SQFlex Problems

System · August 2007

Hi
I am new here. I have a problem that has totally stumped me {so far}
I have a SQflex 11 SQF 2 200' down our well. I have 6 Mitsubishi 130 panels on a Wattsun tracker 125' from the well. The SQFlex has been in the well for a year now and has been working just fine with a generator pumping to our 2500 gallon storage tank that is about 80 vertical above the well and then gravity flows to the house and gardens. I got the solar panels all hooked up running through the Grundfos generator interface box. The OC voltage of the 6 Mitsubishis wired in series is 134v. When I connect the pump to the panels everything works great for a while {sometimes 10 minutes, sometimes 2hours} and then the voltage which under load is sitting at about 102 to 105v with the pump running just fine starts jumping up and down for a couple of minutes eg. 50 to 105 and then settles down and stays at 34v which is still running the pump but not enough to push any water out of the well. If I flip the breaker off and back on it immediately starts again and runs just fine for a while. I have checked and re-checked all the wiring. I have run the tracker on a seperate power supply, tried wiring past the generator interface box, tried takin out 1 panel at a time and running on 5 panels and still the same thing with the voltage dropping to 34v. I have spoken to Grundfos tech support a couple of times and they feel the problem is in the panels and not the pump because it works OK on the generator. Anyone have the answer?? or ideas??
Thanks Roger

BB. · August 2007

Re: SQFlex Problems

Is there a Linear Current Booster installed somewhere in the circuit?

If there is no LCB, and if the SQflex controller does not have that function, I could see the pump motor starting strong (high voltage, not too high of current since well water starts high and there is little head (if no check/foot valve), then pump starts loading down (drawing more current, panel voltage falling)... Then a little power drop in the panel output (rising temperature, shadow from cloud, bird, etc.) or the pump draws a little extra current, and all of a sudden the panel voltage falls (panels, more or less, give a fixed amount of current based on the amount of sunlight) and the pump slows down (wanting more current--but there isn't any extra)--and you are at a stable situation--pump turning slowly, not enough current, voltage low, and it does not recover.

If there is a working LCB installed--perhaps there are not enough panels (or enough current?) for this system? Or, the LCB is bad?

Have you measured the current (and voltage on generator and on panels?).

Just a guess.

-Bill

Windsun · August 2007

Re: SQFlex Problems

It could also maybe be a wiring problem or just a loose wire or terminal in the panels - are they all hooked in series in a single string?

We have never seen a Mitsubishi panel go bad, and the fact that the voltage drop is so much leads me to believe that is not the case here, unless perhaps a bypass diode got zapped.

One easy way to check the panels would be to just put a 125 watt bulb on them for a while and see what happens.

System2 · August 2007

Re: SQFlex Problems

Thanks for the replies so far. There is no LCB in the circuit. The SQFlex has a "soft start" and has a controller in the pump which has a MPPT function while running on DC {as I have read and understand it so far}.
The panels are all connected in one string.
I have had this problem for a few days now and have checked and rechecked the wiring a few times. This morning I went out early to reconnect the panels after checking them with 5 at a time yesterday. So far this morning the pump has been running about 1 and a half hours at a fairly steady 107/108 volts a little higher than before but maybe this is because the panels are cold. Yesterday the pump ran until about 10 am and then did its voltage drop voodoo. I am going to see how long it lasts today. Being an off grid rural home I do not have any 125 watt light bulbs sitting around but I will try to find one from somewhere and try that. If the panels are connected in one string and a bypass diode was zapped, would I still read voltage for all panels, which I do?? OC voltage 134v .
Roger

System2 · August 2007

Re: SQFlex Problems

Update.....Pump ran till about 10.30 today then same thing. I restarted it and now it will only run 10 minutes before voltage oscillations and drop. Still trying.
Roger

BB. · August 2007

Re: SQFlex Problems

Can you measure the current from the solar panels (and the voltage at the same time)? The current, if the panels are working OK, should stay the same (or increase slightly) when the pump voltage drops from 110 VDC to 34 VDC--if the current drops (while the voltage is low) there is a solar panel wiring/panel problem.

Can you measure the AC power from your generator? Is the pump taking more power (watts) than the solar panels are capable of driving?

It appears the pump should slow down if power drops (?).

I assume no error messages are displayed?

I don't know for your pump, but for a simple centrifugal pump, you can put a valve on the pipe and cut the volume down--and the pump's power requirements will also drop. Test to see if the pumping stays stable at reduced flow.

Your pump appears to be a more along the lines of a positive displacement pump (because of the high head). I am not sure that adding a flow restriction valve will reduce power requirements or actually increase power requirements. If you don't leave the valve closed, the pump should not overheat if you experiment with reducing flow by 1/2 and see what happens.

-Bill

Windsun · August 2007

Re: SQFlex Problems

The fact that it drops out after the sun is up a while indicates there might be a heat problem someplace, possibly in the panels or associated wiring or above ground controls or something.

1. Try letting the pump rest a while. If it does the same thing after it cools down for 30 minutes or so, that pretty much eliminates the pump.

2. Water down the panels to cool them off, see if that makes any difference for a few minutes.

3. Check any solder connections if there are any inside the J-boxes on the panels. Not likely but a cold solder joint is possible.

But without knowing what the actual amperage from the panels or to the pump is doing, kind of just guessing here. It does not really sound like a panel problem with that large a voltage drop. I could see one panel going bad and dropping by that amount but not 75 volts.

It is also possible that you are underpowered with those panels when they start getting hot and putting out less power, and the pump is trying to "downshift" the LCB. If that is the case, then spraying them down with water should fix it temporarily.

BB. · August 2007

Re: SQFlex Problems

I would still like to see the current/voltage numbers when the pump is running correctly vs oscillating, vs "failing"...

However as a another approach to debugging the problem--I am wondering, whether or not one could place one or two 12 VDC 100AH (car type batteries) in series with the existing solar panels to simulate adding 1-2 additional 12 volt (or 1x 24 volt panel) to the string. Saves spending $xxxx just to find out the problem still exists.

Using real car batteries (in otherwise good shape) should run 8 AH for 2-3 hours of testing without causing any problems (~20% discharge). Adding 10-15 amp fuse to the battery lead would help reduce the chances of dangerous currents.

The SQ Flex seems to be limited to less than 9 amps in all configurations (assuming my skim of the product manual is accurate). So, adding high current capable batteries should not be a major issue for testing.

Anyone can thing of any reason to not do this for testing of this system? Will it yield meaningful information?

-Bill

System2 · August 2007

Re: SQFlex Problems

Hi Thanks again for all the ideas. I do not have a good way of measuring current which is why I have not posted that info.....but I did call a local friend of mine who is an electrician {with a background in software} and does a lot of work on irrigation pumps locally. He is usually dealing with 3 phase large capacity pumps connected to the grid, so he asked me to email him all the literature for the SQFlex pump and he grilled me on how the system was wired. These were his ideas on the problem....
<<"I am pretty much convinced that the problem lies in the configuration. and under certain conditions the pvs on one side and the MPPT logic on the other start to something like resonate. start oscillating with higher and higer amplitudes until it gets stuck at 30Volts or whatever it is, where something in the pump software does the wrong thing. BTW, when the pump gets stuck in the 30 Volts or so, did you ever try to wait for a while to see if it is going to come up again? Yes i have and it does not
It is like if you are driving your car a certain speed over a certain washboard road without shock absorbers. if you hit the right speed so that your suspension resonates with the road surface your entire vehicle will start to move violently probably to a point that something breaks. this oscillation is dampened by the shock absorbers. electronic means exist to dampen oscillations in electronic circuits as well.">>
He asked me to try loading the panels a little and see if that changed anything, so I wired in a 90 watt floodlight I found to the panels and started the pump again. It ran for the rest of the day, about 6 hours, lamp shining bright and water constantly pumping!! I wanted to run it for a whole day on Sunday with the load on the panels but we had an unusual cloudy day and the pump started running fine without the load. So today it is back to hot and sunny with no clouds. I started the day without the bulb/load and it ran about an hour and a half before the oscillations and voltage drop. I have now put the bulb/load back in the system and have been running about 5 hours so far today.
It seems that the current from the panels is not a problem as they can run the pump and the 90w bulb for hours at a time. I do not really understand the pump electronics or their interaction with the panels but I think and hope he is on to something here. I am going to experiment a couple more days with the load in the system to make sure it keeps working and then see if my friend can call the tech support at Grundfos to see if it is possible to change the parameters in the pump logic
Roger

niel · August 2007

Re: SQFlex Problems

that's a very odd scenario and it seems you did hit onto something here. maybe try a smaller and smaller load to see if it holds the oscillations at bay.

BB. · August 2007

Re: SQFlex Problems

Hmmmm.... Interesting problem. Sure sounds like there is a stability issue with the electronics and the solar panels inside the pump.

I assume there is nothing "strange" or unusual in the pump/panel wiring--no capacitors, inductors, or batteries in the circuit. Connections are "clean" and tight.

Only physical thing I can think of... The plumbing to the pump, is there any siphon effect--if so, put an air break at the top of the siphon and see if that changes anything (pressure rises as siphon starts and drops as it gets going)--depending on the size and routing of the plumping--there may be some oscillations with the siphon effect.

I guess there can be worst solutions than placing a low wattage light bulb on the solar panel wiring.

-Bill

Telco · August 2007

Re: SQFlex Problems

I'm by no means an expert here, but I've been taking info away and wish to contribute, so... here's how it looks to me. Anyone feel free to correct.

Seems to me like the light bulb, being a load, will help regulate the load seen on the PVs. As the pump pulls more the bulb will dim, and as the pump pulls less the bulb will brighten. The effect (may happen too quickly to see any flicker in the bulb would be to have a constant load on the PVs. If the PVs are showing a constant load, then they would not see any backfeed spikes as the bulb would absorb them. Any time you put load on or take load off a power system, power will spike when the load is off and drop when the load is put on. I'd have to wonder if the pump, being as they are high current items, might not be affecting the PVs on a state change.

Are there any batteries on this? If so adding batteries should fix the issue as batteries will act as a load bank. The batteries will accept the spikes and backfeed without affecting the PVs, and the batteries will have ample power available to handle the spikes when the power is turned on. And, the batteries will allow the pump to run at night or when there is no sun as well. A few deep cycle batteries should be plenty for it.

mike95490 · August 2007

Re: SQFlex Problems

Telco wrote: »

Are there any batteries on this?
If so adding batteries should fix the issue as batteries will act as a load bank.

I don't think batteries will work, as there is no controller, and the batteries will run flat on the first night.

What may work, is a bank of the "Super Caps" you can get at Automotive Audio suppliers. The caps will charge up, and help supply a peak surge. If they don't work, take it back, and get a refund. match the peak voltage your panel provides, may need 2 in parallel, mount as close to pump as possible.

Unless, is the pump AC or DC ?? What does the "Grundfos generator interface box" do, what is it's output ?

Telco · September 2007

Re: SQFlex Problems

Would they run flat? What would cause this to happen? I wouldn't think that a water pump would be running continuously, so didn't think it would kill the batteries as long as there was a diode to prevent current from flowing from the batteries to the PV.

Thanks, asking to learn here.:D

System2 · September 2010

Re: SQFlex Problems

I wanted to let everyone know that I finally got the SQ Flex working correctly. The very simple solution was add more panels therefore increasing the voltage.

I am still amazed that no one else has ever had this problem of the pump stopping working and having to be reset constantly, and also that Grundfos did not suggest this idea. Although my six Mitsubishi panels were slightly under the optimum voltage for the pump, Grundfos only expect a loss in efficiency with lower voltages and not complete stoppage of the pump.

I have added 2 more panels to my tracker and now the pump has been running for about 6 months without a hitch.

Quite frustrating but at least it is now working.

BB. · September 2010

Re: SQFlex Problems

Wow--Thank you for the update... That was a long and frustrating battle.

-Bill

zozomike · August 2011

Re: SQFlex Problems

I know I'm late to this party, but I have almost exactly the same issue, SQflex11, CU 200 controller, 6 KY 130s in parallel with 105 vdc@ load. very similar spec. Exactly same symptom. Grundfos has an issue here. Gone through several levels of experts, with resultant peripheral changes, no avail. This issue only arose when I upgraded panels, previously I had only 2 Solar World 165s and the system worked swell/slow but swell. The issue is not in the CU 200, as I have taken it out of circuit, ( which meant installing a normally closed pressure switch instead of the normal open the CU 200 requires) and had the same issue. My latest idea is to try to separate the six into two strings thereby lowering the voltage? Spec on the pump is 30-300 vdc, but there is something here the SQFLEX does not like. What do you think?

Dave Angelini · August 2011

Re: SQFlex Problems

" 6 KY 130s in parallel with 105 vdc@ load."

How do you get this voltage with parrallel panels?

waynefromnscanada · August 2011

Re: SQFlex Problems

Must have meant to say series?

zozomike · August 2011

Re: SQFlex Problems

yes, my mistake, series of course.

Dave Angelini · August 2011

Re: SQFlex Problems

Lowering the voltage will lower the output. The controller is bad or really rare would be the pump or wiring insulation breaking down.

zozomike · August 2011

Re: SQFlex Problems

Thanks, If I split the 6 panels into two strings using a combiner I would be lowering voltage, but doubling amps, how does that lower output? I have removed the controller from the system to test that and found the same issue. Perhaps it is the pump, but I did not have the issue until I upgraded from 2 panels to six. Actually the same spec as the previous guy.

Dave Angelini · August 2011

Re: SQFlex Problems

Look at the output curve for your pump and factor depth in. That will tell you the DC level that is best if you can supply the current. Test it by measuring the water output vs time. If it is not doing that then the pump is probably in need of service. I have also seen issues where people really do not know what model is down there. Good luck

zozomike · August 2011

Re: SQFlex Problems

I'm sure as a professional you have seen many issues and much ignorance, thank you for taking the time to reply. You will note I stated the pump and system in first post.
SQflex11, CU 200 controller, 6 KY 130s in series with 105 vdc@ load.
Both the pump and controller are rated 30-300 vdc
Water depth measured 85' well depth 130'
The system falls within all specs and none of this explains ( to me) why the pump would work fine until sporadic rapid cycling ( we are talking seconds with a literally bouncing pressure from 0-30) and when reset will resume normal operation. For a time. This only occurs in free flow mode.
Thanks for all your ideas.
Michael

BB. · August 2011

Re: SQFlex Problems

Did you try a ~90 watt bulb and see if it changes the pump's behavior. (don't know why it worked--but it did). Strawfly:

He asked me to try loading the panels a little and see if that changed anything, so I wired in a 90 watt floodlight I found to the panels and started the pump again. It ran for the rest of the day, about 6 hours, lamp shining bright and water constantly pumping!! I wanted to run it for a whole day on Sunday with the load on the panels but we had an unusual cloudy day and the pump started running fine without the load. So today it is back to hot and sunny with no clouds. I started the day without the bulb/load and it ran about an hour and a half before the oscillations and voltage drop. I have now put the bulb/load back in the system and have been running about 5 hours so far today.
It seems that the current from the panels is not a problem as they can run the pump and the 90w bulb for hours at a time. I do not really understand the pump electronics or their interaction with the panels but I think and hope he is on to something here. I am going to experiment a couple more days with the load in the system to make sure it keeps working and then see if my friend can call the tech support at Grundfos to see if it is possible to change the parameters in the pump logic

Later in this thread, adding panels fixed his problem....

-Bill

Dave Angelini · August 2011

Re: SQFlex Problems

zozomike wrote: »

I'm sure as a professional you have seen many issues and much ignorance, thank you for taking the time to reply. You will note I stated the pump and system in first post.
SQflex11, CU 200 controller, 6 KY 130s in series with 105 vdc@ load.
Both the pump and controller are rated 30-300 vdc
Water depth measured 85' well depth 130'
The system falls within all specs and none of this explains ( to me) why the pump would work fine until sporadic rapid cycling ( we are talking seconds with a literally bouncing pressure from 0-30) and when reset will resume normal operation. For a time. This only occurs in free flow mode.
Thanks for all your ideas.
Michael

My point is that the pump that is in the hole may not be the one that is in there unless you know this for a fact. It sounded like that was what you were told. If doing what Bill suggests works then you probably have a bad controller. Good Luck!

SQFlex Problems

Comments

Categories