Voltage drop of panels

BB. · April 2016

Vic,

For me, a <cntrl-Z> will "restore" the previous edit (including the times when I have deleted the entire draft post). There are (for me) multiple levels of ^Z (I can go back several "edits"). Also, if you go back too far--You can use ^Y to restore (reverse the ^Z).

-Bill

Vic · April 2016

Hi Bill,

Thanks, as always, for the tips ... will play with that, but, I was not really Editing the Post ... errr rrr Comment, but just TRYING to Paste a previously Copied Link, am a bit tired, and forgot that only here, really must do a Ctrl V ... sorry, but R Click/Paste IS truly a reflex ... do it without any thinking, and POOF goes about half of a post/Comment, and there appears to be no recovery. Will test the Ctrl Z, and appreciate all of your tips, and level-headed detailed Comments for the benefit of the Community ...

Thanks! Cranky Vic

BB. · April 2016

I have (on firefox) tried the right click (and scroll down to paste in the right click menu)--And it does what I expect...

If I can figure out how to reproduce.... Perhaps I can get something fixed.

-Bill

Vic · April 2016

Hi Bill,

Being the plain vanilla person that I am, this Toshiba Laptop is Win 7HPE, with IE 11. Previously, others, here, have tried to replicate this issue, and said that they cannot, and it must be my computer, which it may.

I do realize that when programmers need to try to accommodate Smart Phones, Hockey Puck Phones and the myriad of all of the devices around, that either there is not a lot of testing of each environment, especially with each of the far-too-many versions of Windoze OS and IE versions, or perhaps in the rush to ship product that the permissible operating environment gets narrower without specifically noting the narrow limits ... have been in the Computer biz, and the old times were much, much simpler with very few variations on OSes, and possible environments ...

Eventually I will learn to generally not R-Click here, and will play with your noted Ctrl Z & Y to see what happens.

Thanks again, Vic

Vic · April 2016

vince said:

All my batteries have been charged again, filled up with water and having an equal charge of 7.56V-7.58V after charging. When I hooked it up, the system went in absorb mode for another hour until it showed a full charge and 200 ah in the systems with 56.7 V. I have a WBjr and shunt and Temp adjustment. The good panel is producing 73-80 V and at times above 400W. That is good enough as my daily use is less than a kW. if I don't use the split level heat pump.
I checked both of the bad panels and they have almost the same Voc at the MC4 connectors of 59.8V and 60.3V. The amp function of the multimeter no longer works and a new one is coming in this week. No obvious discoloration, condensation, tears or holes. Actually, everything still looks new. I'll contact the company I bought these from on Monday and see what the deal is. Thy don't have any of these panels in stock anymore, so I have no idea what they will do.
With 120 cells in a panel, are all these cells in series or could a section of 30 cells be disconnected? Would checking with an infrared thermometer be able to identify cells that are dead?

Hi vince,

So, am wondering how your batteries got 200 Ah charge ... was this in one day? Or was that accumulated over a period of close to one week?

And, if you are saying that your Absorb voltage (Vabs) was at 56.7 volts, that is quite a bit too low for Flooded batteries, unless the batteries were quite warm. You correctly mentioned that you would try to deal with Setpoints later, But, if you are using Default charge settings, these are usually quite low for your Flooded batteries. If your batteries were at 77 - 80 degrees F, when you observed 56.7 volts and that WAS the Vabs, I would guess that this is about 2.5 volts too low for a Vabs. This is all based upon some guessing, which may not be that accurate.

It is important to NOT rely very much at all on what the KID/WBjr states as the SOC, at this point, unless you have had the time to carefully calibrate the information entered as battery Parameters, based upon watching the SG measurements that you have taken over a period of time, and changing the charge parameters, so the SOC indication is fairly accurate for the way that your system is used and charged, on average. There are many variables in the way that batteries are charged and discharged, that affect the actual Capacity remaining, based upon discharge current, charge current verses time, and so on, and many battery Monitoring devices are simply unaware of these factors, and a number more of them.

Does your Inverter have a built-in Charger?

Do you have a Hydrometer or a Refractometer, to use in measuring the cell SGs?

OH, forgot to mention about trying to bypass some of the PV Cells in the questionable Cells, this not something that could be normally be done in modern PV modules. Many times, the PV's junction box is glued shut. But, I am unfamiliar with SunPower PVs.

Thanks for any additional info, Vic

vince · April 2016

It charged up in two days, well an afternoon and a morning so basically one full day or 7-8 hours. I have an hydrometer and will use it the next time. When I used it before, getting accurate readings took some practice. I thought that the WBjr was the best way to get a reliable indication of SOC. The Kid is on the default setting for WBjr. The kid with one good panel pretty much charged it up. The settings that I have were based on this recommendation:
1. BULK STAGE = 10% of Ah (C10) or 21.5 amps
26 amps (max.current) per series string.

2. ABSORPTION STAGE = 2.44 vpc (x 3 cells) = 7.32 v x the # of batteries in your strings(4)=29.28V (29.3v) X2=58.6

3. FLOAT STAGE = 2.34 vpc (x 3 cells) = 7.02 v x the # of batteries in your strings(4)=28.08V (28.1v) X2= 56.2

4. EQUALIZE STAGE = 2.53 vpc (x 3 cells) = 7.59 v the # of batteries in your strings(4)=30.36V (30.4v) X2= 60.8

This is at -3 mV / cell / °C1

My inverter has no charger so I just bulk charge them in pairs with a regular car battery charger to 80% and let the panels do the rest. I found two cells that were discolored in one panel and a scratch of white paint on another panel but neither could account for a 25% drop. The junction box is glued on and shows no sign of damage. Everything seems working ok today, system was floating and charging 68W.

Vic · April 2016

Hi vince,

Well, very good that you DO have that Auto type charger, and that you were finishing the charge from the one PV + the KID.

To me, even that 58.6 Vabs may be a bit low, but you can fine-tune the Vabs, and the Absorb time, based on the SG readings on your next time out. Vfloat seems high, and EQ seems low, BUT, these batteries may well be Lead-Calcium types, which seem to be more common in the past few years. It has been about 20 years since I have run GC2 batteries ...

What is the way that the Absorb stage is ended -- only by a time setting in the KID, or by Shunt End Amps?

Regarding cell discoloration, some cells from different lots, can change color in varying ways, sometimes, one never knows the reason for and the effects of cell discoloration (at least not me).

Here is a very good article on Measuring SG of Flooded batteries from Surrette:
http://support.rollsbattery.com/support/solutions/articles/4347-measuring-specific-gravity

Believe that your Target SG may sell be in the 1.277 - 1.280 range for full charge, so any SOC vs SG table may need to be adjusted by about 5% upward ... have not looked at that article, recently, but almost all Surrette Solar batteries have a 1.265 SG Target,

OH, and the Defaults on the WBjr may not be appropriate for most Flooded batteries. Entries like the battery Efficiency factor, and Ah Capacity, will probably need to be changed for your batteries. Most Flooded batteries have an Efficiency factor of about 85%, and may well diminish with battery age. The Reference Temperature should be 25 C, which should be the Default ...

More Later, glad that you seem to have made good progress this weekend, and perhaps you can get back to your retreat being more retreat-like in the future. Thanks, Vic

vince · April 2016

I think that I have the absorb stage end by time, 2 hours. I may have to increase that to 3 or 4 hours, or use 1% of the 215ah so 2.15 A, or both?

vince · April 2016

Just found out that the junction box can be opened and that likely one of the four wires might be disconnected. Hope that is something I can fix.

westbranch · April 2016

Vince, can you post some pics?

vince · April 2016

westbranch said:

Vince, can you post some pics?

The inside of the junction box?

westbranch · April 2016

yup..

vince · April 2016

Ok, don't see how to post a pix. Didn't see any lose connections although I measured from left to right, there are four strips of metal, a voltage of 20V for strip 1-2, 60V for strip 1-3 and 0 V for strip 3-4. So most likely the fourth strip is not adding any voltage.

westbranch · April 2016

to add pics , look at the icons starting with the "BOLD B"above the new post box, to the one that looks like a folded corner on a page, that dumb looking thing is for attachments...

Vic · April 2016

vince said:

Ok, don't see how to post a pix. Didn't see any lose connections although I measured from left to right, there are four strips of metal, a voltage of 20V for strip 1-2, 60V for strip 1-3 and 0 V for strip 3-4. So most likely the fourth strip is not adding any voltage.

Hi vince,

It is possible that a Bypass Diode has failed, on each of the two PVs with 60-ish Vmps. Such a failure would probably not be obvious, although you could look at the solder connections on the diodes, although those might be crimp-connections these days.

Vic

vince · April 2016

How can I get something like that fixed. I can get a refund but with the limited space, these panels were ideal.

vince · April 2016

Inside the junction box.

westbranch · May 2016

If that is the diode , the black round piece, on the right side of the picture, it got real HOT at some time... Vic or others can tell you if that is a diode, I would just be guessing...

BB. · May 2016

Yes, they are diodes.

With the panel in light, you can measure the DC voltage across each diode. If they are they same (roughly 1/3 of Voc rating), then all is OK. If one of them reads low, then probably the diode is shorted (diodes fail both shorted and open).

If the diode is failed shorted, you can remove it (with a soldering iron)--And if the voltage goes back up to equal the other diodes--Then that diode is bad and needs replacing (can get some from Digikey or similar).

-Bill

vince · May 2016

Bill, I took these measurements: Starting from + side Across from diode 1, I measured 21V, across diode 2, 42V and across diode 3 there was no voltage. My Voc is about 80V but I'm only getting about 60V. Seems that the third diode has failed?

BB. · May 2016

Can you measure across each diode?

I am not sure why you get zero volts... If nothing else, you should get 60 volts on that last terminal.

What are the Voc/Vmp/Isc/Imp rating of the panels?

-Bill

vince · May 2016

How can I tell if it is a failed bypass diode versus an interrupted circuit that forces the flow through the diode. I checked the temperature of the cells with a infrared thermometer and noticed a difference in temperature among cells of 145 versus 130 F. What would that suggest?

vince · May 2016

BB. said:

Can you measure across each diode?

I am not sure why you get zero volts... If nothing else, you should get 60 volts on that last terminal.

What are the Voc/Vmp/Isc/Imp rating of the panels?

-Bill

I believe that is what I did. Rated Voltage V mpp 72.9 V; Rated Current I mpp 5.97 A; Open-Circuit Voltage Voc 85.6 V; Short-Circuit Current Isc 6.43 A; It's zero volts over the third diode. I thought that it would be a third but it is not.

BB. · May 2016

OK, I thought you left one probe on the left contact and then measured with the other probe. That would account for 21v then 42 on the next... If you measured across each diode, I am not sure why you got this range of voltage.

Guessing, each diode should see ~85.6 volts / 3 = 28.5 volts across each diode (three strings of cells).

It would not hurt anything if you unsolder the diode with "zero volts" across it and replace it with a new diode. Bypass diodes are only needed if you shade one or more cells. If you lift the diode and do not replace it--Make sure you do not load the panel and shade any cells... That can cause damage to the actual solar cell.

It looks like the diodes are 10SQ100 (D suffix means?):

http://www.dccomponents.com/products/Rectifiers/Diode/Schottky/10SQ030-10SQ100.pdf

-Bill

vince · May 2016

Can't find a place to buy them online. Must be looking the wrong way unless I order large quantities from China. Found these on Amazon but they are 1000V. http://www.amazon.com/TOOGOO-Molded-Plastic-Rectifier-Diodes/dp/B009IN1KB8?ie=UTF8&ref_=cm_sw_em_r_lm

Vic · May 2016

vince said:

Can't find a place to buy them online. Must be looking the wrong way unless I order large quantities from China. Found these on Amazon but they are 1000V. http://www.amazon.com/TOOGOO-Molded-Plastic-Rectifier-Diodes/dp/B009IN1KB8?ie=UTF8&ref_=cm_sw_em_r_lm

Hi vince,

Am also having trouble finding a Domestic USA/NA supplier for that diode -- 100V 10A Schottky Axial Lead ...

AND, in also looking for one at 12A or 15A also yielded no results ... will try to look further, today.

Schottky diodes are a bit on the fragile side. Higher voltage Schottkys are less common than are lower voltage units. And, soldering them into a junction box, requires a bit of finesse. One probably needs a soldering iron with a fairly fat tip to retain heat well, as the diode is soldered onto those heat-sinking bus bars in the J-box, so the soldering job is accomplished quickly ... and so on.

You have said that in the initial months after installation, that you system did work well (IIRC). Several things could have caused a failure of these diodes; Shading of the subject PVs, Transient voltage event, like Lightning in the area or a transient within your power system[perhaps], poor Quality Control, or infant mortality on/of the diode, etc.

Will keep looking. Good Luck, Vic

Dave Angelini · May 2016

I am thinking that these panels are the Sunpower rejects resold from the used car lot in Miami. I hope I am wrong!

BB. · May 2016

It could also be a failure elsewhere in the panel... At this point, I am not sure that the diode(s) are shorted.

Any chance for Warranty Replacement?

-Bill

Vic · May 2016

vince said:

How can I get something like that fixed. I can get a refund but with the limited space, these panels were ideal.

vince has stated that the supplier of these three SunPower PVs has none available, but he could get a refund. However, with limited space, and his Tiny Home being set up for these exact PVs, assume that this exact, and somewhat unique PVs would be the ideal replacement.

Two 72 Cell PVs should have a String Vmp almost exactly the same as the SP PVs, although, one has to wonder about the future reliability of the one SP PV that appears to be working ... probably they all should be returned for a refund, and perhaps there is some source for a few of these exact PVs, although, as stated before, believe that these PVs are sold to installers, not into the Retail channel ...

FWIW, Vic

vince · May 2016

Yes Vic that is the dilemma I'm contemplating. I can get a refund but for a 48v system, there are few panels doing the trick.

Voltage drop of panels

Comments

Categories