The volume dependant compression damping is mostly needed for absorbing big high speed impacts that would otherwise bottom the suspension. On these big high speed impacts the speed at which the suspension compresses may remain quite high through much of the suspension stroke. Keeping the volume dependant compression damping low at low suspension compression speeds helps to give the smoothest possible ride over very big bumps at rather low vehicle speeds. The final concept in suspension systems is progressive damping, which is typically a substitute for better functioning volume dependant damping. Progressive damping can allow a smooth ride with little compression damping during the first few inches of suspension travel while providing much more compression damping towards the bottom of the suspension travel to prevent bottoming. The main problem with this progressive compression damping is that it causes a harsh ride if a heavier load is carried, and even repeated hits that cause the suspension to ride lower in it's stroke will tend to feel excessively harsh. A small amount of progressiveness in the compression damping can be beneficial for some suspension systems, but too much progressiveness causes large problems and in the end progressive damping is no substitute for volume dependant compression damping. Progressive rebound damping is sometimes used, allowing a low amount of rebound damping for the first few inches of the suspension travel so that packing-up is less of a problem while still providing the necessary stiff resistance to extension of the suspension after hitting a big bump. Progressive rebound damping tends to be much more useful than progressive compression damping, but still is not often used. Particularly if the spring rate is highly progressive (such as with leaf springs) progressive rebound damping is just the ticket.