20 of 10000mAh NiMH D cells for 500W motor

What you are proposing is very logical , your 500 watts may have momentary much higher peak power drawn , would not be suprised to see 40 to 60 amps peak surge .
Recommende taking a close look at the internal impedance of those 10A/H cells ,anything over 5-6 milliohms and you are throwing away quite like of energy . I have worled with D cells by Panasonic " HHR650D " rated at 6.5 a/h they are about 2.0 milliohms impedance .
If you expect to bring down the Cell to close 100 % DOD , would recommended that you matched the cells , although have found the HHR650D stock cells to be within 1%-2% incrediably good ( purchased from Digi key )
The one down side is their cost about $15- $18 in OEM lot.
In summary get a impedance reading on the cells that you considering , the lower the better , have seem "D" 10A/H with 15 milliohms . they are suitable for nice slow drain less 2 Amps , they would not work at all for the 20 A + required by a 500 W motor load .Have not been able to find " D " cells with lower impedance than the HHR650D

If you are considering the Chinese D cells, then that may be too much for them from what I've heard. Also, the chinese cells don't perform to their rated capacity especially at high rates.

I would just use the lead-acid. There is a lot of misinformation on the net, especially from forums like these. People will claim that you will kill lead-acid AGMs quick if you discharge them deeply, but they don't reference any manufacturer testing to back that up.

From the spec sheets from B&B batteries, discharging deeply is not a problem. As long as you don't discharge beyond about 10v you are fine. Once the batteries drop to 10v then they are almost fully discharged and will drop real quickly, so beyond that is a risk of reverse charging a cell. But, when you get to that point, you've probably used >95% of the capacity depending on the discharge rate. And the controller may have a low-voltage cutout to protect from that. If it doesn't, then just stop running the scooter when it gets noticeably slower.

A greater risk of damage is balance problems. Even a two-battery string may get out of balance. I would get two 12v chargers and charge each battery separately, or put in a double pole double throw switch to switch the batteries in parallel and charge them with one 12v charger.

That looks like it will work. The sub C cells will probably have an equal or lower impedance than the larger C chinese cells.

If I was to do this, I would go with a high quality manufacturer. SAFT makes some D size NiCad cells which are 5000 mah. It may be more difficult to fit 20 of these, but the quality, life, and performance will most likely be a far cry from the chinese cells. Check the VRE D 5500 towards the bottom here:
Batterystore SAFD D Nicad

I don't know of an easy way to monitor the cells for low voltage. If a cell gets reversed because of having less capacity, you can always test for that and replace the cell. A 20v string voltage cutoff should provide fair protection.

That is an interesting problem though. I'm sure there is an integrated circuit (IC) that could provide that functionality. Maybe even one available at Radio Shack. The way I would set this up is have it cycle through all of the cells sampling the voltage. Try and get a high enough sampling rate so that it cycles through all of the cells >5 times a second. Once one drops to <1v then have the IC trigger an automotive relay. I think some of these can handle switching 30-40 amps. The circuit to switch the relay would need to be amplified as the IC could probably not handle the power needed to control the relay.

A more advanced approach may be to have the IC send a reading of the lowest cell voltage to a digital meter that you can read with fair resolution like down to 10 mv increments. then once this drops to less then 1v, you would know to stop. Or both systems could be used having a display and an auto shutoff. This would no doubt require a fair amount of programmability from the chip.

To recharge the batteries, two strings of 12v would be a cautious approach. I would definitely recommend doing that if you wouldn't mind going through the trouble to set that up.

In summary get a impedance reading on the cells that you considering , the lower the better

How do you do this??

Mr. Mik

Hi Mik

Cell impedance (or internal cell resistance ) can be measure several ways , the one that I prefer is the standard that most battery manufacturer use .
1 Khz AC milliohmmeter such as HP4328A or HP4338 series is the instrument refer to when a impedance figure is provided .
There is a DC method that can be used to calculate( NOT MEASURE )the impedance of a cell . however have found that the charging or discharging current used in the DC method is not as quick and depend more on state of charge , temperature and other variables .Measurement in both method are fairly close (within 10-15 %), my preference is the AC method since it is very fast and repeatable .
The actual impedance number is of lesser importance than the numerical change between charge and discharge state of a cell or impedance cell variation in one production run .
AC impedance measurement is very useful when comparing similar size and amp/hour cells from different manufacturer .
Believe that impedance it is far more important than A/H figures when working in High C discharge application ,when used in low discharge .
There is direct correlation between low impedance and A/H capacity ,so in a group of identical cells the lower impedance ones will most likely the higher capacity.
The one exception that I have found , some cells have internal partial short ,they will have lower impedance , however since they are self discharging , their A/H rating will be much lower .if the A/H test is perform right after charge (within a few minutes ) , it will not be as apparent and completly missed .In my 40 years working with power generation , storage ,conversion and management have found the most important factor is impedance , since it is the one directly influencing efficiency .
Have found most batteries user either do not understand or accept the importance of impedance , have been reading comment on this and other site .Impedance is very rarely used ,like it was a dirty or curse word .
Without any question ALL BATTERIES FAILURE ARE ALWAYS IMPEDANCE RELATED ...........

Bear in mind that lead may work fine for you if the battery has the range for your needs. If the bike sits all day uncharged, that is an issue too. But while lead does have it's inconveniences, it is still a cheap way to go. If you do a deeper than 80% discharge once a weekday, you still get almost a year of riding at 200 cycles. If you have money and just want more range and convenience you would love a good sized lifepo4, like 20 ah.

From everything I have read, you'll never get more than 1C from D cells (meaning rated capacity, ie. 5000mah = 5ah no more)

Go sub-c and you can get 5-10C.

I've been looking at 10 x 7.2v RC packs (36v x 2)

When I was racing RC cars we used a simple device called a "battery bug" which was plugged on our 7.2-volt battery pack and left on until the next charge and use. We used a normal LED and a 1.2k Ohm resistor in series. When the battery bug was connected to the battery it would take several days to discharge the battery down to the point where the LED would go out. Because the LED required 1.2-volts to forward bias and come on the battery was never discharged to a point lower than 1.2-volts and the cells would NOT reverse if left on too long. This worked well for NiCad packs and if you made enough of them for your pack you could place 1 on every pair of cells and because the discharge is so slow the cells would be discharged to .6-volts without the fear of reversing the cells.

Grandpa Chas S.

hp79,

You are right we are not talking about Saft Batteries. The ones we are talking about are lower quality.

Thanks,
Grandpa Chas S.

Regarding "D " cells I have assembled a cell using a Pansonic HHR650D 6.5A/h welded 10-23 post , was able to take out over 300 AMps into a dead short for almost a full minute .Those cell have 2 milliohms impedance when fully charged The peak current that you able to take out of those cell is E/2R or 300 AMPS or almost 50C
Have to see a "C" or any other size cell of any chemistries matching this .
Love to heard otherwise .As before keep going back to impedance .................................lower the better .

Several years ago assembled a 36 volt 5000 Watt/hour pack for a electric plane.