Roadtrek E-Trek Battery Replacement

[markopolo]

Was that actually at rest (no input or output) for a week or still being charged by solar panels?

[photolimo]

Originally Posted by markopolo

Was that actually at rest (no input or output) for a week or still being charged by solar panels?

No loads, just the solar float. No other charging.

[markopolo]

The 6.8's and 6.9's are being charged. Were the measurements taken in daylight?

If so, that could indicate A & F have lost capacity and are not able to reach the voltages the other batteries get to while being charged.

[booster]

Originally Posted by photolimo

I did a simple load test by turning on a few high load appliances for 15-20 minutes. I took voltage readings before any load and during the test.

After resting for a week:
A: 6.33v
B: 6.43v < New Cell
C: 6.94v
D: 6.88v
E: 6.43v < New Cell
F: 6.34v
G: 6.94v
H: 6.88v

Under 50% load:
A: 5.31v
B: 5.39v < New Cell
C: 5.32v
D: 5.34v
E: 5.32v < New Cell
F: 5.14v
G: 5.36v
H: 5.32v

Does this tell me that cell F is also experiencing failure?

Attachment 2945

G and H and C and D pairs are the 24v only batteries and look to be in decent shape and match pretty well. A and B are in the front with long wiring so they will see less voltage when on charge and probably hold voltage better on discharge because of the wiring. It does look like F may be heading south.

[markopolo]

If you add up the voltages posted you can easily see the problem of voltage imbalance that can develop in series connected batteries.

Those posted voltages are from the solar system in float mode. Float is often set near 13.2v in 12v controllers & 26.4 in 12v controllers.

Ideally you'd see each 6 volt battery at around 6.6v float.

Bank 1
6.33 under
6.43 under
6.94 over
6.88 over
The 26.58 volts total is what you'd expect, but two batteries are clearly being overcharged and two batteries are being undercharged while in float mode.

Bank 2
6.43 under
6.34 under
6.94 over
6.88 over
26.59 volts total is what you'd expect, but again two batteries are being overcharged and two batteries are being undercharged while in float mode.

The solar setup is doing its job by maintaining approx 26.6 volts but the individual batteries in each 24v group are imbalanced.

[photolimo]

Another thing I noticed is that the new batteries are not the same brand or capacity as the old ones.

Original Batteries: 190Ah@20Hr Bulk 6.9-7.3v; Float 6.7-6.8v
--2 New Batteries: 210Ah@20Hr Bulk 6.1-6.2v; Float 6.6-6.7v

Should we be concerned about that type of mismatch?

[booster]

Originally Posted by photolimo

Another thing I noticed is that the new batteries are not the same brand or capacity as the old ones.

Original Batteries: 190Ah@20Hr Bulk 6.9-7.3v; Float 6.7-6.8v
--2 New Batteries: 210Ah@20Hr Bulk 6.1-6.2v; Float 6.6-6.7v

Should we be concerned about that type of mismatch?

Everything that I have ever seen says that they should be the same style, brand, model number, and even age and production run for best results. The more batteries in the bank, the more important it is they match well is often a common thing also.

So I would say it definitely is a potential problem.

[photolimo]

Originally Posted by markopolo

Was that actually at rest (no input or output) for a week or still being charged by solar panels?

I took some readings this week after disconnecting the solar and letting them rest for 24 hours.

After rest (no solar charger):
A 6.26v
B 6.40v < New Cell
C 6.43v
D 6.43v
E 6.40v < New Cell
F 6.26v
G 6.43v
H 6.43v
E-H 25.43v

Under 33% load at 24v inverter for 1/2 hour:
A 5.98v
B 6.13v < New Cell
C 6.06v
D 6.08v
E 5.09v < New Cell
F 6.92v
G 6.08v
E-H 24.11v

After revving above 2500rpm and at idle:
E-H 27.12

[markopolo]

A&F pairing are much weaker than the others. I see them as being 65% to 70% at best. Is the problem A or F or both?

E won't last long if being 100% discharged like that. It will drag down B.

So a load causes charging at AF?
AF pair is 12.9V under load and 12.52V at rest.

Equalizer (E in model #) should give 1/2 of 24.11V to AF.

Converter (C in model #) would give (try to give) 13.5V if 22V+ seen at H. H was 24.11V.

Photolimo - can you see a C or E in model # of the Cooper device?

[booster]

Here is another 8v battery I stumbled across. Don't know anything about the brand, though. It is 200 amp hours, in the same case size, so that's interesting. That would still give you your normal 800ah at 12v that the etrek always speced.

UBGC8-45968 8 Volt 200AH Golf Cart Battery

I don't know what quality batteries Roadtrek uses, but it looks like they have changed?

[photolimo]

Originally Posted by markopolo

A&F pairing are much weaker than the others. I see them as being 65% to 70% at best. Is the problem A or F or both?

E won't last long if being 100% discharged like that. It will drag down B.

So a load causes charging at AF?
AF pair is 12.9V under load and 12.52V at rest.

Equalizer (E in model #) should give 1/2 of 24.11V to AF.

Converter (C in model #) would give (try to give) 13.5V if 22V+ seen at H. H was 24.11V.

Photolimo - can you see a C or E in model # of the Cooper device?

F is the cell I was worried about but after this test I think A may be near dead as well.

I updated the numbers exactly as I mixed up one cell last night when I copied them by hand incorrectly. So many numbers to keep track of!

After rest (no solar charger):
A 6.26
B 6.40 < New Cell
C 6.43
D 6.43
E 6.40 < New Cell
F 6.26
G 6.43
H 6.43
E-H 25.43

Under 33% load at 24v inverter for 1/2 hour:
A 5.98
B 6.13 < New Cell
C 6.06
D 6.08
E 6.09 < New Cell
F 5.92
G 6.09
H 6.08
E-H 24.11

After revving above 2500rpm and at idle:
E-H 27.12

[booster]

A and F are the remaining two on the 12v section, so it makes sense that they would be right behind the two that were replaced in terms of degradation. At least that would make that section match batteries.

With the change of size, brand, age, charge specs, and the known problems it has had, they really should be replacing the entire bank of 8 IMO.

[gregmchugh]

Originally Posted by booster

A and F are the remaining two on the 12v section, so it makes sense that they would be right behind the two that were replaced in terms of degradation. At least that would make that section match batteries.

With the change of size, brand, age, charge specs, and the known problems it has had, they really should be replacing the entire bank of 8 IMO.

I agree, I would push Roadtrek to replace them all with 8 new matched batteries and try to get them to use one of the better rated brands if possible.

[markopolo]

It would be great if RT would immediately replace A & F to at least match the newer B & E. With new A and F we'd see what that would do to help the very poor balance of voltages under solar charging float mode.

Obviously, there's still the underlying problem of maintaining the A B E F bank that needs to be dealt with.

I don't know when the equalizer was installed. If that is fairly new then there's a chance things won't look so dire if A & F are replaced now.

Replacing A and F will show if the setup is functional to a low to moderate standard or if reconfiguration is absolutely required to avoid continual premature battery failure.

[gsm]

photolimo

I would use a clamp-on Ammeter to check currents and balance among all eight batteries. On voltage alone, a battery that does not carry its share of current, can continue to look good. In a similar way I would check the charging-currents and determine balanced charging. I would also make sure that terminals are all torqued properly. At the currents flowing in this battery system, miniscule resistance, as well as difference between parallel current-paths, play a large role in current-sharing among batteries, and batteries tend to mask the weakness of the paralleled partner. While the large currents are flowing through connections, check the millivolt drop between the battery terminal itself and the cable-termination-lug. Large drops of the order of several tens of millivolts and variations between connections, is a good clue. Sometimes temperature-differences between terminal-posts/lugs can be felt.

A inexpensive Clamp-on Ammeter is good enough. Calibration precision is not needed. You are essentially only comparing currents within your system.

Regards
GerryM

[booster]

Checking the amps, as gsm suggests, is a very good idea.

Here is a link to a thread that talks about what I found with our batteries on the bench, checking for possible imbalances or other issue before installing them. Pretty surprising results, at least to me.

http://www.classbforum.com/forums/f5...ical-3660.html

[gsm]

photolimo, booster and All

1 - Booster's experiences with his 12 volt-bench-setup are magnified at six-volts, i.e., lower voltage, high currents. In the E-Trek we have 24 volts, and a stack of four of these.

2 - I have a great respect for booster's knowledge of battery specifications and their significance. It is important to pay close attention to what booster says. However, few RVers have any real experience with multiple stacked battery-pairs, triplets, etc.. Anything beyond a stack of two six-volt batteries is very new, even for RT. That is why RT and we are in this mess.

3 - I hate to say it again. RT is fudging their mess by offering batteries and the like to AGM E-Trek owners to get them out of their hair. However RT appears to continue to refuse to correct their fundamental design errors on the electrical power system of this product.

4 - I look at RT's design and it makes me sick, as a competent electrical engineer I realize that I can offer owners struggling with this design, only placebos, unless they are willing to pursue my suggested reconfiguration using as many of the components already present, putting them to proper use, while adding a modicum of new hardware, replacing unsuitable components and ones that have gone and continue to go defective as a result RT's engineering errors.

5 - The best recipe I can offer requires that owners to do so take the bull by the horns., and follow through on the approach I outlined in Posts #141 and #150 of the thread “ E-Trek Battery Monitor Install”. Or follow the approach suggested by booster in that thread, by six, eight-volt AGMs.

6 - In my view, only once the 24-volt battery-system is performing properly should a battery management/monitoring system be considered. Right now it will only report that there is a problem. It will not allow you to isolate problems within the battery-bank.

7 – Work on the 24-volt battery-banks load/charge terminal only. Keeping a battery balancer, voltage-tap-point, load, etc. disonnected their presence can muddy the true picture.

8 - It is so very important that in each member of a pair of batteries connected in parallel has the same amount (length and gauge) of cable in series with it. This means that the power tap points for series connecting the pairs of batteries from the bottom pair up, has to be located diagonally across each pair. If you take a minute and sketch the wiring of batteries A, B, E and F including the long cable runs front to back, and the termination to battery pair CG, it becomes clear that this cannot work adequately as configured by RT.

9 - I am really puzzled by the connection of the separator. I like you to verify whether the separator shown connected between the 12 volt point on the battery and the starter battery is actually present. I doubt it is there, it could be a mistake introduced in our schematic. But if it is there it has implications as described in Posts #141 and #150, referred to above.


10 - I am sorry I cannot be more immediately helpful than I have been.


Best Regards All
GerryM

[booster]

I think I understand what you were saying, but to clarify, we have a 12v only system, but the batteries are 6 volts. Four Lifeline GPL4CT golf cart batteries at 220ah at 6 volts.

Most of the numbers in the link were from the original test, which had the batteries wired series first then parallel of the series pairs. I later changed to parallel first, series of the parallel pairs, as it made for cleaner wiring. Our cable lengths match nearly perfectly for all 4 batteries.

There was one, very, very, interesting thing that came up when I switched to the parallel first setup. My return (ending, terminal, tail, float transition) amps went down considerably from about 1.7 amps to less than 1.0 amps. This is letting it run in absorption until the amps quit dropping. I have no explanation for it as the I think the impedance should be the same for both ways. The only thing I can think of is that maybe having the pairs of batteries "averaging" with each other at 6 volts is more effective than doing the averaging after they are at 12v?

The return amps that low is substantially below what Lifeline considers fully charged, which would be at 2.2 amps or .5%. We are closer to .2%. The Magnum charger will only program in 1 amp steps, so I am not sure where I will put it for the transition to float. Being close to 1 amp could cause a problem if it varied a bit do to temp (even with temp correction) and wound up locked in absorption until it hit max time, which I also need to determine by doing a deep discharge or two to see how long it takes to get back to full. They float at .1-.2 amps at 13.2 volts.

I have our setup running in the van finally, with everything mounted and wired as it will be in use, with the exception of getting the inductive pickup at the coach battery (on the positive) wired into the ammeter I have in the cockpit so I know when to shut off the alternator charging. I hope it will give some insight once I get a few test cycles on it.

The more I look at the Roadtrek setup, the less I think it can work in the current orientation, even with rewiring, and such. Having those two batteries up front at the end of the long cable and in the heat, just make it essentially impossible.

As I have mentioned, if it were mine, I would push for Roadtrek to put in the 6 eight volt batteries, and if that wasn't an option, I would gulp hard and ask to have the two front batteries moved to the rear, even if they had to go inside the van to find the space. It wouldn't be perfect, but it sure would be better. I would hope there is a place to put them underneath, though.

[markopolo]

To keep track of topics - the other current e-trek topic is here: http://www.classbforum.com/forums/f5...tall-3984.html

In other models Roadtrek used Sure Power 12V separators that activate at 13.2v and disconnect at 12.8. It's likely the same in this unit but confirmation would be useful.

Both the Separator and the Equalizer provide 12V+ to the FE/AB group (in the design below). I added the Cooper equalizer chassis ground: (it could be routed to the front)

E-Trek Battery balanced layout equalizer.jpg

[markopolo]

And here it is with a DC/DC Converter supplying 12v loads and with the Separator removed (no charging chassis battery from solar or when plugged in).

E-Trek Battery balanced layout converter.jpg