DC TO DC efficiency
 

I would like to 'recycle' some forklift batteries in an ev conversion. Is it practical to use the dc boost stage (after voiding the warranty and taking off the cover) of a modified sine wave inverter to get 150 VDC from each of the 6 batteries?

The batteries each weigh about 150 lbs, so 6 fits in my GVRW. Then I need to series these 6 voltages to get the 900 VDC that my industrial VFD wants as input.

I searched the archives, but likely missed the answer there.

Most such inverters are not designed to operate in a series synchronized system like that ... care is needed else the magic smoke escapes.

How practical it may or may not be will depend greatly on the design of the DC-DC stage ( may not work like that at all ) , the person doing it, and how such this option compares with other alternatives for time and $.

What is a VFD?
I would guess that a DC to DC converter like you are talking about is going to be about 85% efficient, just judging from other DC to DC converters that I've seen, but finding one that will deal with 6v is going to be hard and finding one that can deal with the high amps that you are talking about (500 amps or more) is going to be the really hard part.

I would be happy with 85%. Re-using fork truck batteries is inexpensive and carrying a couple more for decreased efficiency seems reasonable. I need 35 HP out, so I'll need about 41 HP in.

A VFD, or Variable Frequency Drive, also referred to as an Adjustable Speed Drive (ASD) is an industrial motor controller. It takes three phase AC in, converts to DC - usually with 6 diodes in a bridge, has some DC filtering capacitors and inductors, then uses pulse width modulation to 'build' AC at the frequency that you want your motor to run and sends it out. This would replace the 'Curtis' controller, or the ReVolt Cougar.

The off-the-shelf DC to AC inverter I've tried takes 12V in and puts out 150- 155VDC. With 85% efficiency, that would be 12V in at 170A continuous, 340A peak. Output would be just over 11A continuous at 150VDC.

I should have said 'two fork truck batteries in series' for each inverter.

The DC output of the inverters appears to be isolated from the input terminals, and more importantly, isolated from the heat sinks. I'm testing so far - planning to put multiple inverters in series to add voltage, perhaps multiple strings of inverters to get the required current.

I need about 30A at 900VDC. That should give me 35 HP, about what I expect that 'SalvageS10' will need to make highway speed.

I've read about all of the AC conversions that I can find. No one seems to be using this method. I expect there are one or more reasons for that. I'm trying to figure out what the issues are and work around them.

I would be surprised if the original designers had envisioned this as a possible end use. I'm trying to be careful.

There are two 'boards', the boost converter that takes 12V and makes 150V, and the output inverter that takes 150VDC and makes 120V AC modified sine wave.

I'm breaking into the DC between the stages, so I'm trying to be paranoid about isolation. The DC has some ripple. The DC 'bus' capacitors take up a lot of room - I have no schematic so it's difficult to determine if it's safe to series these inverters.

The input terminals appear to be isolated from the DC bus - and +, the case/heat sink, and the output AC. Similarly, the DC bus - and + appear isolated from the case/heat sink. The DC bus is not isolated from the output AC.

The efficiency is also a function of how much you need to change the voltage iirc, like going from 12 to 15 volts isn't bad but 12 to 900 volts you can expect an efficiency hit. Not sure though

I think that is the reason people don't use this method.

If you manage to get ~85% efficiency ... getting 30A @ 900V from 12V looks like about ~2,650 Amps from the battery side :eek:

That is some serious current ... need huge connectors and cabling... be careful with those huge cables or you will also be generating some serious size magnetic fields from that electromagnet.

For continuous operation of ~30A @ 900V ... from that type of inverter you're looking at about ~18 of those inverters.... 6 in series to Voltage 3 in parralel for current.

That is a lot of places to something to go wrong ... and a good size investment in time , money , space , and weight , in inverters ... plus the question of how the battery will take ~2,650 Amp discharge rates... so before you go out and buy ~20 inverters I would look into comparing to other options ... such as more batteries to make up the voltage instead of inverters ... or a different motor controller / motor that doesn't need 900V... the other options might not work for you ... but I would still recommend looking into them before seriously investing in this path.

I see a silly math error in my example, where 6 batteries would service 6 inverters. And I have not been clear that there is a separate pair of fork truck batteries feeding each DC to AC inverter. My apologies.

I agree that 2650 amps would be far more difficult to deal with than purchasing different equipment.

Let's try this example again with better numbers. 6 pairs of fork truck batteries. Each pair gives 12V at 500A to one AC to DC converter (DC Converter) which is sized better. Each 5000W (instead of 1750W) DC Converter puts out 30A at 150VDC, assuming 85% efficiency. String the 6 DC Converter outputs together in series to get 30A at 900 VDC.

This discussion of theory ignores the fact that I'm over the GVRW for the truck, and assumes that I successfully cool the heat sinks well enough that nothing melts.

The two biggest costs for an EV, from what I've read, are the battery pack and the controller. I have access to cheap big heavy batteries that are low voltage, and a cheap controller which is very high voltage. I've come up with this goofy idea to make the two work together. And I want to do it safely.

I'm nervous because I have not seen this method used thus far. I'm not arrogant enough to think that I'm the first to think of this. It's been thought of before and rejected for some very good reasons that I'm looking for. If I can find a solution to those good reasons - I'll try it.

One reason is inefficiency. Losing 15% of your battery power is a huge penalty. Not only is it lost, but I must find a way to get rid of the heat from all of those heat sinks - fans, liquid cooling, whatever. If I can carry 15% extra (cheap) battery to balance that off, and find a reasonable way to get rid of the heat, that one seems surmountable.

That is sound advice for any project.

Perhaps the effort of using the cheap parts won't work out, but a mental exercise is always worth it!:)

6 inverters each rated 5,000 watts each , don't usually come very cheap... unless you have a very good source for them.

Also remember those peak rating numbers on inverters are only for surges ( a few seconds ) ... if you run the inverters for any significant amount of time at more than the continuous rating , you have a good chance of burning them out to an early death.

What kind of Lead Acid batteries are you considering? Do you have any model numbers or specs on them?

Lead is heavy to start with ... and Lead usually doesn't like high Amp current rates very much either ... plus the Peukert effects of the harder you push them the less you get out of them ... if you are pushing them too hard they might not have a very long service life under the stress either.

And as you already noted ... even if you get ~85% efficiency ... those inverters alone will be pumping out over ~5kw of heat ... plus the additional heat from the batteries themselves ... plus the additional heat from the 900V VFD ... plus the additional heat from motor.

What kind of motor is it? Do you have any model numbers or specs? ... If the only reason to need this step up to 900V is to use the VFD as a motor controller ... than we can look at other motor controller prices ... which will put a cap on what the VFD + Inverters would have to be under to be a cost effective alternative.

I'm not very picky. I expect that it will be a mix of whatever is traded in and has not been sent to the recycler. Some of them will have enough life left to help me. Some will not and I'll get them traded back. Sorry - no part numbers.

I agree. I don't know much about fork trucks or their batteries. I heard from the service guy I was talking to that they get traded in when they won't last a shift, or half a shift. He said that there is still some life in them when they are traded in for the core charge.

It is a GE industrial motor, 575V, three phase, medium efficiency, rated 1766 rpm at 36.8 amps. Class B insulation (affects maximum temperature), service factor 1.15. Totally enclosed, fan cooled. This thing weighs a lot - maybe 500 lbs? I'm not really worried about this motor overheating. It will take a lot more abuse than I can dish out.

The VFD is quite a bit over-sized. It is an Allen Bradley 1336 Plus II (circa 1998) rated at 302A, 150% for 30 seconds.

What I would do is find someone who installs solar electric systems and offer to trade them a few forklift batteries for a few golf cart batteries, pound for pound you are going to be better off because for the same weight you can have a big pile of golf cart batteries giving you a higher end voltage... altho 900 volts is still way up there.
Either way, I think it would be worth trading up to equipment that will better suit a car,

Brand New Lead Acid batteries are usually less than about ~40wh/kg ... these older ones they are replacing will most likely be significantly less than that ... but you can test them easily enough when you get them.

Given what you know already about the batteries ... 12V and weighs about ~150 pounds ... that means it is safe to assume when they were new they had something around ~230 Ah... until you test them to better quantify where they stand now it is reasonable to expect them to be somewhere less than ~80% of the original capacity ... which means , until you can quantify differently I would expect these batteries to offer less than ~180 Ah @ 12V ... when discharged at a ~20 hour rate.

The rate is important because of Peukert effects on the amount of usable battery capacity at a given discharge rate.... Depending on the type of Lead Acid it can range from about 1.05 to 1.6 ... The high Amp rate needed for the battery side for this DC-DC step up concept forces a large Peukert hit to the useful battery capacity ... a far larger Peukert hit than one would see from a higher voltage lower current battery system for the same power output ... if we assume a conservative Peukert k value of ~1.2 ... @ ~500 Amps we might expect to see something around ~80Ah usable ... 12*80=960*6= ~5.7kwh expected ... of course you can improve the accuracy of that estimate with some better quantified specs from testing these batteries to get on real numbers and not just estimates.

You should not be at full power 100% of the time ... so if your vehicle can manage about ~4 miles per kwh average vehicle efficiency ... you might expect to see around ~23 miles of range ... +/- YMMV.... if you do 100% DoD ... which is a bad idea for Lead Acid battery service life... and you will also loose additional usable range when the batteries get cold... Maybe somewhere around ~12 Miles per charge would be reasonable to expect in the winter time.

Extra Large Motor + Extra Large Controller + additional inverters = a lot vehicle space and weight is getting used by non-battery / non-energy storing devices.

But it should be functional ... I still have some doubts about how practical it will be.

Well that has a lot of documentation for it.
Link

It "should" work as long as each converter is working, but what happens if one should shut down(for what ever reason) and suddenly sees -750v (yes negative) inside it? (in a place it shouldn't be possible no less since you hacked it)


(edit)
After posting i thought of two things:

1. if the outputs are galvanically insulated, you can make one 2s6p pack for all converters, instead of giving each converter it's own 2s pack, this eliminates the risk of one converter running out of input-juice.

2. You can give each converter a bypass diode to protect it if it shuts down.

Adding one diode between inverter stages and one as a bypass for each inverter sounds like a good fix.

What do you put between the old and unbalanced batteries to allow them to be run in parallel? Diodes?

With one 2s pack each, the supplies are isolated from each other. The loss of one reduces the output voltage by 150V but can be survived. Hopefuly I can figure out a way to 'enable' each inverter separately and can shut them down when the battery voltage drops below 11.0 V. Schottky diodes that can deal with 100 amps or so will drop about 3V. 5 diodes, 35A continuous, another 175W of precious battery power dissipated as heat ...

Thank you for the ball park figures.

My numbers for the drive to work are 30 HP (for 40 minutes), 90 km/h average - maybe 15 kw-h at 100% efficiency. Assume some efficiencies: battery to VFD = 85%; VFD to motor = 90%; I need just under 20 kw-h. Your numbers give me about 5.7 kwh, less than 25% of what I need. Maybe with 3 times the battery, the battery draw would be lower and the kwh would rise a bit, maybe not.

I guess it's a start.

I have a 1250 lb 'budget' for batteries. At 35 lbs each (lowest I can find for info online) plus battery frame, that would get me 25 - 30 batteries. The kwh add up pretty close, though.

So if you traded your current free parts for golf cart batteries and a fork lift motor you could come in at about the same weight and have a 144v system that you know is going to work and work well while giving you full highway speeds.
I do like the idea of doing something different like this, but at the same time it's going to be complex, inefficient and heavier then using more common parts.

I don't know that I could trade the older fork truck batteries for others - I'd have to check that out.

If I can make these parts work out, I should have a good source of batteries for a few years at least.

I'd like to have something reliable (aka no experimenting) but I'd like to experiment ...

Peukert effect are exponential ... for example:
For instance the same batteries with a Peukert k value of 1.2 and 180Ah at a 20 hour discharge rate will give.

~80Ah when discharged at ~500Amps.
or
~111Ah @ ~100Amps
or
~176Ah @ ~10Amps

etc.... etc.

The thing to keep in mind about Peukert Effects is that the Wh of energy are not really all lost ... it is just that the chemical reaction of the battery can not keep pace with the faster rate of the electronics ... for example if you discharged these same batteries at ~500Amps ... you might expect to estimate running out of ~500 Amp load ability in about ~80 Ah ... but if you were to then let the batteries rest for an hour or so you will be able to pull more from them again ... as the chemical reactions inside the battery that produce the electrical potential take time to catch up with the faster electronics ... Some old time BEV people call it 'growing Amps'.

Not all batteries have the same Peukert effects ... some handle high discharge rates much better than others ... A123 is among the best ... Lead Acid is among the worst... but even in Lead acid there is a significant variation from battery to battery.

Sure it's a start ... and with that many inverters you could power a whole house from your battery pack in case of power outages... ~30kw of Inverter AC is well over 200 amps of house 120 level AC ... easily an entire houses 100% full electric load running just about everything in the house at once.

Using that weight as your battery budget we can work backwards ...

Unless you can test and confirm better than the gross estimates.
~12V ~180Ah per ~150 Pound Battery = ~8 batteries max = ~17kwh
~17kwh from Batteries into 85% efficent Inverters = ~14.4kwh output
~14.4kwh from inverter to ~90% efficient VFD = ~12.9 kwh output
~12.9 kwh from VFD to ~90% efficient Motor = ~11.6kwh to wheels....

Unfortunately the batteries are not ideal batteries ... and you will have Peukert effects to deal with making this less than ... and you will have cold temperature effects to also deal with in the winter also making it less ... and it is not a good ideal to run 100% Do, also making it less.

Sorry ... I don't see this setup of batteries and components getting you ~15kwh to the wheels in ~40 minutes or less.

Maybe look at ways to reduce that ~15kwh ...

Or get some tested numbers on batteries that have more than ~180Ah @12V @ ~150 Pounds @the Discharge Rate you are aiming for ( not at some ~20 Hour Rate )

The modified sine wave inverters don't have a way to synchronize with each other. There would be lots of runtime on a single inverter, which could run the furnace, the water pump, the fridge, the deep freeze and maybe a few lights.


I think you are right. And as has been mentioned, a DC system would work well and could have the range that I need, but I won't be setting up a DC EV system. I'll work on the AC system with a bunch of old junk that costs me nothing, or almost nothing, and see what I can learn.

I don't want to marginalize the batteries. They appear to be the center of the system, and the rest of the system revolves around them. They weigh as much as the rest of the truck, you need to build big sturdy boxes for them and they have to be connected properly, charged, and maintained.

They are just not very interesting compared to tuning the VFD, measuring the torque produced, estimating acceleration, trading off the size of the motor versus weight versus temperature rise and reduced life of the electric motor, driving the instrument cluster, etc. A couple of people are using industrial motors and controllers, but there is not a lot of real-world data out there (that I can find, anyway).

I am not a die-hard EV fan. If I have to add a 10 - 20 HP Briggs and Stratton engine to make some electricity to extend the range of whatever I come up with - that's also an option. Less gas is not my issue. Less cost would be nice, but again is not the issue. Understanding what is going on and troubleshooting myself is my goal. I *HATE* paying mechanics to connect a computer to my car and say 'I need to replace the Oxygen sensor'. But you just did that a month ago. What's causing the problem? 'Uh - the computer says I need to replace the Oxygen sensor - that'll be $600'.

If I ever get off my butt and build something, I'll start a build thread and I'm sure there will be much further discussion, about many things.

I don't think it is a DC or AC thing.
I think the issues are still the same weather it was a DC system or an AC system.

• You have a target range.
• You have a target weight limit.
• There is a net total system efficiency from battery to wheel.

The batteries store the energy for all the vehicle needs for power and range... they limit everything else... a 1,000 HP electric motor and controller only has 10HP if that is all the batteries will give... and range is a function of energy use rate and again limited by what the batteries give.

I completely understand they might not be as interesting to you ... I find them very interesting myself ... but to each their own.

DIY HEVs like that have been made by the DIY BEV crowd years before the OEMs got into HEVs.

The short version is that the ICE generator isn't free ... not in $ , not in space, not in weight ... what you spend on it ( $ , or L , or kg ) you are not spending on batteries... so there will be some break even points along that line.

• Smallest ... Regular non-Plug in HEVs like the Prius.
• Medium ... PHEVs for increased net vehicle efficiency.
• Large ... PHEVs Primarily runs as BEV , only using ICE rarely to extend range.

The size and weight of even ~10kw continuous output generator eats more than some people expect... the ICE itself is not the source of the ICE benefits ... the ICE benefits mostly come from the massive energy density of the fuel they run on, not on the ICE itself that is needed to make use of that fuel's energy density.

And a good goal it is.
And a benefit of BEVs is their comparative simplicity to ICEs.

I think you've already started ... the trick is not to get too impatient and expect it to be done fast or easy ... you've begun the work of ironing out what your finished goals are , for energy use ... power output , weight , cost , etc ... lots of progress.:thumbup: