Optimum panel configuration for heating only

[Noel Barlau]

I just received my shipment from you, and I'm almost ready to connect the arrays to the DMPPT.

If I'm going to use the DMPPT for heating only without also charging a battery, what would be the optimum configuration, considering I have 41x 245w panels available?  Also the wiring run is going to be longer than I'd like, probably about 150 feet.  I've already purchased enough 2/0 aluminum wire to run two lines in parallel to each input.  Also, trying to maximize utilization of the DMPPT, is there an optimal maximum capacity solution?  I'm ready to buy more panels if I can utilize the additional power.

I was going to just use 39 panels as indicated in the manual, but after reading another thread I now realize that's only optimal if you're also charging batteries.

If I could add another couple questions, do you have a recommended solution for lightning protection, and do you recommend circuit breakers somewhere in the circuits for the DMPPT?

Thanks in advance, and I just have to say how impressive the hardware is.  I've heard that art is something you look at and wish you'd created yourself, and your products definitely fall into that category.

[Noel Barlau]

Edit:  It's 2 gauge, not 2/0 alumin(i)um wire.

[Julien]

I think you take into account the Al / copper oxydation ? I didn't thought aluminium cables with the AL/ copper transition pieces was cost effective with this (relatively short) distance but it may be the case. Nevertheless it will be a impressive setup. Waiting for some pictures of it :)

[Dacian Todea]

Noel,

Is that PV cable or will you have that underground ?  I think people will need to start using Aluminium for large PV arrays as there is not that much copper and copper is more expensive.

The DMPPT450 connectors can handle both copper and aluminium directly so that will not be a problem. Not sure if you need two lines in parallel for each input.

150 feet to the PV array is not to bad. You will need to have individual fuses for each panel as you likely want to parallel more than two panels before connecting to this aluminium wires.

You can use all 41 panels not sure what you want to use as heating elements.

You will also need to provide proper cooling to the DMPPT450 that will mean also active cooling with a DC fan in front of the DMPPT maybe two 24V fans in series connected to one of the PV inputs so they are powered directly by the PV array since you will  not have a battery.

As you will not be charging a battery the PV inputs can be distributed more equally so you can have 9+9+8+7+4+4 

Not sure about the sort of heating elements you want to use. Will you use 36V DC heating elements or want to use silicone copper wire?

Lightning protection will depend on your circumstances. There is a direct strike not possible in my installation as there is no return path in my case with a fully insulated house from ground and there is indirect induced current in to transmission cables. You will need to check local regulations as they differ.  There is no lightning usually in winter so you may just be able to isolate the system in summer when it is not used.   

 

[Noel Barlau]

Thanks very much Dacian,

The wire is underground rated, and I'll be burying it after the final configuration.  For testing purposes I've simply propped up the panels at the optimum angle for this time of year at our location using pressure treated lumber and securing attachments which I cut out on a bandsaw.  It's a real hack job at the moment, but I want to figure out the capabilities of the total system before choosing a permanent location for the arrays and DMPPT.

You previously recommended using 36v 1200w water heating elements, which I ordered enough of to create the proper resistances on the DMPPT outputs.  The elements will be mounted into 55-gallon steel drums during the testing phase.  After I'm convinced it's the correct solution for our situation and I can proceed with confidence in the final setup, I'll order the thin stranded copper wire to embed in thin-set in the basement slab, then cover it with tile.

It's great to know the terminals on the DMPPT are rated for aluminum.  It'll save me a transition to copper for the termination.  The paralleling of two lines was, admittedly, overkill.  That's in my nature.  I'll try it with one run first and see what the losses turn out to be, then decide if it's worth running the second lines.

Also it's super to have that number of panels paralleled information.  Will the required output resistances still be the same, or will that change with the number of panels?  I'm sorry to ask this, but I still can't get a grip on the internal logic of how the system works and I want to get it absolutely right lest I let the magic smoke out, which has happened to me in the past.  And I'm going to do my darndest to not have it happen with your equipment.

We're in a very active thunderstorm corridor, and our house is an electrical nightmare thanks to the previous owner.  I'll need to have plenty of lightning protection.  I'm leaning towards the Midnite Solar SPD line at the moment, but if anyone has other recommendations I'm all ears.

Julien, after much research with the same concerns you expressed, I've decided to go with solar wire from the panels - two panels in parallel per run to stay under the current rating of the wire.  These will terminate using split bolt connectors which will be slathered in anti-corrosion compound, then heat shrinked and waterproofed.  This is the best I can do, having done all the research online I could possibly do before deciding on this method.  It seems that the bigger problem than galvanic corrosion is the differing expansion rates of copper and aluminum, which can eventually work the fittings loose.  The split bolt can apply significant pressure which forestalls the joint failure.  That's what I understood from the general consensus online, anyways...  Everything I'm doing with this setup is experimental, and I'm sure there will be failures and a steep learning curve.  I'm expecting surprises to say the least.  If I find any helpful hints along the way I'll post them here.

And I'll try to post some pictures along the way.  But I'm pretty embarrassed at the half-assed nature of my setup at the moment.  I just can't drop 10k on permanent mounts before I'm sure this whole idea will pan out.  If it does, I'll gladly drop the cash into a permanent solution.

This brings up another point - the legality of this type of system.  It really doesn't bother me from a moral standpoint to do experiments with temporary setups, but when this is ready to become a permanent part of the house, how am I going to explain this system to the electrical inspector, and how do I describe it on the building permit application?  Usually things need to be UL-approved to be used as appliances, don't they?

[Dacian Todea]

Noel,

As you have about 10kW of PV panels you will need about 8 of those 1200W heating element's (they are 2x 600W if I remember) so basically 16 x 600W

I guess you can have two of those 1200W elements on each barrel so 4 barrels if is just for testing else you may use 8 barrels for larger thermal storage.

You can have them connected like this

TLD2  -   600W   (so half of one element).

TLD3  -   1200W (one full element).

TLD5  -   1800W (one full element and a half can be the remaining half from TLD2)

TLD1  -   1800W

TLD4  -   1800W

TLD6  -   1800W

 

This is 7.5 elements but still good as is best heating elements to be slightly lower power than PV array as that is STC power rating and in most days you will not be able to get that and with this configuration you get slightly better overall efficiency.

It will be able to select the max power point power in 600W increments.

So all you need to do if you just use the DMPPT is to connect the PV array as mentioned earlier then the heating elements as mentioned above and you need two 10K NTC B3950 thermistors on the side white connector this will enable the heating one will be for TLD 2,3,5 and the other for TLD 1,4,6 but are set to fixed +60C  You can just add a small switch in series with those NTC to disconnect them that will allow you to stop the heating or enable or you can add your own temperature controller to set any other higher temperature and that can disconnect this NTC's to stop that. (you can just use 10KOhm resistors with a switch to enable and disable using an external temp controller).

If you use 4 of those 55 gallon 208 liter barrels that will require 1.16Wh to increase temperature of 1 liter of water by one degree Celsius so say +25C to +75C will require  (75C - 25C) * 208L * 1.16 = 12kWh per barrel so 48kWh thermal storage with 4 barrels or 96kWh with 8 barrels

So in a full sunny day in February when day is longer and still cold you may produce 48kWh in a single day heating 4 barrels to +75C assuming you start for +25C   

[Noel Barlau]

Excellent information, Dacian.  Thank you so much!

I already have the thermistors and somehow it never crossed my mind to use the inputs as switches.  Super easy solution!

It's also a great relief to hear that my thinking was correct about the elements.  It's exactly as I'd worked out on my own and luckily I wasn't thinking incorrectly, which happens quite often when I'm attempting unsupervised math.

[Noel Barlau]

Hello again Dacian,

Finally getting around to testing the DMPPT450.  I have it hooked up as you suggested using the 36v 1200w water heater elements.  There are 39 panels wired as shown in the manual.

The 10k thermistors are hooked up to both ET1 and ET2.  In full sun, TLD3 came on indicating one full 1200w element with two loops was being powered.  Then when it got cloudy TLD3 disconnected and TLD2 lit up, powering only one 600w element.  TLD 1, 4, and 6 never lit up.  I disconnected ET1 to verify it was working properly, and then TLD4 lit and was powering three individual elements for a total of 1800w.  When I re-inserted ET1, TLD4 extinguished and TLD3 lit up.  Then a cloud passed over and TLD2 lit up and TLD3 extinguished.  Now it's full sun again and TLD2 is still lit.

Granted, I'm running each string on 10ga wire, so it does get warm - but not too hot to touch.  Might this be affecting the power output significantly?

I assumed the thermistors would go from ET1 to the ground above it and the same for ET2.  This must be correct because the outputs do work, although not how I was expecting.

I'm sure I've done something stupid to make it behave this way, but it seems like it should be powering more than one element in full sun.  Please let me know what you think it might be.

Also another question - I thought I read somewhere that all 39 panels were providing power at all times.  Is this true?  If so, why is there only one panel on TPV2?  Also, should more than one output be on at a time, or will there only ever be one on at a time?

That's all I can come up with for now.  Thanks in advance for your advice.

[Noel Barlau]

I came up with another question after submitting the prior one - How does the DMPPT450 decide which outputs turn on and in what sequence?  It's acting like it's waiting for what would be the TLD 2, 3, and 5 to hit the 60c set point for hot water storage before turning on TLD 1, 4, and 6 for what would be in-floor heating.  I wonder if I heated up ET1 to simulate the water barrels at 60c if TLD 1, 4, and 6 would become active?

[Josh Peterson]

Did you try to switch the temp sensors?maybe one is bad?

[Noel Barlau]

I didn't actually swap them, but since after I unplugged ET1 to test it (and it tested ok) and the outputs using ET2 started working, which enable the outputs which weren't before - I figured the thermistors aren't the issue.  Might still be wrong on that though, I'm wrong quite often.

[Dacian Todea]

Noel,

Is the DSSR20 connected to the SBMS120 communication port ?

If so you can look at page 5 monitoring that will show what is going on.

My guess is that panels are hot and max power point is not to high and maybe you still charge the battery also thus remaining current for heating is just about 1800W

The way it works from your description seems correct just limited power available form the PV array.

Yes TLD2,3 and 5 have priority but if enough power then also the TLD1,4 and 6 will start working.

The DMPPT450 considers max power point needs to be in the range of 28.5 to 32V so when DMPPT connects a heating element and voltage on the PV array drops below 32V but above 28.5V it will consider that it is in the max power point range and stay there. If voltage is still higher it will go to the next heating element say TLD3 and if it is still higher it will also switch the TLD2 now both are switched ON and if voltage is still higher it will continue with next heating element until voltage of the array is in that max power point range.

This was designed for Winter when panels are cold so max power point voltage will be higher than in summer when panels are hot.

 

[Noel Barlau]

Excellent info Dacian, thank you so much.

It's standalone right now, but I should probably try connecting it to the SBMS120 to check the monitoring.  I hate to say it but I still haven't even fired up any of the SBMS units you sent.  Haven't had time, somehow!  (Our 3-year old might explain it to you...)

The panels are just laid on the ground with no air flow under them, and they're hot as hell under the Atlanta sun.  Probably explains a lot!  That's what I get for doing a half-arsed test in the absolute opposite of the correct season the hardware was designed for.

As long as I'm on the subject, I'm planning on using the two SBMS0's for monitoring a 48v Lifepo4 bank.  Have you tried this, and do you think it'd work to use one SBMS0 per 24v partition both for monitoring and also for charging?

[Dacian Todea]

It is OK stand alone. I have not tried the dual SBMS0 setup as I do not have to similar batteries to test but there is no reason not to work is just that is not an elegant solution as you see just half of the pack on both and so monitoring will show you the power and energy for half the pack. You need to manually add the values to get the real power and energy used.