under-load voltage abnormally low (1.61v per cell)

[JohnnieB]

Originally Posted by Volt_Ampere

............A DVM is not very good at reading the load voltages though.

FWIW: I've got a couple DMM's with Min/Max functions and they work out fairly well for measuring stuff while on the move, or other situations where you can't keep your eyes on the meter all the time.

[sportcoupe]

I don't know how you found that info Johnnie but thanks. I just shot Delta-Q an email, Monday I will be real busy in meetings at work. Maybe I'll get an updated user guide with newer profiles through email.

When I started the individual battery load test, the pack was at 73.49V or 60% SoC because I didn't let it rest once I got to the "open area". Still it wasn't far drive and the SoC should have been much higher.

I just went for a second ride without charging from this mornings testings and sure enough, the under load volts dropped again way down to 55v. Not good. The drive was maybe 1.5 miles.

I took volt readings after 3 mins and again after waiting 15 mins.

Pack 70.9 --- 73.6
1) 8.09 ---- 8.20
2) 8.04 ---- 8.17
3) 8.05 ---- 8.17
4) 8.08 ---- 8.19
5) 8.26 ---- 8.33
6) 8.26 ---- 8.33
7) 8.07 ---- 8.17
8) 6.42 ---- 7.89
9) 8.10 ---- 8.20

I think I may change profile and charge, noting cutoff voltage if I can catch it. I don't know if #8 will recover or not, it's pretty bad.

[sportcoupe]

Changed profile to #4 (per newest user guide) and charged yesterday. Last night I saw 93.1v and it was still going. This morning it's at 90.7v and still going. I'll let it shut off and wait 12 hours for results.

[JohnnieB]

Originally Posted by sportcoupe

Changed profile to #4 (per newest user guide) and charged yesterday. Last night I saw 93.1v and it was still going. This morning it's at 90.1v and still going. I'll let it shut off and wait 12 hours for results.

Technically, a cell (or battery) isn't fully charged until the Specific Gravity of the electrolyte ceases to increase while a charging current is passing through it. Since the voltage tracks the SG, it can be inferred that the voltage also ceases to rise when the cell is fully charged.

Delta-Q uses dV/dT (Change in Voltage over Change in Time) to terminate the charge cycle and the batteries have been chronically undercharged, so it will take a while to get them fully charged, especially #8.

If you can lay your hands on an 8V charger, it would be a good idea to give #8 some individual charging.

BTW: 93.1V is well within the US Battery charging specs and since Delta-Q uses dV/dT, it might climb higher than that before the batteries get equalized and things level out. I wouldn't start getting concerned unless it went over 100.5V. More correctly, 11.2V across any one of the nine 8V batteries.

-------
If Delta-Q sends you any info, send me a copy. PM me for an e-mail address if you don't what to post it via the forum.

------
Also, when I do serious battery testing, I let them rest about half an hour after the cart has moved, before taking voltage readings.

There are a ton of variables determining how much the voltage will recover after the load is removed from the battery, but in 30 minutes they recover 92.3% of what they will recover in an hour. In 15 minutes it is 84.6%.

As long as you are aware the voltage will climb for an hour or more and you consistently wait about the same length of time, the comparisons should be reasonably valid.

[sportcoupe]

Originally Posted by JohnnieB

-------
If Delta-Q sends you any info, send me a copy. PM me for an e-mail address if you don't what to post it via the forum.

------
Also, when I do serious battery testing, I let them rest about half an hour after the cart has moved, before taking voltage readings.

I will certainly post any data Delta-Q sends.

Half hour in between each battery test would make for a long day. If I had 9 DVM's with min/max functions I'd be set.

PS, I've looked for an automatic 8v charger before but didn't find one.

http://www.buggiesgonewild.com/elect...y-charger.html


.

[JohnnieB]

Originally Posted by sportcoupe

Half hour in between each battery test would make for a long day. If I had 9 DVM's with min/max functions I'd be set.

That would make a long day, but there is a difference between At-Rest voltage for estimating SoC and Under-Load voltage testing for finding batteries that failing under load.

The Under-Load voltage is measured in Real-Time, so you just switch the DVM to the next battery and start driving again. If possible, drive over the same short route for the Under-Load testing of each battery.

When estimating SoC, or trying to identify which batteries lose more voltage, that is when you need to let the rest for while before measuring the voltage.

If it takes you 15 seconds to measure and record the voltage of each battery, the last battery will have rested 2.25 minutes longer then the first one, so they will read progressively higher. If you only wait 5 minutes before measuring the first battery, that difference will be significant, but if you want at least 15 minutes, the time induced error will be far less. The longer you wait, the less difference 2.25minutes will make in the readings.

I don't usually measure individual battery voltages when I do battery tests since my pack is fairly well matched (Typically, all seven within 0.03V before and after a 10 to 12 mile run), but when I do, I let them rest at least 1/2 hour to more or less eliminate time induced errors.

Of course, it depends on what you are looking for. A bad/weak battery ought to show up no matter how little rest time you allow.

[Cart-Buddy]

Not to be presumptuous since I'm a new guy here but after working on military battery systems for the past 15 yrs one of the things we found is that when the voltage on batteries charged in series is more that .2 different (and sometimes less than that) the lower battery will not recharge along with the rest of the pack. Internal resistance gets to be too high and they just don't accept current properly so the suggestion to recharge number 8 individually is probably the only way you have a shot at getting your pack properly balanced again and fully recharging as a pack again.

[JohnnieB]

A belated welcome to BGW!
Glad to have someone with prior battery knowledge/experience on-board.

What battery chemistry, type and voltage for the 0.2V tolerance?

The difference in the At-Rest voltage given in Post #1 is 0.21V and they are 8V, 121AH, Lead-Acid, Wet-Cell, Deep-Cycle batteries made by US Battery company, that have apparently never been fully charged, and a dV/dT type charger is being used and it is now set to the correct charging algorithm, so they might equalize. At least they are likely to get closer together in voltage unless they've been irreversibly damaged

It definitely would be better to give #8 and any of the other lower voltage ones some individual charging, but 8V chargers aren't all that common.

I had prior knowledge of Gel-Cel, MiCd, NiMh, early Li-ion and other relatively new technologies, but the old-school FLA (Flooded Lead Acid) deep-cycle technology used in golf carts was overlooked territory for me.

They are far more tolerant of typical abuse and far more forgiving than the later stuff.

My cart batteries would only charge to 85% when I got my cart, but they were up to almost 100% when I upgraded to a 42V system. All I did was charge the living daylights out of them and added distilled water as needed.

As said earlier, welcome aboard and Semper Fi. (USAF 63-75)

[sportcoupe]

Here are new individual SoC readings after 12h off charge with Delta-Q charger set to profile #4. I listed the original readings from 12h off the charger with profile #1 as reference in the first column.

As a note, it is much colder this morning then when I took the original readings. I don't know exactly how much that affects volt readings but I still have a marked improvement.

pack 75.6 --- 75.9
1) 8.38 --- 8.45
2) 8.36 --- 8.43
3) 8.35 --- 8.41
4) 8.38 --- 8.43
5) 8.52 --- 8.47
6) 8.52 --- 8.48
7) 8.37 --- 8.42
8) 8.31 --- 8.36
9) 8.41 --- 8.46

Now I plan to run back to back chargings all week and do another load check this weekend. I have ordered a new DVM with averaging feature (min/max), that will make much more accurate loaded volt checks. I suspect #8 will still fail miserably.

I have no idea why the newer batteries in #5 and #6 position went DOWN. That's odd.

[JohnnieB]

Looking good.
The pack went from 88.5% SoC to 92% SoC with just one charge.
The spread between highest voltage and lowest voltage went from 0.21V down to 0.12V.

Batteries 1, 2 & 3 had the greatest improvement, but 8 showed a significant improvement also.
Not sure why 5 & 6 decreased either, other then the improvement in the rest of the pack shut down the charger sooner, so they didn't charge as much as they had been charging, relative to the rest of the pack. The actual charge time is the same for all since they are charged in series, but how much they charge depends on their initial SoC and health in general.

As for temperature vs battery voltage, I'm pretty sure colder batteries read a different voltage than hotter batteries, but I'm not sure which reads higher. I've seen articles that says it increases as temperature decreases and other articles that says it decreases as temperature decreases and many have what appears to be valid proof of totally opposite things happening as temperature changes. So I haven't got it figured out yet.

Whatever it does, it doesn't do much and all the batteries are the same temperature, so comparisons between batteries within each set of voltage readings are valid even though comparisons between different sets of voltage reading may be a little skewed. Anyway, we are looking for gross changes and differences in voltage readings, so the temperature probably doesn't matter all that much.

Repeated back-to-back charging and charging after very short trips is how I got my original battery pack to go from charging to 88% SoC to charging to 97% SoC in four weeks.

Here is what mine did:
17-Jun-11 ----------- 20-Aug-11
1 = 6.32 ------------- 1 = 6.36
2 = 6.31 ------------- 2 = 6.34
3 = 6.32 ------------- 3 = 6.37
4 = 6.28 ------------- 4 = 6.34
5 = 6.28 ------------- 5 = 6.35
6 = 6.28 ------------- 6 = 6.33
Pack = 37.79--------- Pack = 38.09
Spread = 0.04 ------ Spread = 0.04

I was originally planning on replacing them in the spring of 2012, but after being on float charge all winter, they were charging to 98% and I was running more or less stock PDS, so I decided not to. When I finally got around to upgraded to 42V in July-13, they were charging to 100% SoC.

-------
You had mention that the batteries were of different ages in an earlier post.
Do you would happen to know what the ages of the specific batteries are?
Or at least which ones are older and which are newer.