Batteries

So I spent a lot of time playing batteries and IID while I was on a recent trip to Exmouth and back caught out in poor solar weather. The D4 is a trick beast when it comes to voltages.
The IID can show you the battery voltage (the voltage applied to the battery by the BMS), battery current (current being consumed or being drawn into it (charging)), alternator current (everything sucking its life), State of charge, and body battery voltage (voltage as shown to accessories etc).
I noted that it charges the battery as per a battery charger with high voltage being applied, lower during absorption etc. These voltages did not reflect the voltage displayed in cabin. Interestingly the in cab voltages fluctuated as has been said before, with them rising on decel presumably to take advantage of fuel saving.

What I took away was that you can’t assume the in cab voltage has any reference to the battery. When the battery was charged (0A inflow), the battery voltage was maintained at 14V yet the in cab voltage was 12.2V.

Also it showed that over 24 hours of driving my 2.5 year old AGM could not get higher than 84 percent SoC on the car. I’ve yet to try and Ctek it and check to see what it reports after a Ctek charge.
All very interesting.

Interim update for the MF88H after 1259 days (3.45 years) in service.

I got bored/interested, so I disconnected the battery and charged it at 14.8V until the current dropped below ~200mA (it's full mate). Let it sit for an hour or so and then commenced a constant current discharge test. The battery is rated at 90AH (C20) or an RC of 160 Minutes. If I conflate the 2 (90AH @ 4.5A & 66.6AH @ 25A) that gives a Peukerts exponent of about 1.17 for a new battery, which this isn't.

The problem I have is when I tested the original cast-off back in 2019, I tested at 2A to 10.8V. Thats a C45 rate, and with the Peukerts exponent means the battery should show ~102AH (which it didn't because it was ****ed). Back then it returned ~20AH. So it was down to 20% of new and still *serviceable.

So, this time I'm testing at 4.5A (C20) to a cut off of 11.5V because Supercharge says "don't discharge below 11.5V". Now I recently characterised ~20 UPS batteries between 7 & 55Ah and reviewing the logs there's maybe 5-8% available between 11.5 & 10.8V on a really generous day. Additionally it's cold outside (average temperature over the test has been 13C) so I'd expect a best case of ~81AH (5% for the high cut off and 5% for the low temp).

I've had 3 "severe discharge" incidents where it wound up below 9V due to "operator error" over that time, but nothing catastrophic, so this should be pretty "typical". I figure if I see 50% (40.5AH) after 3 and a half years I'm probably doing ok and should be good for another year or so.

*where serviceable meant it'd start the car after a few wakeups but no more.

It sustained 4.5A for 8 hours 50 minutes for a total of 39.698AH (close to my target). It got very close to starting the car, but due to the low voltage the car wouldn't engage the glow plugs and it was cold this morning. It spun it over for quite a while but wouldn't fire.

I reckon it'll do me for another year but I might start keeping my eyes on the sales.

Attachment 180910

48 and a half months (1482 days). Down to 30.7 AH. Not seeing any issues yet, but it does sound like it's turning over a bit slower on a cold morning. the discharge test behaviour is odd. The voltage bounces around +/- 200mV once it gets right into the meat of the discharge. That's why the curve is fuzzy and there is 4142 data points in the test. I've never seen that before. The test in September had 51 data points (it only logs a point if the voltage or current changes). The internal resistance must be increasing, because as soon as I put the charger on it jumped up to ~13.4V at 10A.

Attachment 185106

Haven't been watching the sales, but I suppose I should really start doing that.

Hi Brad, have you by any chance carried out a Maintenance Charge routine at all.

The reason I ask is that your battery is still showing a decent CCA behaviour but as you stated, the voltage jumping around like that is a bit strange.

Quote:

---

Originally Posted by drivesafe

Hi Brad, have you by any chance carried out a Maintenance Charge routine at all.

---

Yep, for the last week the car has been backed into the driveway and straight on charge every time I got home. That's what prompted me to re-test after I started to see no improvement in the charge behaviour.

I charge using a 10A bench supply with reasonably accurate metering. It took about 5 days for the float current to gradually drop to ~80mA, and it certainly recovered faster after each drive. I probably should have done a test before I started, but the reality is at > 4 years in a car where it's the only battery it lives a relatively hard life.

Bulk charge at ~4A, absorption at 14.8V and float at 13.8V. Temp has been cool enough to not need to apply much compensation to the voltages.

I have quickly read through this.. So I may have missed a point or two.

I had a supercharge mf88 in my d4. I used to put them in everything (still do) it was excellent. But it always had a resting voltage of 12.3v even when not in the car.. I tested it multiple times and it passed on CCA. Didn't do a AH test. I changed it out of fear it's now at dads and still going as a battery for his pump on his sprayer.

I'll be going back to one when my ss88 fails me.

Quote:

---

Originally Posted by BradC

48 and a half months (1482 days). Down to 30.7 AH. Not seeing any issues yet, but it does sound like it's turning over a bit slower on a cold morning. the discharge test behaviour is odd. The voltage bounces around +/- 200mV once it gets right into the meat of the discharge. That's why the curve is fuzzy and there is 4142 data points in the test. I've never seen that before. The test in September had 51 data points (it only logs a point if the voltage or current changes). The internal resistance must be increasing, because as soon as I put the charger on it jumped up to ~13.4V at 10A.

Attachment 185106

Haven't been watching the sales, but I suppose I should really start doing that.

---

Been racking my brain about your graph as I have never seen a battery behave like that.

I am not ruling the battery out as the cause, but and just a suggestion, even though you are probably connecting your load, for the test, directly to the battery's terminals, have you checked all your cable connections, particularly earth connects?

Or, and this might be more feasible, could you have a relay oscillating ( continually cutting in and out ) and the reason I suggest this is because your first test had no initial negative spike ( for want of a better name ) and while the second test has that initial negative spike and then as the voltage dropped, you got that oscillation with about a continuous and comparatively even 3 second cycle?

Yeah, it's certainly an odd one.

The first test I let the battery rest for an hour. The second I didn't.

The load device is a SykyRC BD200 electronic load (up to 20A or 200W whichever comes first). It connects directly to the battery terminals with decent size croc clips and has nice short fat cables. In this instance it was only imposing a ~4.5A load.

I wrote some custom software that sits on a laptop to drive it. So aside from not letting it rest for an hour between charge and discharge the test methodology was identical.

Do you test the battery while it is connected or do you disconnect one of the leads?

Always disconnected. Can’t get an accurate test when you don’t know what else is waking up.