Fischer Panda: The Final Chapter

March 2008

We have been sitting out some rainy and windy weather here in the northland New Zealand while awaiting the delivery of a new generator. I (Duncan) decided it would useful to sit and document some of my thoughts regarding our deceased Fischer Panda generator. I hope this will be a constructive log that will result in future improvements in marine genset products and processes. Prospective genset buyers may benefit from our experience.

As anyone that follows us knows, our Fischer Panda AGT-4000 12V DC 285 amp generator was by far the largest mistake we made when outfitting Whisper for cruising. We have spent a significant amount of money (>$17,000 USD so far) and time on this genset over the last six years all while enjoying a love-hate relationship with it. When it ran, we loved it. When it sat quietly with one or more unresolvable-in-the-field mechanical and/or electrical problems — which was about 15% of the time — we hated it. This log is the closure I need to put it all behind me and start anew with what we now hope will be a superior solution.

This log is aimed at two audiences:

1.) Anyone that owns a Fischer Panda AGT-4000, or is looking to purchase a marine diesel generator (of any brand) as their primary means of battery recharging and power generation aboard their boat. An additional qualifier is that you don't have deep pockets and you plan to operate your generator well away from qualified service personnel and parts departments. If you are never planning to go offshore for extended periods, will always be near Fischer Panda qualified service personnel, and spending a few thousand dollars a year on generator servicing (after an initial outlay of $15,000) doesn't phase you — you don't need to read this log. Also, if by trade, you are a qualified diesel mechanic and electrician, you won't need to read this either.

However, if you, like us, are average-Joe offshore cruisers on a budget that use 150 to 250 amp-hours per day and enjoy plentiful hot and desalinated fresh water, refrigeration, instruments, computers, a TV, and a few small appliances, you probably need something more than solar and wind and a small engine alternator. If you don't have or don't want to run your main engine every day with a large alternator, a DC genset might be a perfect solution. Then if you, like us, are/were easily persuaded by the Fischer Panda boat show demonstration of an impressively engineered, clean, tiny, quiet, genset humming along while putting out 200+ amps at 14+ volts — please read on. This log may save you thousands of dollars and many years of frustration.

2.) Manufacturers of marine generators: Any genset manufacturer looking to deliver a better customer experience and improve their products. I hope Fischer Panda will read this and take some of these suggestions to heart in the design, promotion, and support of their next generation of products.

In this log, I'll highlight what specifically was wrong with our Fischer Panda AGT-4000 generator. During that process, I'll suggest or imply areas for possible improvement in future generator designs. I'll also briefly touch upon what could be improved in the manufacturer sales and support processes and share some ideas that would make any genset easier to maintain in the field. I will also highlight what I wish the Fischer Panda genset could have done that it didn't do. Then, at the end of the log, I'll outline what we probably should have purchased (and now have purchased) based on our now clearly articulated needs.

After this log, I'll consider the Fischer Panda topic closed and I won't mention it again. Really... No really...

However, that doesn't mean you won't hear all about our new generator! Yay! Hopefully, it will be GREAT and we'll just be celebrating how smart we were the second time around! We can't wait to get our new solution delivered, installed, and running in just a couple more weeks!

So what precisely was wrong with our Fischer Panda AGT-4000 generator? Was it a lemon from the get-go, or was it just not a great solution for the way we were using it? Did we have any specific complaints with the way we were treated by Fischer Panda over the years? What compelled me to waste a few rainy days writing this log? All this and more...

Mismanaged customer expectations

One of the main complaints we had with Fisher Panda was that we don't believe our expectations were managed well during the sales cycle. There were implications that the unit delivered outputs that matched its specifications, that it was reliable, and that it was an "ideal solution" for our needs. The reality after the sale was much different. Here are the main areas where (with experience) we can now take issue...

The "true" output capability: The AGT-4000 was marketed and sold to us as 285 amp, 12V DC generator. However, after purchasing and suffering through years of support iterations with the manufacturer, we were informed the genset should only be operated for extended periods at 125 amps or less. This is less than one-half the output as specified in all the Fischer Panda marketing materials, the Genset Manual, and on their website.

In 2005, Fischer Panda informed us via email that anything near 250 amps continuous output is overloading the generator. Most of our run time during the first 1000 hours of its life was between 125 and 250 amps — or a "full" load. According to Fischer Panda field service personnel here in New Zealand, we accelerated our engine wear by running the generator at >125 amps for up to two hours per day. We were told we should not expect more than 1500 hours on the engine before replacing this engine (we were told this at 1170 hours). Rebuilding the tiny Kubota engine is not recommended. A full rebuild with labor and parts is 80% the cost of a brand new engine. Of course, a brand new engine is 80% of the cost of a brand new generator. Jeesh, talk about up-sell.

An updated operations manual provided to us in 2005 now clearly states that the unit should not be run at full output for "extended periods", but then immediately states that 2-3 hours is fine. Is that every day or over the 1270 hour life of our genset? Either way, you will see a link to an email later in this document saying anything over 125 amps is overloading. Well, that tidbit is about 1000 hours-of-use too late. There was no discussion of this 125 amp limitation prior to the sale.

According to Fischer Panda support, we were overloading our genset for the first 1000 hours. According to us, Fischer Panda misrepresented the true output capability of the genset prior to the sale. We may both be right, but hit an impasse on what to do about it.

Reliability: Fischer Panda plays the "German engineering" implied-quality card during the boat show demos — with the assertion that these are reliable pieces of sophisticated engineering and worth the upwards of $10,000 USD you'll pay. Yes, they sure do look good when they are shiny and new! Kind of like looking under the hood of a fine German automobile. However, after six years of this, my blind trust in the reliability of fancy new marine generator technology has been shattered.

We estimate that for the 700 days the Fischer Panda provided away-from-the-dock charging for us over the last six years, it was "down" at least another 70 - 100 days that we would like to have used it. We spent these "down" days either in expensive marinas and plugged in to shore power (while we worked on the generator), or we limped our way to the next marina running only our engine for charging. Once plugged-in, we usually hired qualified service people to assist us if we didn't solve the problem while out at anchor. The genset seemed anything but reliable. It seemed very expensive for parts and labor and unreliable to an almost humorous extreme. Based on our current "investment" and operational costs so far (at 1270 hours), this solution has cost us over $20 per day for battery charging.

A quick orientation: our Fischer Panda AGT-4000 12V DC genset

When the genset is down, the Volvo and Balmar can take from three to five hours of daily runtime to recharge our 735 Ah house batteries. A large 76 horsepower 4-cylinder engine (consuming about a gallon of diesel an hour) is an inefficient way to provide low amp battery charging.

We still think a DC genset is the best solution for us, and will not only acquire a new DC generator solution, but also supplement it with a couple of small solar panels. More on our new genset at the end of this document.

So, what specific problems did we have with our Fischer Panda AGT-4000 12V DC genset?

The "Sooty Burning" Problem

From leaving the dock in 2002, at any/all loads, this Fischer Panda has burned "sooty". The exhaust gas is clear coming out the stern, but the "super silent" water separator spews sooty water and bubbles into every anchorage. Over the years we have chased several wild geese looking for the root cause of this without closure.

Our current theory is that the under-powered, one-cylinder, 6 HP Kubota diesel engine can not handle our charging, watermaking, and water heating demands without carefully managing the loads to remain under 125 amps at all times. In order to do this, we would need to run the genset over three hours per day.

We were recently told by a (New Zealand) diesel mechanic that the general rule-of-thumb is 1 HP of engine for every 25 amps of alternator size. Therefore a 285 amp unit should have been powered by a minimum 10 HP engine to deliver 250 amps continuously. It was not sold to us with this "overloading" caveat, and it was only mentioned by Fischer Panda after four years of trying (without resolution) to fix our sooty burning problem.

Whatever the root cause, the exhaust soot eventually clogs the exhaust manifold and exhaust port in the head. This exacerbates the problem in a downward cycle of more soot, a clogged exhaust port, more exhaust back-pressure, and eventual depowering of the engine. Every 200 hours, the clogged exhaust port must be cleaned. This works out okay because the exhaust elbow and gasket also fail routinely and this is a good time to replace them as well.

This email thread recaps all our anticipated repairs and diagnostics when we first arrived in Auckland, New Zealand in 2005. It reveals a first admission that 125 amps is the true sustained output of the unit (mentioned above). It also documents the very generous $400 USD contribution Fischer Panda made to help us out with diagnostics of the sooting problem on a then four-year-old engine (and already three years out of warranty).

To this day, the root cause of the problem is unknown and the Fischer Panda leaves a sooty mess in the anchorage and down the side of Whisper. We have now given up on this problem realizing we cannot solve it without additional time and money expenditure. We try not to think about the environmental harm it's done, and we had bigger problems than sooty burning...

Stuck Throttle

Our second pesky problem with the Fischer Panda occurred within the first 35 hours of its life. We had an unexplained voltage climb resulting in a blown 400 amp internal fuse and complete electrical failure and shutdown. We had a long series of email exchanges with up to four Fischer Panda support and field service people looking at several red-herrings; a tiny fuse within the VCS, getting a new VCS swapped-in and sending our current one FedEx to Florida for diagnosis, replacing the blown fuses with new (twice), then with thick copper bars (bypassing any internal fusing). After four months of angst, Shea Weston emailed that he had seen our same problem on another AGT-4000 related to a stuck throttle.

On start-up, but only intermittently, the throttle motor and worm gear drives the throttle armature all the way to the end nut where it "sticks" up against the stop-nut. Once stuck, the VCS can no longer control the voltage. It climbs until the fuse blows and the genset shut down. This was very difficult to troubleshoot because it was only an intermittent problem.

Once properly diagnosed, Fischer Panda offered no solution other than keeping the gear and nut lubricated, but even lubricated, it would still occasionally stick. A simple split-plastic washer stuck over the worm gear by the end nut solved this problem for us.

During the three times this voltage-climb occurred, the wiring harness inside the unit got very hot, turned brown, and the plastic and cloth wire sheathing melted into a sticky goo. Fischer Panda insisted this was normal, not to be alarmed, and no, they would not replace the wiring harness under warranty. To this day, the brown, sticky, goo-covered wiring is still in there glopping-up anything it touches.

Exhaust Mixer (Cooler) Elbows

About annually, the extremely hard to remove exhaust cooler elbow fails. A Fischer Panda custom-built $350-$400 USD part (not including shipping), an exhaust cooler elbow (or two) is a required spare for the AGT. Clogging (from the sooty burning), exposure to the cooling salt-water, and the exhaust heat ensure pinhole failures start in the elbow welds every 100 to 200 hours.

The first couple of elbows that failed, I received free replacements under warranty, but the subsequent four were all at my expense. If the elbow failure is not caught immediately (as they are starting to fail), a real mess is made with sooty exhaust and saltwater spewing all over the inside of the sound capsule. The location of the three manifold nuts behind the elbow housing itself making it very tedious and difficult to remove and replace. A stubby custom-bent open-ended wrench is required to get the nuts on and off — one nut facet at a time — times three nuts. A few times, the metal exhaust manifold gasket rusted and failed / leaked before the elbow itself. Either way, it is a huge hassle to get the thing off and on, and it is prudent to always replace the exhaust elbow with a new one at the same time. You really only want to do this project once a year.

The exhaust elbows were the first suspicion that the Fischer Panda might need to improve the engineering of the AGT-4000. It either wasn't designed for the marine environment or it was designed for planned obsolescence to keep the parts and service revenues rolling in. But wait, there's more.

The Crankcase End-Plate

I can't really blame Fischer Panda for poorly engineering this crankcase end plate, but I can suggest they should not hang a bunch of wire brackets off of the bolts. A hairline crack in the lower corner of our cast aluminum crankcase end-plate caused an oil leak. The oil had nowhere to go but onto the foam in the bottom of the sound shield. The black-sludge absorbed into entire foam lining in base of case, then once fully engorged, the remaining oil leaked out into engine room. Nice.

The open-cell foam insulation under the engine and throughout the inside of the sound capsule is basically, a giant sponge. Although the foam under the engine is coated with a thin plastic sheet that is supposed to prevent any absorption, it looks like oil is one substance that easily disintegrates the coating. Now, once the engine is warm it smolders blue-smoke from under the engine.

Fischer Panda, having seen the catastrophic result of someone's internal engine mounts disintegrating from an oil spill, sent us four new internal engine mounts to replace ours. Lifting the engine from within the sound shield to replace engine mounts is not an easy task, so we have not yet done it. We're not sure if this was a fire hazard or not — we watched it carefully.

Raw Water Pump

When still in California in 2002, we had the LA Fischer Panda guy sell us a bunch of spares. One of the spares was a new water pump and belt. When in the south Pacific, we discovered the belt he sold us was 65mm too long and the Johnson pump he sold us had not yet had the Fischer Panda "custom groove" machined out of the foot.

The custom Panda water-pump hose-tails/hose-barbs were fabricated out of stainless and brass — two dissimilar metals screwed into a bronze pump-body and all touching saltwater. Genius. These failed in a year. When we rebuilt the pump in Mexico, we insulated the hose tails from the bronze housing, but the welds of the fitting threads to the stainless pipe still failed again in New Zealand. We corrected this by having the old ones re-manufactured out of 100% stainless. We also just purchased the latest generation hose tails from Panda that allegedly don't have this problem and are using these now with liberal amounts of thread tape.

In 2008, while in Whangarei New Zealand, we also had the local marine diesel service guys fix, clean, and repaint the corroding heat exchanger (left) and rebuild both water pumps (as well as machining out the groove on the spare pump foot to match the original pump). About $400 US and a week later, we were ready to go again, right? Wrong. We still have an oil pressure issue to find and fix.

Oil Pressure problem

With our newly rebuilt-head on the engine, serviced injector, and everything ready to go for our 2008 season of cruising (though still burning sooty), we left in mid December 2007 and off we went to the local New Zealand waters to shake down until next April. We hoped we could get another 300 hours out of the Panda (about 1500 total hours by the end of 2008) without any major or expensive genset issues.

The $17,000+ USD "investment" in this genset and the annual maintenance spend was starting to make us feel like we needed to join gamblers anonymous. "Just one more little spend on this last maintenance gamble, and everything will be OK for the next year". We hope. Maybe. Let's give it a try. Roll the dice. Lose. Okay, double-or-Nothing. Two out of three. One more $2000 spend and we'll get another 300 hours out of it...

As you know from the recent log/blog updates, our most recent $2500 USD worth of service and maintenance gambles did not pay off. 70 hours of runtime later, the genset started failing on low oil pressure and then low alternator output. It shuts down after about 10 minutes into a daily cold start-up and only delivers a maximum of 150 amps. This is our latest Fischer Panda problem and the reason the genset is has been sitting quiet for the last month.

For about the last 500 hours, after every oil change, its been progressively more difficult to re-start the genset because the oil pressure would not com up immediately. It went from five, to ten, to fifteen tries on starting before it would start. This was attributed to a likely clogged "Oil Screen" located in an inaccessible part of the engine. It is specified to be cleaned every 500 hours, but because the engine must be "lifted from the internal engine mounts and tipped in the sound capsule" in order to access the screen, this was a very low priority on my project list. There is no traditional oil filtration system on this tiny engine, and although we replace the oil every 100 hours, the oil turns black-as-tar after only an hour of use.

When the engine recently died mid-run with an oil pressure idiot light, the first thing I tried was the long-delayed oil screen cleaning. It was a big project, but no, I didn't lift the engine. I got "creative" and added a porthole to the sound capsule. I cleaned and replaced the screen. It will be much easier to clean next time. Needless to say, this didn't fix oil pressure problem.

The second thing to try was to see if the oil pressure sensor might be failing. I ordered a replacement immediately on the gamble that that's all it was and had it sent ahead to our next destination. You must remove the throttle actuator assembly before you can actually reach the oil pressure sensor, it's behind it. Again, not so convenient. Then, two of the worlds tiniest screws hold two tiny wires which must be removed and replaced all while hanging upside down in the hot engine room. Lots of swearing that day. I put the new sensor in. It worked, but only for about ten minutes. Once the engine warms up, it still shuts down on a red oil pressure idiot light.

The third thing to try was to pull, clean, inspect, and replace the oil pressure relief valve. Extra-high oil pressure can cause a relief valve to trip a "bypass" back to the sump, which could cause a severe oil pressure drop and possibly shut down. This valve is impossible to reach without removing the air intake silencer. After about a four hour project, I learned the valve (a hollow-bolt, ball bearing, and spring) was fine. It was cleaned and replaced, but still shut down after ten minutes.

The fourth was the trochoid oil pump inspection. Took it apart, sent photos to Fischer Panda. It looked okay to both of us. Put it back together. Another red-herring, but we needed to look at all the "easy" (read: inexpensive) possibilities first.

The fifth was to pay a mechanic to come measure the oil and fuel pressures with manually fitted external gauges. We did this and verified the slow drop in oil pressure over five minutes to below the 7 PSI threshold. The fuel pressure was about 1 PSI, 2-3 PSI lower than the specified pressure. We had just replaced this $350 Hardi fuel pump in Tonga before coming to New Zealand and were hoping to not have to replace the fancy expensive pump again. So we bought a $100 US fuel pump (4 PSI) from a local auto parts store just as a test to see if it was related to the low output. It ran for 5 minutes at low output and failed on oil pressure. Another red-herring.

A sixth recommendation (not from Fischer Panda) was to try a heavier viscosity oil to see if it would run longer. We tried 20W-50. Still failed after 5 minutes. Two weeks have gone by, sitting in an expensive marina, grease and oil up to my elbows, troubleshooting via email with full day turnarounds, burning money on parts and diesel mechanics, and getting increasingly frustrated.

The final recommendation from Fischer Panda was to go into the crankcase to replace a possible "spun bearing". They didn't say which one, but we are guessing the one on the connecting rod — crankshaft end. Fischer Panda support provided the caveat, that this still might not fix the problem and it might not be the only problem encountered on the disassembly.

The email thread regarding this oil pressure issue is here. This will probably be our last interaction with Fischer Panda. You may detect increasing frustration with our "Fischer Panda experience". We were coming to grips with the fact that our options are now limited and everything looks like it is going to cost significantly more money and time. In the end, after all the money is spent and the weeks are consumed — we will still have this genset. The devil we know, or should we look at something new?

Okay, so far, everything has been about what our Fischer Panda genset has done that was basically bad. What about the things we wanted it to do that it never could do (besides run reliably at full output for long periods)? What about documentation or processes that could be improved? Next...

No hot water heater thermal (coolant) take-off

Other than our 12 Ah once-a-day 120V AC coffee pot, and our TV, we have very few and very small AC loads aboard Whisper. The one exception to this is hot water heating. Our 800W AC water heater draws about 75 amps at 12V DC for an hour if we run it through an inverter. Ideally, when we run the genset, we would just use a thermal coolant take-off from the engine to provide all the hot water we need.

Fischer Panda specifically recommends against any attempt to take off hot water from their genset. So instead, we charged our batteries until the charge got to about 125 amps in, then would switch on the inverter and the hot water heater. This extra 73 amp load for another hour is what Fischer Panda has implied has caused our myriad "overloading" problems. According to Fischer Panda, the AGT-4000 can only handle a maximum 125 amps sustained. Anything more is considered "overloading". You won't find this fact anywhere in the Fischer Panda marketing materials!

No temperature compensated charging

The AGT unit does not offer any temperature compensated charging option. We added remote digital thermometers to our batteries in each of two the battery boxes and we monitor them manually. We manually adjusting the VCS as we go from warmer to hotter climates and vice versa. This should be an option on any battery charging solution.

Limited in-field troubleshooting and diagnostic tools

Fischer Panda offers no automated or computer integrated diagnostic tools or capabilities beyond the few "idiot lights" on the control panel. Even then, these idiot lights are overloaded (four different sensors may trigger one "temperature" light), leaving you to pursue several possible avenues when troubleshooting. More robust monitoring, logging, and alerting should be a part of any sophisticated battery charging solution. It should not cost a fortune, it should uses standards, and it should be simple. Automatic shut-down should still be provided, but with companion tools and diagnostics to see trends, aberrant behavior, and alert to anything outside the norm. Even attach-points for Murphy gauges would be great.

Poor Documentation Quality

Only a "beginners" troubleshooting guide is provided. An email to Fischer Panda and the accompanying turn-around times, time-zone delays, and single threaded nature of "try this next" spoon-fed troubleshooting leads to a very frustrating customer experience. I would rather have access to a robust troubleshooting guide on paper or on my computer 7X24. Unfortunately, Fischer Panda would then be revealing all their dirty laundry (problems with their gensets) in the process! This must be kept secret if they ever want to sell another generator! Caveat emptor: Let the buyer beware.

Although there has been some effort made by Fischer Panda to split the English and German versions of their genset documentation, they still need to split the installation from the operation, from the maintenance/troubleshooting/servicing/repair guides (the latter doesn't really exist for end customers, only a checklist of when certain things should be done — not how to do them).

It was very nice to see the documentation updated and significantly cleaned-up in 2005! It now outlines ways to avoid some of the problems we had encountered for our first three years. We're glad we could provide the free in-field testing and maintenance recommendations for new customers of the AGT-4000 generator.

Improve their in-the-field / Offshore User Support philosophy

A huge frustration with our unit was the inability to narrow-in quickly on the likely root cause of any problem. With no diagnostic tools, gauges, monitoring, logging, or other tools or documentation to assist, everything was down to a long list of "most likely" causes spoon-fed to you one email at a time over weeks until by dumb luck, you might finally stumble upon the real root cause.

I have included a few of our email threads that illustrate this. The first thread is in chronological order and covers email correspondence with Fischer Panda from 2002 through 2005. It touches upon all the problems we had with the Panda starting after only 30 hours of use. You may have already seen this starting with our stuck throttle / fuse blowing problem above.

This second thread was our attempt to fix the Fischer Panda once-and-for-all when we arrived in New Zealand. Fischer Panda contributed $400 USD toward the sooting problem diagnostics, but even after our $2900 USD in repairs we still had not found the problem.

The third and final thread is our most recent (and last) correspondence with Fischer Panda regarding our current oil-pressure problems. You may have already seen this one under the oil pressure issue mentioned above.

Please note, these illustrate a typical interaction and all are very friendly and cordial. Some of the folks involved early on (Todd Rickard and Shea Weston) we consider good friends and excellent businessmen and electrical/systems technicians and would work with them again today on any solution. As a matter of fact, Todd was influential in the selection of our next genset solution (whether he knows it or not) when he visited us here in Whangarei a few weeks ago!

Everyone we have dealt with has been quite helpful and the Fischer Panda support network has always been very responsive — even if it took some time to get to the actual problems. We never really had a "people" issue with Fischer Panda. It has always been a "product" issue: custom technology where off-the-shelf would have sufficed, poor or impractical design, overly complex-engineering, frustrating offshore troubleshooting/service scenarios, and light quality assurance and testing before bringing a product to market. Of course, ultimately this IS a people issue, but not with any of the people we have worked with.

I have also included a few typical repair invoices so you can see how expensive repairs can be if you don't DIY. We had two major rounds of repairs in 2005 and 2007 while we were both working in Auckland. The 2005 repairs were to try to solve all the yet-unresolved sooting issue -- and to keep the Panda running over our brief NZ holiday escapes into the Hauraki Gulf. The 2007 spend was intended to get us through the 2008 cruising season.

• Invoice for first round of 2005 repairs
• Invoice for second round of 2005 repairs
• Invoice for 2007 Repairs and Parts

Note the prices are all NZ dollars, worth about 3/4 of a US dollar. So that's about $2900 USD spent in 2005, and another $750 USD spent in 2007. And these are just samples of our major parts and service bills in those two years.

Now the Panda sits dead with it's probable "spun bearing" problem and we face a current estimate for repair somewhere between $1000 USD (do it myself, buy parts directly from Kubota) and $5000 USD (new/refurbished Kubota engine installed myself). As we have been warned, "you won't know exactly what's wrong until you get in there, it could cost more".

If I do it myself, it would take 2-3 weeks of my time. We would be sitting at a dock in a marina for $25 per day, and we would not be out cruising and enjoying the short New Zealand summer on Whisper. If I pay someone else to do it, we are on their schedule and instead we are burning our time waiting, and even more money on someone else's time. Either way, many (more!) thousands of dollars and several weeks later, we would still have the same lemon generator that has plagued us with problems for the last six years. So what to do?

Based on everything we now know and have experienced, and after a thorough analysis of this year's cruising kitty and the budget for generator repairs in 2008 (the $3000 we set aside annually), we have decided to stop "throwing good money after bad" and pull the Fischer Panda out of Whisper. This was a very big decision for us, and it was not taken lightly. It is very painful to have spent over $17,000 USD on a generator to only get 1270 hours out of it. I guess at the time we purchased the Fischer Panda, we had "more money than sense". We feel the pain for the mistake we made, but it is time to move on.

Okay, so based on all of this, what would I recommend manufacturers do differently and buyers look for?

A "better" DC Generator (for offshore use)

Simplify everything. Use simple, reliable, proven, off-the-shelf technology. Use properly sized and marinized diesel engines, un-modified water pumps, off-the-shelf heat exchangers, proven couplings, heavy duty brand alternators, and multi-stage voltage regulators with robust temperature compensated battery charging. Remove all custom bits, proprietary software, and embedded firmware that can't be in-field-user-upgraded.

Design for ease of in-the-field preventative maintenance, repair, and servicing. Ease of access to any/all parts that wear or fail. Quick, easy visual checking of all operational aspects without having to remove a 20 lb sound shield daily. Ports and hatches in sound shield, or ideally, a quiet engine requiring no sound shield.

Improve in-the-field monitoring, logging, and alerts capability. Provide gauges or digital readings to look at when idiot lights do shut things down. Discreet facts of problems and symptoms will speed troubleshooting and repair. Four temperature sensors feeding one idiot light is not useful.

Provide a hot-water heater take-off. This is obvious. If a solution is designed primarily for battery charging, and it has an engine that gets hot, why heat water with an AC element through an inverter? Recommending we buy an AC generator for our 90% 12V DC loads just to heat water and make coffee daily, watch TV for a few hours a week, and recharge our 735Ah of batteries daily through a puny 100 amp AC charger is not a good solution either.

Have a better oil filtering / screening system. Especially one that does not require lifting the engine from its internal mounts every 500 hours.

Parts: Inexpensive non-custom parts that are available anywhere in the world. Remove all proprietary parts where "you can only buy this part from the manufacturer". I know this would hurt the manufacturer's parts and service revenues, but this is just the right thing to do. For parts that fail frequently, figure out a way to provide replacements very inexpensively. If purchasing/custom manufacturing something using more expensive materials will make it far more reliable, do it. If not, don't do it.

Provide localized manuals, written by fluent local-language speaking tech writers. Split the Installation into one manual and Operation, Troubleshooting, and Maintenance procedures into another manual. Beef-up troubleshooting for advanced users and in-field diagnostics (assume we are primarily away from Internet, email and a people support network). Invest more in documentation and less in interactive support. Offer the diesel engine and alternator maker's parts manuals and shop manuals with every new generator.

Whisper's requirements and selection of a new DC Genset

Although we went through about two intensive weeks of research and decision making, we ultimately decided to purchase another DC generator. NOT another Fischer Panda of course. We won't go into the details of all the options considered and how we arrived at our selected solution, but here is a quick summary of what we learned:

• Gensets are expensive. A DC genset is marginally more expensive than an AC genset because the DC battery charging capability (the AC to DC diodes) is included with the DC genset. With the AC genset, that's another spend on top of the genset, so in our opinion, it's really a wash.
• DC Gensets that meet our criteria are few.
• Getting anything shipped into New Zealand is expensive.
• Building or buying a custom-built solution here in New Zealand from available engines and alternators would be just as expensive as purchasing a new one and having it shipped in.

First we need to recap our requirements. What were our objectives? We want a solution that will:

• Deliver a quick, "healthy", recharge of our 12V house bank — 735 Ah of Gel-AGM hybrid batteries. Battery charging is the primary need aboard Whisper because everything except the Krups coffeemaker and the TV/media center runs off 12V (disregard hot water, because we require the new solution to have a hot water take-off.
• Replace the 150 to 250 Ah used daily in about 1.5 hours of run time (supplemented by solar).
• Heat water using the generator coolant thermal heat and only optionally, the AC inverter — not just the 120V electric water heater on the AC Inverter alone (as we do today).
• Provide simple engineering and design using lots of commonly available and familiar bits.
• Be extremely reliable, tested, and proven technology under "hard service" circumstances (i.e. realistic marine use).
• Be very easy to maintain and service in-the-field with normal tools and basic mechanical and electrical skills. Be easy to find parts and service mechanics familiar with the engine (type) anywhere in the world.
• Be easy to see if anything is going wrong or starting to go wrong -- visually or via monitoring and gauges provided.
• Be as quiet as possible (given it will probably not have a sound shield to satisfy some of the above requirements). It does not have to have a sound shield. If it is quiet enough without it, we would prefer it not have one to encourage frequent and easy monitoring and checking. Ideally, it would be less than 100 decibels at full output with minimal vibration.
• It must fit in our available space and weigh less than 200 pounds (90 kg).
• Must have a larger, more reliable engine — sufficient to power a large alternator. 250 amps would require 10 HP. Find a reliable two or three cylinder marinized diesel engine in this power range. Look for an unmodified (or minimally modified) diesel engine, marinized in the simplest possible way.
• Use an off-the-shelf alternator: a reliable brand, proven technology, "hard-duty/heavy duty" high-temperature 200-250 amp alternator. It does not have to be a "marine" alternator.
• Direct drive or belt drive between the engine or the alternator, using standard available engineering components (pulleys, gears, or a spider coupling).
• Vibration insulating frame that holds the engine and alternator in correct alignment with minimal fuss or need to frequently adjust. Frame and mounts should not rust at their first introduction to salt air.
• Temperature compensated, fully adjustable, multi-step voltage/charge regulation with the ability to charge up to 14.9 volts bulk/absorption and 13.4 float (at 20C/77F).
• Simple monitoring and diagnostic tools, with the capability to add more (Murphy gauges or computer logging / monitoring / alerts) later if desired.
• Engine shop manual and parts manual. Common spares for engine.
• Alternator parts manual and in-field rebuild kit.
• Off-the-shelf pumps and pump rebuild kits available world-wide, ideally using commonly available bearings, seals, and impellers.

What solutions did we seriously consider?

Our options narrowed in on two possible solutions. The first was the Genie 12/280 built by Ample Power in Seattle, and the second option, a similar solution, the DCGen482M-12/200, a 200 amp solution from OnSite Power in Queensland, Australia (marketed by Power Stream). Here is a quick summary of both solutions:

Ample Power Genie 12/280 DC Genset

OnSite Power DCGen482M-12/200

Here is what we learned about the DCGen 482M-12/200:

• Marinized 2-cyl Kubota 482 engine direct driving (spider coupling) a 200 Amp Prestolite 4860RJ alternator (off-the-shelf) with a Sterling Pro Digital Alternator Regulator. It should put out 150-175 amps at a fixed 3000 RPMs without too much engine effort and decent fuel efficiency.

• Meets about 90% of our requirements.

• It has only a three week delivery time (can get installed and tested before we go to Fiji).

• Will be about the same price ultimately as the Ample Power solution.

• Based on a proven generator platform used by the Australian mining industry (marketing pitch says so anyway).

• Have the option to add engine-driven water maker, refrigeration compressor, dive compressor, or hydraulic pump to drive remote hydraulic motors.

• We have no idea how loud it is.

• Sales guy was very responsive in answering questions.

• But still appears to be a very small company without much in the way of a "support network".

• Would have liked a bit larger alternator to be more on par with the Ample Power solution. The 200 amp alternator at only 3000 RPM will deliver 150 to 175 amps continuously, but without strain. This engine should last much longer than our Fischer Panda's 1270 hours!

• We are sailing toward Australia and the manufacturer.

What did we purchase?

You can probably guess that we went with the DCGen 482M 12V 200 amp solution from OnSite Power (sold by Power Stream, Australia). Manufacturing is complete on 10 March 2008 and we should have it to Opua, New Zealand and installed on Whisper by the end of March 2008. The Panda will be out of Whisper and probably sold on TradeMe (New Zealand's eBay-like website) for scrap, parts, or a science project. We are VERY excited to be starting fresh!