Omnipower batteries & PSS Inverters - BIG problems...

[GooseD]

Good day all!

Thank you very much for reading this - I really hope I can get some insight as to the problems we are experiencing....

We have multiple sites running with PSS SH-series inverters (http://pss.co.za/wp-content/uploads/2017/06/PSS-Distributors-SH-Series-Inverter-Brochure.pdf) together with Omnipower OPR batteries. When we install the system, we test current draw to estimate backup time.

Example A:

For example we have a site with a PSS SH-600 inverter and one Omnipower OPR120 battery. We set the charging of the inverter to 20A as per battery specification. We have a current draw of 5.56A from the battery during power outage, and this estimates at least 10-15hours backup time - and it works nice at first. But after the batteries reaches it's first "Low battery" state, it seems like it never charges to full again. We can confirm that the inverter switches off at 11V and this dictates a flat battery if we look at the specifications.

This site was left on "charge" for 36hours, and this was the discharge voltage graph - it only lasted 1hour15mins before it had low battery shutdown:

 

We are experiencing failure of these combinations one after another, and most sites are not even 1 year old. We are not getting any reliable assistance from the suppliers so far - with only a confirmation that the "PSS inverters and omnipower batteries are compatible".

 

Example B

Another example is with a PSS SH-2000 inverter and 2x Omnipower OPR240 (to get the 24V needed) - and I used this to test my theory of the charging problem. This setup has a current draw of 15A when there is a power outage. I have started measuring voltage and charge current just after the low battery alarm started on the inverter:

time    voltage    current   
19:39:00    22,8    -15    low batt alarm starts
19:41:00    27,1    24,5    charging begins
19:43:00    27,3    25   
19:46:00    27,2    26   
19:53:00    27,2    29,5   
20:00:00    27,4    29   
20:05:00    27,2    30,5   
20:10:00    27,5    28,8   
20:18:00    27,4    29   
20:30:00    27,8    29,6   
20:40:00    27,9    29,1   
20:50:00    28,5    27   
20:55:00    27,7    15    bulk stopped as soon as voltage spiked to 29V - but this seems to soon as we expect bulk charging up to 85% capacity?
21:00:00    27,8    13   
21:07:00    27,8    12   
21:15:00    27,9    11   
21:20:00    27,8    10   
21:35:00    27,8    7   
21:45:00    27,8    7   
21:55:00    27,9    6,5   
22:05:00    27,9    6   
22:45:00    27,8    4   
07:45:00    27,9    2    

 

Voltage graph of charging:

 

After the batteries was left on "charge" for 19 hours, we had a power interruption and this is the discharge voltage graph:

Discharge amps was constant 15A and these are 200ah batteries. We expect 9-11 hours out of them, but got a low battery alarm only after 2 hours...

 

Please any assistance / ideas will be greatly appreciated...

Thank you very much!

Edited October 18, 2019 by GooseD

[Jaco De Jongh]

2 hours ago, GooseD said:

- I really hope I can get some insight as to the problems we are experiencing....

Welcome Goose

I am afraid things dont look good for you. To have that volt drop (Sagging) with low loads like that and looking at the little bit more than 2 hours it takes for Example B to go from low alarm (Under Voltage) to full, your batteries might be damaged. I assume there is no battery monitors on either of these systems?

Do you have any data for the time when the voltage really  starts to drop away, What was the real current draw at that time? Can you maybe post that as well?

Just maybe you are lucky and its a nice big load creating the voltage to drop to the Cut off voltage. Look at example B, the moment it switches off the voltage goes straight back up to over 24.  

EDIT: For example A, do you have current readings from 16H20 - 15H00 and for example B, current readings from 17H20 to 18H00? Please.

[GooseD]

Hi Jaco,

Thank you for your time

2 minutes ago, Jaco de Jongh said:

I am afraid things dont look good for you. To have that volt drop (Sagging) with low loads like that and looking at the little bit more than 2 hours it takes for Example B to go from low alarm (Under Voltage) to full, your batteries might be damaged. 

I hear you... But surely it cannot be that we have 6x OPR240 batteries that are faulty from manufacturing.... Example B batteries are only about 3 months old with only 3-5 cycles to about 50%... The only time they were taken down to "Low Battery" was now with my testing. And immediately it started behaving the same as the other sites.

4 minutes ago, Jaco de Jongh said:

Do you have any data for the time when the voltage really  starts to drop away, What was the real current draw at that time? Can you maybe post that as well?

For example B Current draw was constant at 15A

 

New test:

Something I just tested now: The OPR120 battery with the PSS SH-600 was charging at about 3-4A (Where I was expecting 20A) and I then added a CTEK MXS 5.0 to the mix on AGM mode. It immediately took the voltage higher up to ~14.9 from the original 13.8V that the PSS provided on charging. The CTEK stated charging at 4-5A while the PSS stopped its charging.

Here is the voltage graph for this scenario:

[Jaco De Jongh]

7 minutes ago, GooseD said:

Something I just tested now: The OPR120 battery with the PSS SH-600 was charging at about 3-4A (Where I was expecting 20A) and I then added a CTEK MXS 5.0 to the mix on AGM mode. It immediately took the voltage higher up to ~14.9 from the original 13.8V that the PSS provided on charging. The CTEK stated charging at 4-5A while the PSS stopped its charging.

As a battery charges and gets closer to full, the charge current he can accept gets less and less, so it might take the 20 amps after if was discharged for a while, but you will see the current drop like in your previous post.

After 17H40 on example B's graph you will se a few places where the voltage saged and came back up again, What was causing that? that is a load switching on and off a couple of times.    

[GooseD]

30 minutes ago, Jaco de Jongh said:

Just maybe you are lucky and its a nice big load creating the voltage to drop to the Cut off voltage. Look at example B, the moment it switches off the voltage goes straight back up to over 24.  

Unfortunately that jump in voltage is when I turned on the generator, load was constant at 15A at that stage.

 

30 minutes ago, Jaco de Jongh said:

DIT: For example A, do you have current readings from 16H20 - 15H00 and for example B, current readings from 17H20 to 18H00? Please.

Sorry for the confusion, Example A constant draw of 5.6A @12V and Example B constant draw of 15A @24V

 

12 minutes ago, Jaco de Jongh said:

As a battery charges and gets closer to full, the charge current he can accept gets less and less, so it might take the 20 amps after if was discharged for a while, but you will see it drop like in your previous post.   

It makes sense, but is seems like it stops taking charge way too early

 

30 minutes ago, Jaco de Jongh said:

I assume there is no battery monitors on either of these systems?

I only have these voltage graphs for monitoring, unfortunately no current graphs, but the current draw is constant.

 

Edited October 18, 2019 by GooseD

[Jaco De Jongh]

1 minute ago, GooseD said:

It makes sense, but is seems like it stops taking charge way too early

That is my concern as well. 

2 minutes ago, GooseD said:

but the current draw is constant.

If this is correct, then you have problems as your Thread name states. I am worried about that little drops in the voltage (Saw Tooth Effect) close to the end of Example B's Graph. That is typically caused by additional load being applied to the inverter/UPS.

[Coulomb]

Discharging any lead acid battery to 11 V, even "deep cycle" batteries (many so-called deep cycle batteries aren't really) is very bad for life. You only get about 4 cycles from them that way, then they're ruined. So you have to discharge to a minimum of 50% state of charge, wasting half the capacity. For better life, discharge to only 80% SOC (leaving 80% capacity remaining, using only 20%). That means you need a battery 2x or better still 5x as large for the same load, with corresponding increases in cost.

That's one of the main reasons people are switching to lithium batteries, though you still have to "waste" (not use) 20% or so for good life.

[GooseD]

11 hours ago, Jaco de Jongh said:

After 17H40 on example B's graph you will se a few places where the voltage saged and came back up again, What was causing that? that is a load switching on and off a couple of times.

Load was constant at 15A throughout the test. When I look at the discharge voltage specification of the Omnipowers, it seems "normal" for the voltage to stay high, and only sag at the last part. But if i discharge at 0.1C I only expect to see that sagging from 2 hours and then flat at 10 hours:

 

8 hours ago, Coulomb said:

You only get about 4 cycles from them that way, then they're ruined

Hi Coulomb, thank you for your post! I understand the batteries out there can be quite "dodgy", but to my understanding Omnipower is a reputable brand?

We use them only for backup purposes, so looking at the Cycle life we expect at least 300 cycles if we were to discharge every time to 100%, and this does not happen very often.

Some of the sites starting to give trouble was never brought down to 50% - It is as if with each discharge, it was never charged to full again, and then after a couple of discharges it is brought to low battery.

8 hours ago, Coulomb said:

That's one of the main reasons people are switching to lithium batteries, though you still have to "waste" (not use) 20% or so for good life.

It is worth noting: it looks like PSS cuts out on 11V under load but the Omnipower specifications states that 10.8V under load is the 100% mark.

For the every now and then power interruption, even with load shedding, we do not see the use for lithium at this stage as we have a very little, constant discharge current. Capacity at this stage is more important...

 

I am still waiting to hear from PSS in regards to this problem, could it be that they are not compatible with AGM / GEL? Or is the voltage shown in the graphs enough to charge them?

[Jaco De Jongh]

21 minutes ago, GooseD said:

Load was constant at 15A throughout the test.

Hi Goose

Something here just dont add up. What do you use to measure this constant load?

If in example A you draw 12V * 5.56A = 66Watt and the batterys goes from full to low voltage alarm in 1 hour, your batteries are stuffed, bottom line. 

Same with example B 24V x 15A = 360 watt and the battery reaches low voltage alarm in 2 hours, the batteries are also damaged in my opinion. 

Was your amps measured on the DC side?

[GooseD]

29 minutes ago, Jaco de Jongh said:

What do you use to measure this constant load?

https://www.major-tech.com/product/400a-compact-ac-dc-clamp-meter/

 

30 minutes ago, Jaco de Jongh said:

If in example A you draw 12V * 5.56A = 66Watt and the batterys goes from full to low voltage alarm in 1 hour, your batteries are stuffed, bottom line. 

Same with example B 24V x 15A = 360 watt and the battery reaches low voltage alarm in 2 hours, the batteries are also damaged in my opinion. 

Sure, but I am thinking they are not charging.... I unfortunately do not have any other charger that can give the required charge current for testing.. I only have the PSS inverters...

 

31 minutes ago, Jaco de Jongh said:

Was your amps measured on the DC side?

That is correct. The DC amps made sense as we know what the constant load was on the AC side - The load is always constant at our backup sites (perhaps a deviation of 1AMP on the DC side now and again)

[Jaco De Jongh]

56 minutes ago, GooseD said:

The DC amps made sense as we know what the constant load was on the AC side -

Please tell me what the constant AC load was?

 

 

56 minutes ago, GooseD said:

but I am thinking they are not charging.

 

17 hours ago, GooseD said:

ime    voltage    current   
19:39:00    22,8    -15    low batt alarm starts
19:41:00    27,1    24,5    charging begins
19:43:00    27,3    25   
19:46:00    27,2    26   
19:53:00    27,2    29,5   
20:00:00    27,4    29   
20:05:00    27,2    30,5   
20:10:00    27,5    28,8   
20:18:00    27,4    29   
20:30:00    27,8    29,6   
20:40:00    27,9    29,1   
20:50:00    28,5    27   
20:55:00    27,7    15    bulk stopped as soon as voltage spiked to 29V - but this seems to soon as we expect bulk charging up to 85% capacity?
21:00:00    27,8    13   
21:07:00    27,8    12   
21:15:00    27,9    11   
21:20:00    27,8    10   
21:35:00    27,8    7   
21:45:00    27,8    7   
21:55:00    27,9    6,5   
22:05:00    27,9    6   
22:45:00    27,8    4   
07:45:00    27,9    2    

This tell me it is charging, but to quick, so either the batteries are shot, or the PSS has got a very unusual charging algorithm with some kind of current control. 

[Coulomb]

11 hours ago, Jaco de Jongh said:

This tell me it is charging, but to quick

Yes. It's like there is no absorb stage at all; it seems to go from bulk straight to float. Maybe that's because the PSS inverter is essentially a UPS.

The specifications list the charging voltage as 27.6 V; that's float voltage. 29 V at the end of the bulk phase is just over 28.8 V, the usual end of bulk and start of absorb voltage for flooded lead acid batteries.

@GooseD, I think you need to find a way to get the PSS inverter to do a proper absorb stage, use a separate charger, or use a different inverter-charger. Hopefully your batteries will survive this chronic under-charging without too much damage. Lead acid batteries don't appreciate under-charging. I don't know if AGMs are hardier in this respect.

[Richard Mackay]

Do the different sites have the same equipment?

What is the backup power being used for?  

[PJJ]

On 2019/10/18 at 8:01 PM, GooseD said:

Omnipower OPR240

We have one of these doing fairly light work at a building with no AC, but the total system consumption is only around 40W.

It just power a few routers, 12V Lights and a few IP cameras.

One average it does 13% DOD daily, the lowest I have ever seen it go was 52% SOC (3 Days of terrible solar production).

Its just over a year old now, and its still looking good.

[Richard Mackay]

I presume all the sites have a similar setup? (It just power a few routers, 12V Lights and a few IP cameras)

Are you using POE? If so what voltage do the cameras use?

[PJJ]

1 hour ago, Richard Mackay said:

I presume all the sites have a similar setup? (It just power a few routers, 12V Lights and a few IP cameras)

Are you using POE? If so what voltage do the cameras use?

We only have 1 location with that battery

I didn't want to use the DC from the battery because the battery voltage does go to 14.5V in absorption (13.6V float) and my IP cameras specify 12VDC +-10%

So that means only 13.2VDC max.

We needed AC power however for the lighting, so there is a small 250VA Victron Phoenix inverter, so I just added 12V powersupply to its output.

I could have probably just added a DC-DC buck converter, but I didn't have one at the time.

[Richard Mackay]

2 hours ago, PJJ said:

We only have 1 location with that battery

 

And what PSS inverter is this connected to?

[PJJ]

3 minutes ago, Richard Mackay said:

And what PSS inverter is this connected to?

No PPS inverter, that's why I only qouted the battery model, since the only thing I have in common to the OP.

Charging is only from a Victron 150/35 Solar Charge Controller.

AC is from a Victron 250VA Phoenix

[Richard Mackay]

Take one or two batteries to the main agent (not the outlet where you bought them) and get them to check them. 

Do this whether they are under warranty or not to clarify what the issue is.

It is possible for new bats to fail. I had  an Energiser 102Ah deep cycle that wasn't performing. After lots of hot air they tested it over a day or two,  admitted it was faulty and replaced it.  

Edited October 24, 2019 by Richard Mackay

[Anarchy2007]

@GooseDI'm sorry to hear about your issues. And can only contribute to your agony. We have installed 4 sets of Omni power batteries and every one of these had failed well before the 1 year guarantee period. All with the same issues as described in your post. A battery monitor was connected to the last set of 4x OPR240s. I do not have the graphs with me at this moment, but can tell you that the highest temp recorded was 24 degrees, the lowest discharge rate was around 70% SoC. The highest charge rate was around 35A - 40A. Cycle wise, none of these batteries had more than 50 cycles at an average discharge of around 10% per cycle (90% soc). Two sets were credited under warranty, the 2nd last which coincidentally was connected to the battery monitor, was rejected. Within a month (during a single cycle) the battery monitor showed a total disacharge of eg. 30A and once charge completed (charge rate of less than 1A) the battery monitor would indicate only 91% SoC. Checking this it was noted that the recharge rate was almost always less than the discharge rate. This set off red flags. These batteries were returned to Sintech and the warranty claim was rejected. Their reason, possible overcharging. With a data sheet that claims 50A max charge rate and up to 60 degrees Celsius temp rate. That is kind of impossible.

The reality is that these batteries are really bad quality and the specs claimed in the data sheet is nowhere near the real values. I have had clients that used realy cheap rubbish batteries that lasted much longer than these under the same conditions.

[Okkie]

On 2021/04/29 at 11:46 AM, Anarchy2007 said:

@GooseDI'm sorry to hear about your issues. And can only contribute to your agony. We have installed 4 sets of Omni power batteries and every one of these had failed well before the 1 year guarantee period. All with the same issues as described in your post. A battery monitor was connected to the last set of 4x OPR240s. I do not have the graphs with me at this moment, but can tell you that the highest temp recorded was 24 degrees, the lowest discharge rate was around 70% SoC. The highest charge rate was around 35A - 40A. Cycle wise, none of these batteries had more than 50 cycles at an average discharge of around 10% per cycle (90% soc). Two sets were credited under warranty, the 2nd last which coincidentally was connected to the battery monitor, was rejected. Within a month (during a single cycle) the battery monitor showed a total disacharge of eg. 30A and once charge completed (charge rate of less than 1A) the battery monitor would indicate only 91% SoC. Checking this it was noted that the recharge rate was almost always less than the discharge rate. This set off red flags. These batteries were returned to Sintech and the warranty claim was rejected. Their reason, possible overcharging. With a data sheet that claims 50A max charge rate and up to 60 degrees Celsius temp rate. That is kind of impossible.

The reality is that these batteries are really bad quality and the specs claimed in the data sheet is nowhere near the real values. I have had clients that used realy cheap rubbish batteries that lasted much longer than these under the same conditions.

 

It is a pity I did not see these comments sooner. I would just like to share my experience as well which will confirm the statement made re the bad quality of the Omnipower batteries. I have had my 5KVA Inverter for 4 years and after 3 years had the replace my 4x100a/h batteries. Replaced by 4x120a/h Omnipower batteries. More expensive than the first set I had. After 1 year, exactly one week before end of the year, the batteries would not last 1 hour during load shedding. In the past we would enjoy at least stage 2 with the 4 and half hours load shedding. Had these batteries checked in at Sinetech via Communica. After following up with Sinetech numerous times, they came back after 3 weeks saying that there was nothing wrong with the batteries. I collected the batteries from them . Had the Inverter checked out by Mustek and there report also showed no default on Inverter. Had batteries checked out by an electrical expert and his report showed that batteries were at fault. Did not hold the voltage. This report was printed and forwarded to Sinetech. It was followed up a week later with a second email to Sinetech, again with no response. This just conforms the previous comments on this Forum that the quality of the Omnipower batteries is not up to standard and want to add that the backup support is also non existent.   

[BenBB]

Yes agree.. Omnipower batteries are crappy to say the least and Sintech is not a company I will ever use again. Poor service; bad response warranty avoiders and that from the CEO!! And they are making bags of money selling loadshedding systems by the dozens and not carring about the projects failing because of poor quality batteries!