48VDC generator head

ChrisOlson

Re: 48VDC generator head

papa wrote: »

Fact is, the larger displacement engine can actually be more efficient, using less fuel, at the lower RPM than the smaller screamer at rated 3600.

Only if you reduce the valve overlap event so you're not sending raw fuel from the intake valve right out the exhaust due to a low mean piston speed and low mass air flow to achieve proper cylinder scavenging. Once you start running four-stroke engines at low mean piston speeds and lower than optimum volumetric efficiency with valve timing events designed for a higher speed you run into the same problems that two-strokes have with a big hole open right from the intake port to the exhaust port, and a nice flow of heat leaving the exhaust port that sucks the incoming charge right out the hole.
--
Chris

inetdog

Re: 48VDC generator head

ChrisOlson wrote: »

Only if you reduce the valve overlap event so you're not sending raw fuel from the intake valve right out the exhaust due to a low mean piston speed and low mass air flow to achieve proper cylinder scavenging. Once you start running four-stroke engines at low mean piston speeds and lower than optimum volumetric efficiency with valve timing events designed for a higher speed you run into the same problems that two-strokes have with a big hole open right from the intake port to the exhaust port, and a nice flow of heat leaving the exhaust port that sucks the incoming charge right out the hole.
--
Chris

Have you looked at using either variable valve timing or a fixed offset to valve timing to produce an Atkinson-cycle engine of the type used in the Prius? You get less maximum power per cc and per pound, but you get better combustion and higher fuel efficiency as a tradeoff.

papa

Re: 48VDC generator head

ChrisOlson wrote:

Only if you reduce the valve overlap event

"Only?"... I would argue that hypothesis. At a minimum, you'd also need to include both combustion chamber shape, piston crown shape, and exhaust system ineffiencies to even approach a ballpark estimate due to cam swaps. And even if you did optimize valve overlap at 2000 RPM, that 'estimate' would likely be measured in micro-fractions of HP/TQ, since were discussing improvements with an engine producing only 2.75 HP (at 2k rpm) to begin with. Reality is, you'd gain more usable output at 2000 RPM by simply reducing the venturi area (read carburetor swap), than an expensive camshaft swap on this engine. But again, what's the point? since the Gx160 doesn't have the 'beans' to produce 3kW at 2000 RPM anyway.

You only need to examine the torque curve graph below to see at what RPM the GX160 is most efficient at, with its existing OEM valve timing. Optimum efficient usually occurs at or near peak 'torque'. On the GX160, peak torque occurs between 2300 and 2700 RPM,... not 3600. In fact, the GX160 produces more torque at 2000 RPM than it does at 3600 RPM. Also notice the gold horizontal bar at the bottom of the graph, where it says, "Recommended Operating Speed Range" ("2000-3600"), Not solely 3600 RPM.



But really... proof is in the hardware, so lets compare numbers when yours is up & running.

ChrisOlson

Re: 48VDC generator head

papa wrote: »

"Only?"... I would argue that hypothesis. At a minimum, you'd also need to include both combustion chamber shape, piston crown shape, and exhaust system ineffiencies to even approach a ballpark estimate due to cam swaps.

Valve overlap timing, and how much to use, is based on the principle that the intake charge has mass. Anything that has mass stores kinetic energy as a product of its mass and velocity. So the kinetic energy in the intake charge can be used two ways to improve volumetric efficiency:

- you can increase the intake valve duration event by closing it later after the piston makes the transition at BDC on the intake to compression strokes. Even though the piston is moving upward on the compression stroke the kinetic energy in the intake charge overcomes buildup in cylinder pressure for a short time so the cylinder continues to fill.
- you can increase the intake valve duration event by opening it earlier on the exhaust stroke. This takes maximum advantage of the kinetic energy in the intake charge mass by causing it to "pile up" in back of the valve head after the valve closes at BDC on the transition from the intake to compression strokes. So when the intake valve opens on the exhaust stroke the stored kinetic energy in the charge begins filling and scavenging the cylinder earlier.

All of the above is dependent on mean piston speed and number of cycles per unit of time, and velocity of the intake charge. The faster an engine runs the faster the velocity of the intake charge, and the more you can take advantage of increased duration on valve events to improve volumetric efficiency. Engines tuned for peak efficiency at one speed can NOT achieve that peak efficiency at any other speed without using variable valve timing. If you advance the camshaft timing by 2-4 degrees you will typically see an increase of torque numbers by ~5% and a drop in peak hp. If you retard camshaft timing by 2-4 degrees you will typically see an increase in hp numbers at rated speed by ~5% and a drop in peak torque. I won't go into all the gory details as to why, but it is all related to cylinder fill, scavenging and mean cylinder operating pressures at the different speeds.

Whether the combustion chamber is hemi-style, semi-hemi, wedge, polysphere, semi-poly, et al does not matter. Piston crown shape does matter but these small engines are using flat top pistons - I have not seen one yet that has popup pistons in it.

Hemi-style chambers will have an advantage in volumetric efficiency because the valve and port angles are more ideal so the intake charge and exhaust gases don't have to go around as many corners to get in and out of the hole - and you have room for bigger valves. The hemi also has an advantage in lighting the fire at the center of the chamber so the flame front is quenched at the outsides of the chamber, meaning you can use less advance on ignition timing and take better advantage of supercharging. There's a reason hemi's have been used in some of the most powerful piston engines on earth (like NHRA Top Fuel) since before Moby Dick was a minnow. Wedge style chambers light the fire on one side and have to quench it on the other side but they have the advantage of less complex valve operating mechanisms.

But in all cases, variations in valve timing have the same effect on all of them.

You only need to examine the torque curve graph below to see at what RPM the GX160 is most efficient at, with its existing OEM valve timing. Optimum efficient usually occurs at or near peak 'torque'. On the GX160, peak torque occurs between 2300 and 2700 RPM,... not 3600. In fact, the GX160 produces more torque at 2000 RPM than it does at 3600 RPM. Also notice the gold horizontal bar at the bottom of the graph, where it says, "Recommended Operating Speed Range" ("2000-3600"), Not solely 3600 RPM.

papa, this is a misconception. BSFC increases at peak torque. If you run a GX160 on the dyno you will see that the carb throttle plate is not fully open at rated rpm and power. As you load the engine to peak torque, hp drops and the throttle plate goes to fully open at peak torque due to governor droop. BSFC goes up as a result because the carb goes to its full rich power mixture range and with a sniffer on the exhaust you will see the HC emissions go thru the roof. I have a GX390 on a hydraulic pump that works in its peak torque range at full load on the pump and it sends so much raw fuel out the exhaust that the muffler glows red running at only 2,800 rpm with the throttle WFO.

Attachment not found.
--
Chris

papa

Re: 48VDC generator head

ChrisOlson wrote: »

Valve overlap timing, and how much to use, is based on the principle that the intake charge has mass. Anything that has mass stores kinetic energy as a product of its mass and velocity. So the kinetic energy in the intake charge can be used two ways to improve volumetric efficiency:

- you can increase the intake valve duration event by closing it later after the piston makes the transition at BDC on the intake to compression strokes. Even though the piston is moving upward on the compression stroke the kinetic energy in the intake charge overcomes buildup in cylinder pressure for a short time so the cylinder continues to fill.
- you can increase the intake valve duration event by opening it earlier on the exhaust stroke. This takes maximum advantage of the kinetic energy in the intake charge mass by causing it to "pile up" in back of the valve head after the valve closes at BDC on the transition from the intake to compression strokes. So when the intake valve opens on the exhaust stroke the stored kinetic energy in the charge begins filling and scavenging the cylinder earlier.

I agree, but you evaded one critical element. The dynamics you presented are severely inhibited and disrupted at part-throttle, so any hypothetical analysis assumed or applied, (absent of internal dimensions, a flow bench, and a dyno)... is flawed, and is nothing more than speculation. Suffice to say, it's blatantly obvious, that the induction and exhaust system alone on the GX160-200 engines, is clear evidence of intentional 'broad' range tuning.

ChrisOlson wrote: »

Engines tuned for peak efficiency at one speed can NOT achieve that peak efficiency at any other speed without using variable valve timing.

Yet again, you continue evade the fact that these small Honda GX engines (and many others like them) are specifically designed and tuned to run in an very broad, usable RPM range and loading without seriously impacting BSFC - exactly like a vehicle engine. Yes, if you look hard enough, there's likely a single most efficient RPM setting and load,.. but by no-stretch-of-the-imagination, will the assumed losses be the serious detriment you continue to promote. Go back and look at the Graph I posted above, and you'll clearly see the absent of a ANY 'sharp' peak in torque - it isn't there. And least I shed light on the fact... that absolutely nothing from Honda specifically restricts or even recommends a set RPM on these 'general use' engines. IMO, you are attempting to make a mountain out of a proverbial ant hill. And FWIW, there is numerous exterior mods that can be adopted to 'efficiently' alter the torque parameters to suit the target RPM (within limitations, of course)... without calling and annoying a Pro cam grinder. Done it time and time again.

ChrisOlson wrote: »

Piston crown shape does matter but these small engines are using flat top pistons - I have not seen one yet that has popup pistons in it.

You'd be surprised to learn then, that throughout the GX160 production, it endured no less than 4 different piston designs (of which only half were true flattops), and at least 3 different cylinder head configurations, both in shape and volume.

ChrisOlson wrote: »

papa, this is a misconception. BSFC increases at peak torque.

"Any engine will have different BSFC values at different speeds and loads. For example, a reciprocating engine achieves maximum efficiency when the intake air is unthrottled and the engine is running near its torque peak." http://en.wikipedia.org/wiki/Brake_specific_fuel_consumption

ChrisOlson wrote: »

I have a GX390 on a hydraulic pump that works in its peak torque range at full load on the pump and it sends so much raw fuel out the exhaust that the muffler glows red running at only 2,800 rpm with the throttle WFO.

Are you implying that this condition applies to all engines run at peak torque @ WOT? I can cite numerous conditions and/or malfunctions that are directly or indirectly associated with this exact symptom (examples, timing and/or weak mag and/or carburetor issues and/or low compression). So a cherry muffler doesn't mean squat sans specifics.

As I stated earlier... take your time, and get your 160 genny up'n running. Once you're satisfied, PM me and we'll compare BSFC numbers. (baring-in-mind, that my genhead is only about 68-70% efficient, which is nowheres near the 93% yours is).

ChrisOlson

Re: 48VDC generator head

papa wrote: »

Are you implying that this condition applies to all engines run at peak torque @ WOT?

No more than the myth is repeatedly propagated that running larger engines at slower speeds is more efficient than using smaller engines at higher speeds. I have spent too many hours on the dynamometer in my career to know better, and repeating the myth over and over does not make it true. Honda doesn't use a GX340 or 390 running at 2,200 rpm in the EU3000 for the same reasons that auto makers don't put small block 350's turning at 1,200 rpm in subcompacts to achieve fuel economy.

As I stated earlier... take your time, and get your 160 genny up'n running. Once you're satisfied, PM me and we'll compare BSFC numbers. (baring-in-mind, that my genhead is only about 68-70% efficient, which is nowheres near the 93% yours is).

Will do - I just got the generator pulled off the turbine yesterday so I can start building the skid frame for the genset. I've been tossing around whether I should use the belt drive mentioned earlier, or use the transmission off the wind turbine and drive it in reverse with a LoveJoy (the gear ratio of the turbine transmission is .400:1). The rotating mass in the generator is roughly 78 lbs and getting two 14" diameter flywheels turning that weigh 39 lbs each might present some problems in getting the engine started without some sort of clutch to bring that heavy generator up to speed gradually. Even with the engine started and running, my guess is that if I put a clutch on it, it will stall the engine if the clutch were suddenly engaged to start the generator turning. This is not a light duty generator - this is just one rotor off it:

Attachment not found.
--
Chris

inetdog

Re: 48VDC generator head

ChrisOlson wrote: »

No more than the myth is repeatedly propagated that running larger engines at slower speeds is more efficient than using smaller engines at higher speeds.

The larger engine will almost certainly last longer though, and maybe run quieter :-)
Diesels running long term at low loads can present a different problem.

ChrisOlson

Re: 48VDC generator head

inetdog wrote: »

The larger engine will almost certainly last longer though, and maybe run quieter :-)
Diesels running long term at low loads can present a different problem.

Again, not necessarily true. The vast, vast majority of the wear in your engine occurs during thermal cycling (startup and warmup), not when it is running at normal speeds and loads. That's why you can buy a car and drive it coast to coast its entire life and the engine will have no problems lasting 350,000 or more miles. But you drive it back and forth to work 2 miles every day and it'll be worn out inside 120,000 miles.

High speed engines have higher loads imposed on rods and pistons because of speed. But large lower speed engines have higher loads imposed on crankshaft, block and bearings - and in the case of long stroke engines with short rods, extremely high side loading of piston skirts and cylinder walls. So low speed engines have to be built massive or they won't stand up.

One of the largest engine manufacturers on earth is making a business out of replacing medium speed engines with more efficient, compact, longer lasting, and more powerful high-speed diesels:
http://cumminscommentary.com/2011/11/02/cummins-unveils-the-world%E2%80%99s-most-powerful-high-speed-diesel/

If you look at the specs on the QSK95 - 1.7 million gallons to overhaul in a 4,000 hp class engine - that is a 20% improvement over the best medium speed diesels in the same power class, available today.
--
Chris

inetdog

Re: 48VDC generator head

ChrisOlson wrote: »

Again, not necessarily true. The vast, vast majority of the wear in your engine occurs during thermal cycling (startup and warmup), not when it is running at normal speeds and loads. That's why you can buy a car and drive it coast to coast its entire life and the engine will have no problems lasting 350,000 or more miles. But you drive it back and forth to work 2 miles every day and it'll be worn out inside 120,000 miles.

I think that I was probably confused by the fact that low cost generators designed for very limited hours of usage are always 3600 RPM units while ones designed for long term usage are mostly 1800 RPM. This does not correspond to the loading on the engine though unless the AC output is coming from a "written pole" generator or from an inverter and so is independent of shaft speed.

ChrisOlson

Re: 48VDC generator head

That's because the high speed ones can be built cheaper. It costs money to build low speed engines because they have to have lots of iron in them to last. And who wants to lug around a big heavy low speed generator on a portable wheel frame?

But there are plenty of 3,600 rpm generators out there that will run the same hours as medium speed ones. Both diesel and gas. And they all break eventually, and the high speed ones are generally cheaper to fix. A lot of people drool over 1,800 rpm Kubota's or Yanmar's or Deutz, or what have you - until something goes wrong like an injection pump. And then they spend more money to fix just an injection pump than a whole generator costs. All it takes is one tablespoon of water and you ruin the hydraulic head in a distributor-style pump and spend $1,200 bucks on it.

When you factor in up-front cost, repair and maintenance costs, operating costs, and replacement or overhaul cost when it wears out (and they all do), I have never seen where small low rpm gensets pencil out. Cummins and everybody else has tried them, and they remain one of the most expensive (kWh/lifetime) units that exist in piston engine gensets. When you get into 30 kW class and above, then it's different. But the engines in the small units are so expensive to manufacture that the upfront cost alone usually outweighs any other advantage they may have.
--
Chris

papa

Re: 48VDC generator head

ChrisOlson wrote: »

No more than the myth is repeatedly propagated that running larger engines at slower speeds is more efficient than using smaller engines at higher speeds.

Perhaps my error for not being more precise. I am certainly not advocating that all larger displacement engines spun at lower RPM is 'always' the more efficient approach. My statement applies only to my combination - because it has already proven just that. Yes, perhaps there are others too, but undoubtedly NOT all.

ChrisOlson wrote: »

Honda doesn't use a GX340 or 390 running at 2,200 rpm in the EU3000 for the same reasons that auto makers don't put small block 350's turning at 1,200 rpm in subcompacts to achieve fuel economy.

Coupling a GX340 engine to a 3kW generator would add at least 15+ pounds and require (at minimum) different gen head internals if spun at 1800. The combo stated would no doubt increase manufacturing and assembly costs, and reduced portability. So it's safe to assume, that generator performance paramaters has little to do with it's unavailability.

The SB Chevy engine weighs approx. 500 pounds... and is easily 2.5 times physically larger than, say, a generic 3L cammer. Again, the Frankenstein swap suggested isn't about assumed 'efficiency' gains or losses due to increased engine displacement... it's about the added tonnage, installation and aerodynamic complexities it would surely entail. But since you mentioned the Chevy's 350 SB....

2001 Toyota Avalon - engine 3L (182ci) - wt 3417 lbs - MPG 21/29
2001 Chev Corvette - engine 5.7L (350ci) - wt 3239 lbs - MPG 26/28

ChrisOlson

Re: 48VDC generator head

I built the base frame for the genset, but didn't get as much time to work on it as I would've liked yesterday. The generator frame will be 30" long x 18" wide x about 22" high with 8" wheels on it to wheel it around like a portable generator. About the same size as a EU3000 Honda except about 5" longer and with better tires on it so it can be wheeled around easier.

I'm hoping that on poor solar days when we only get 100-150 watts from our solar array that I can plug the DC genset into the solar combiner and "fool" the controller into thinking it's a dang big solar panel. The panels normally run from 90-100 volts on overcast days and if I match the open circuit voltage of the panels with the generator I should be able to charge in parallel with the low panel output. I don't know what will happen when the controller does a performance sweep, but I think if the controller "sees" that the genset has the same open voltage as the panels that it won't know the difference.

At night when the sun goes down I can flip the breakers off in the combiner to isolate the panels so the voltage don't go too high and backfeed them, and run 140-145 volt into the controller with a wind curve loaded in it that optimizes the output of the generator.

Gas powered solar panels for when the sun don't work. I think it will save some hours on the big genset in the winter time, and maybe be a bit more efficient than using the inverter/charger.

I think I'm going to fully enclose the engine and generator on the portable frame for winter time operation so the engine can get up to normal operating temperature inside the enclosure in -20F ambient temps.
--
Chris

Cariboocoot

Re: 48VDC generator head

Chris;

Just a caution here: the DC output from a generator is not as 'clean' as PV output. The charge controller may not like it. It could function erratically, or burn up. Talk to the MidNite guys about doing this.

Not for your benefit (as you already know) but for anyone else reading this thread who might not: all generators put out AC. It is rectified to Dc either by the brushes/commutator arrangement or solid state rectifiers. It is 'cleaner' than rectified line current, but no place near as pure as PV or battery output.

We've had a brief discussion on the forum about using Classics in this manner (120 VAC, rectified and fed directly to a Classic 200) but I don't think anyone has been brave enough to try it yet.

BB.

Re: 48VDC generator head

If done correctly--It should be OK... Chris is already doing the same thing with his wind turbines (three phase, full rectification). And I believe Chris as already done this with another "generator".

The question of pulling the fuse/breaker back to the solar array sharing the Midnite MPPT controller--Just in case something goes wrong, voltage spikes do not get fed back to the solar array.

-Bill

inetdog

Re: 48VDC generator head

ChrisOlson wrote: »

I can plug the DC genset into the solar combiner and "fool" the controller into thinking it's a dang big solar panel. The panels normally run from 90-100 volts on overcast days and if I match the open circuit voltage of the panels with the generator I should be able to charge in parallel with the low panel output. I don't know what will happen when the controller does a performance sweep, but I think if the controller "sees" that the genset has the same open voltage as the panels that it won't know the difference.

I would be careful the first time I tried that combination. If the CC is in one of its Wind or Hydro modes with the right curves loaded, I am sure it would react to the Voc of the generator by trying to draw the correct current for max power. But if the CC is in solar MPPT mode it will instead just try to increase the current drawn until it either gets all the power the battery charging needs at the moment or sees the voltage drop to the point where pulling more current actually reduces the power drawn. This behavior is designed to work with a mostly current source panel array rather than the mostly voltage source behavior of the genset. If the CC tries to draw more current than the genset can produce, it will trip breakers or cause other problems.
As long as the maximum power needed by the CC is far enough below the max genset output, you may get away with it, depending on how far the MPPT controller sweeps.

ChrisOlson

Re: 48VDC generator head

I've tried my best to blow these Classic controllers up with wind turbines and so far have been unsuccessful I've had 'em over 200 volts input with burnt clippers and still have not managed to wreck one. When I was going to use a two-phase turbine with one once (which has "dirtier output than three-phase) boB even offered to let me try it for a year or so and then send the controller back in so they could see what the "dirty" two-phase output does to the input caps in the controller long term.

But anyway, based on what I have tested with these Classic 150's, and my numerous (unsuccessful) attempts to blow one up, I think it will work. I'm pretty much convinced that boB and Robin and the guys at MidNite did their homework on the Classic and it's about as tough as they come.
--
Chris

inetdog

Re: 48VDC generator head

ChrisOlson wrote: »

But anyway, based on what I have tested with these Classic 150's, and my numerous (unsuccessful) attempts to blow one up, I think it will work. I'm pretty much convinced that boB and Robin and the guys at MidNite did their homework on the Classic and it's about as tough as they come.
--
Chris

Yes, the guys at Midnight have produced a really rugged and insanely versatile product. What I am saying is that most solar MPPT CCs do not have any operating modes for turbine operation while the Midnight has several. One of the reasons is it so desirable for turbine users is that it provides good design support for that use. But that means when connecting a genset is that you had better put the Classic in one of its turbine modes rather than in its solar MPPT mode. Your reference to "if the controller "sees" that the genset has the same open voltage as the panels that it won't know the difference" is what I am objecting to. You really should change modes when changing between genset and PV.

ChrisOlson

Re: 48VDC generator head

I know I should use a wind curve for a genset. But what I should do and what I actually do is many times two different things, just to see where the most smoke comes from, and how much I get. If I get fire that's even better :cool:

Our 250 watt panels are wired three in series, with four series strings in parallel for the 3 kW I got feeding the Classic 150. I also got two "orphaned" panels right now that are wired in series and hooked to the bank direct until I can get around to putting up a third panel in the spring to parallel them with the other ones into the Classic. The 3 kW I got on there now don't even come close to maxing out the Classic for amps. With the cold temps in the utility room this time of year the controller don't even hardly warm up because it's only putting out like 54 amps.

The panels are ~36 Voc (I'd have to look for sure to get the exact number). So the open voltage at standard temp is around 108. I think today (we got bright sunshine for a change!) the Voc is 116.3. If I fired the little DC genset up and set the Voc to 116, then throw the breaker to connect it in parallel with the panels and input to the controller, it might throw the Vmp off a little bit for the panels. But the controller max amp setting is set to like 84 and I'm thinking the controller will adjust the Vmp accordingly when it loads the genset. It will just see the genset as another solar panel, with even the Voc to match when it does its performance sweep. Using both the solar and genset output I can get the controller up to the max amps setting, which when reached, will cause the controller to unload both the panels and genset anyway to not go over that max amps setting. But with a couple quarts of gas thru the little generator I can get the bank to absorb quicker so my water heating system can start heating water by noon.

That's my theory. And if anybody is going to try it, it'll be me. Ryan has actually blown more stuff up than me. And after they fixed his blowed up stuff it made it harder for me to re-blow it up. So this is going to be my latest experiment on it'll either "go or blow". But I'm guessing it will go.
--
Chris

BB.

Re: 48VDC generator head

My only concern is inductive kick if (for example) a breaker pops while the genset is running--Possible damaging solar panels connected in parallel with the genset. When everything is running correctly--it probably does not matter (but proper design is 20% of the work is the straight forward operational stuff--The other 80% are trying to figure out and address various error conditions in engineering/safety biz).

-Bill

inetdog

Re: 48VDC generator head

ChrisOlson wrote: »

But I'm guessing it will go.
--
Chris

As long as the panels are there to provide a base current which is pretty much voltage independent (and the CC is therefore not going to try to pull more current from the genset than it can handle,) I agree that you should be safe. I was mainly worried that somebody with less experience than you in blowing things up might try it and be disappointed with the results. :-)

I think that it will not even make a lot of difference how closely the genset output is matched to Vmp of the panels. My suspicion is that the worst case would be if the loaded genset output were above Voc of the panels. I would stand back aways while watching you test that combination. :-)

ChrisOlson

Re: 48VDC generator head

That's where I don't know enough about solar panels to say for sure. I have heard that it is possible to backfeed solar panels with higher voltage than their Voc to cause them to heat and thaw ice off in the winter. I don't know if that's true or not, or how much backfeed they can handle.

If the input breaker to the Classic would pop for some reason, the engine on the generator will momentarily surge due to governor droop and suddenly removing the load from it. That will cause a momentary over-Voc situation until the governor brings it back to the Voc throttle setting. Somebody who knows more about solar panels would have to say if that could cause any damage to them or not - or maybe put some of those big 600V diodes on the strings to prevent any backfeed at all.
--
Chris

Cariboocoot

Re: 48VDC generator head

It's true. Put enough Voltage to them and the panels will "reverse conduct" and draw off power.
This is why blocking diodes sometimes are necessary, although it's possible to back-feed through them if there's enough Voltage to overcome the resistance.

Maybe have a relay connected to the gen that simply cuts the PV's out any time the gen is running.

BB.

Re: 48VDC generator head

I would guess that "forward biased" panels is how they melt the snow--Basically controlled forward current through the panels (+ of panel to - of power source, - of panel to + of power source).

"Back Driving" the panels would put the diode junctions in an over voltage condition and cause them to break down--a permanent failure.

-Bill

ChrisOlson

Re: 48VDC generator head

What about putting those big round 600V Schottkey diodes on each string? Seems to me I saw those on NAWS website at one time but now I can't find them on there. Or maybe I saw them someplace else. Seems to me they were rated for like 50 amps @ 600V or some such thing.
--
Chris

inetdog

Re: 48VDC generator head

ChrisOlson wrote: »

What about putting those big round 600V Schottkey diodes on each string? Seems to me I saw those on NAWS website at one time but now I can't find them on there. Or maybe I saw them someplace else. Seems to me they were rated for like 50 amps @ 600V or some such thing.
--
Chris

Several comments to several opinions. (Adding mine to the list).
1. Cutting out the panels when the genset is active is probably exactly the wrong direction to go in.
2. The normal current flow direction in a PV cell is actually the forward direction in many ways. It is, however current-limited based on the amount of light present. By trying to force more current in the "forward" direction than the panels can handle you will cause resistive heating or diode breakdown or both but since this is what happens with uneven shading all the time it seems to OK at least for short term. When you apply a reverse voltage (which is how I think of a voltage greater than Voc but of the same polarity) you are actually causing diode breakdown with a greater chance of localized heating that could cause local damage.
3. A "blocking diode" in the form of a high current Schottky diode would probably protect the panel string if it needs protecting, but a zener diode in parallel (snubber diode) with a voltage higher than worst case cold Voc should perform the same protective function with less power loss. One thing that a Schottky diode would not do is protect the input of the MPPT controller if the generator surge caused a high enough voltage bounce.
4. My guess is that resistive heating of iced panels would be done with a forced forward current rather than with a reverse breakdown current, but I have never talked to anyone who has tried it. :-). There has been a lot of discussion of the heating effects of just short-circuiting the panels, and my personal opinion is that that will be no better than leaving the panels open-circuited (but better than operating them at their current MPP.)

Cariboocoot

Re: 48VDC generator head

inetdog wrote: »

Several comments to several opinions. (Adding mine to the list).
1. Cutting out the panels when the genset is active is probably exactly the wrong direction to go in.

Nope. Guarantee no harm will come to panels that aren't connected.

inetdog

Re: 48VDC generator head

Cariboocoot wrote: »

Nope. Guarantee no harm will come to panels that aren't connected.

I was thinking of the genset! (earlier issue)

Cariboocoot

Re: 48VDC generator head

inetdog wrote: »

I was thinking of the genset! (earlier issue)

I thought we were talking about protecting the panels from the gen output if the Voltage went high (shaded, non-producing panels hit with enough V from the gen, reverse current situation, that sort of thing).

I think Chris had intended and either/or input to the Classic but not both. being able to rig this up automatic (as with AC in to an inverter-charger) would be cool. 8)

ChrisOlson

Re: 48VDC generator head

Cariboocoot wrote: »

I think Chris had intended and either/or input to the Classic but not both. being able to rig this up automatic (as with AC in to an inverter-charger) would be cool. 8)

'coot - that's my eventual goal is to make it automatic shutdown and throttle control. But not necessarily automatic start. I just need to figure out a way to not wreck the panels and it would probably be best to have a relay in there that disconnects them when there's generator input.

The controller I'm not too worried about. If I blow that (which I highly doubt will happen), I'll just get ahold of Ryan and get it sent in and fixed
--
Chris

Cariboocoot

Re: 48VDC generator head

I suspect the HyperVoc function adds a little extra protection for Voltage spikes that most controllers would not have.