Obscure Technology: The Ten K Solar module

Peterator

New member
Joined
Jun 27, 2021
Messages
22
NEW 10/14/21: I'm spoken to a growing number of people that can't run their system at full capacity because there aren't suitable inverters available. I know of two efforts to provide a solution to this problem. We should get more information in the coming months. This facebook group seems to be the best place to connect at the moment: https://www.facebook.com/groups/559492874643037

There is a weird breed of solar panel made by an defunct Minnesota company called Ten K Solar. They operate very differently from the standard solar panel. Every now and then I see a post on the internet asking about these, so I'd like to share what I know about them.

I know this is probably going to help nobody on a battery forum, but some people might get here on a google search or whatever and I wanted to get this information out somewhere.

If you're from the North-Central US and you just got your hands on a second-hand, weird-looking solar panel, then you've come to the right place. For everyone else, this write-up is entirely useless but might be an amusing anecdote.

Ten K Solar (AKA tenksolar or just TenK) was a solar company founded in 2008 in Minnesota, USA. Leveraging some “interesting” design choices, they grew out of the start-up phase thanks to some state incentives in place at the time. I joined the company in 2015, and left just before they shut down in 2017. (Many reports claim that Tenk went out of business due to the unreliability of the inverters. While there were definitely inverter troubles, the real issue was simply the difficulty in maintaining cost-competitiveness.)

Since I left before the shutdown happened, I don’t know what happened to all the company’s assets when they stopped operation. Specifically, the company owned a lot of unique IP which may be lost forever for all I know. I didn’t have the forethought to secretly take a copy of the source with me so the following is based on what I remember.

The short version of TenK’s design is “put it all in parallel”. Specifically regarding the latest 500W “Apex” modules:
Cells: half cells in a 25s8p configuration. The parallel cells are from bottom to top and series from left to right; this way if the bottom row is shaded (as they often are in order to get panels close together), it doesn’t interrupt the string.
Internal electronics: The plastic housing contains a MPPT boost converter to go from the ~15V cell voltage up to ~56V. The converter itself is an 6-unit interleaved boost converter (6 converters in parallel).
String connections: Around 10-15 or so modules are wired in parallel. The limiting factor is the amount of current that the wire can handle. #2 aluminum wire was the most common.
Inverters: The string of modules were wired to 10-12x ~700W micro-inverters wired in (you guessed it) parallel. The inverters were off-the-shelf units modified by the OEM to run in constant voltage mode (instead of MPPT).

Because of the integrated electronics, the TenK modules can’t be used in normal solar systems. In order to even get the module to produce power, it needs to be presented with a valid load. A load can be either:
A voltage source, such as a battery, or another module that has already started up. I forget the exact voltage requirements but I think anything between 20-55V would be acceptable.
A large (1000 uF+) capacitor, such as would be present on the input stage of a microinverter. There's a precharge step so the anything connected to the cap will need to wait a couple seconds before drawing power

So options for using these panels:
Find a Modified micro-inverter. This might be hard since they're not made anymore and they weren't perfect even when they were. LeadSolar and APS built them. Some of the companies that own the arrays do get inverters as replacements and I have a contact at one of them, so it may be possible to get some at full price.
Use a Standard Microinverter. You’d need something with a 60V input voltage. The interaction between the inverter MPPT and the panel electronic’s output tends to be unstable. It may be possible to improve it with some filtering between the two devices but we never experimented with this.
Use a battery. One of the less-frequently-used but fully-supported use of the panels is to directly charge a battery. The panel electronics double as a charge controller; you can simply hook up the panels directly to the battery to be charged.
One caveat: the end voltage is programmed by default to around 56V, which is too high for a 13s but too low for a 14s if using Li-ion cells. Changing this requires recompiling the source and programming with a special tool, neither of which are available.
I’m currently working on a system like this with a 13s battery; I plan to make some sort of voltage-cutoff circuit to avoid overcharging the battery.
Use a battery-input inverter. I haven’t tried this but it should work. By simply using a large capacitor to take the place of the battery, any inverter that runs off a battery should operate, provided that the inverter doesn’t demand more power than the modules produce.

That's about it. If you have questions, I'm active on this forum. I really think the TenK system is pretty cool, but the lack of compatibility with standar solar components makes them troublesome.
 
Last edited:
Thanks for that.

There was a video they made with two guys in lab coats.
Shading the panels, and covering them with dirt.
I think they even shoot them at one point.
It was most entertaining stuff, but showed what could be done.
Think the two science guys on the muppets. Its a bit like that.
 
There was a video they made with two guys in lab coats.
I know the one. Here's a youtube link:
View: https://www.youtube.com/watch?v=IiiEMM2UaRw&ab_channel=TenKSolar

The data off the meters is real I'm pretty sure. Because of the parallel cell connections it's basically impossible to break a string.

Of course the moment you break the glass the module is going to corrode quickly (assuming it's outside) and become useless after a year or so, but that's less exciting.
 
I know the one. Here's a youtube link:
View: https://www.youtube.com/watch?v=IiiEMM2UaRw&ab_channel=TenKSolar

The data off the meters is real I'm pretty sure. Because of the parallel cell connections it's basically impossible to break a string.

Of course the moment you break the glass the module is going to corrode quickly (assuming it's outside) and become useless after a year or so, but that's less exciting.
Looks like they did the 'Random Act of Violence" (1:10) test right after the rifle/bullet test (meters started at 100%) but yet you could still see the bullet holes... so it couldn't be 100%. Looks suspicious.
1627224143199.png
 
Thanks for reaching out on the other forum. I bought 2 new 500 watt panels from them when they shut down. I knew nothing about solar other than the 3 harbor freight 100 Watt systems. For lack of better words I've been begging for help. The inverter manufacturer wouldn't help me out. I do have them hooked up in parallel with 2 rec 310 Watt panels. Hooked in series for 76 v DC. Any help would be appreciated! Thanks Bob
 
Happy to help, Bob. It sounded like you're using these to charge a battery, is that right? If you can tell me about the battery and the charge controller (model number if you have one) we can figure out if there's a way to get them to work.

76V is higher than the TenK panels are rated to so I would not recommend that configuration.
 
1st Thank you for reaching out! I have these at my off grid cabin and this last sunday i was going Scrap them. This is only my second year of playing with solar but have been off grid for 15 years. Please Don't laugh at my hodge podge system.
I'm currently using an xantrex 3624 modified sinewave inverter. 12 Six volt napa 230 amp hour batteries. (3 24 volt banks), I started out with 400 watts of harbor freight panels. using a epever 3010 tracer mppt. Then I saw an add for the ten-k 500 watt panels so i picked up 2 new off the crate panels with the assurance that i could hook them up to a charge controller and all would be good. as you know they didn't work. the next day i hooked my harbor freight system wired to 56 volts and with full sun i believe i was getting some output from the ten k panels? but if so very little. fast forward to last weekend. i picked up 2 rec solar 310 watt 43 volt panels and wired them in series (My highest measured was 76v) seperately to a new Epever 40 amp tracer. my plan was to try and somehow trick them into working. I drove the 3 hours home and saw your message. so now I'm sure hoping you can help me salvage those panels. Again Thanks for reaching out!!!
Thanks, Bob
 
I think people here can appreciate getting what power you can using whatever you can get your hands on!

Sounds like you have some other panels set up so I'm going to recommend using a separate charge controller for the TenK modules. Luckily, you don't need MPPT functions in the new controller since the modules have all that built in so you can get something very cheap. I recently ordered this part from eBay for my own system and I think it should work for you as well but I haven't tested it yet. You may also need some capacitors to get it to start up, I found this one with a quick search.

The problem for your system, which I just realized, is that the output current on TenK panels is limited. So with a 24V battery, the power will max out at around 250W.

Admittedly, putting the standard panels in parallel with the TenKs seems like a workable solution, provided the voltage is right, but the MPPT is not going to optimize power from them well. Running at 56V and getting only a little output from the TenKs is an expected response.

If you're in an experimenting mood, try hooking up the TenKs direct to your battery, bypassing the charge controller. You won't damage anything as long as you don't leave it connected. It will start charging your battery but will likely cap at 250W or so. The only reason you can't leave it this way, as I've mentioned before, is that it will overcharge the battery.

I'm not sure what the best way to get the full 500W into a 24V battery, but it would involve some added electronics or tricking the charge controller. I'll have to think aboout how that might be done.
 
I think people here can appreciate getting what power you can using whatever you can get your hands on!

Sounds like you have some other panels set up so I'm going to recommend using a separate charge controller for the TenK modules. Luckily, you don't need MPPT functions in the new controller since the modules have all that built in so you can get something very cheap. I recently ordered this part from eBay for my own system and I think it should work for you as well but I haven't tested it yet. You may also need some capacitors to get it to start up, I found this one with a quick search.

The problem for your system, which I just realized, is that the output current on TenK panels is limited. So with a 24V battery, the power will max out at around 250W.

Admittedly, putting the standard panels in parallel with the TenKs seems like a workable solution, provided the voltage is right, but the MPPT is not going to optimize power from them well. Running at 56V and getting only a little output from the TenKs is an expected response.

If you're in an experimenting mood, try hooking up the TenKs direct to your battery, bypassing the charge controller. You won't damage anything as long as you don't leave it connected. It will start charging your battery but will likely cap at 250W or so. The only reason you can't leave it this way, as I've mentioned before, is that it will overcharge the battery.

I'm not sure what the best way to get the full 500W into a 24V battery, but it would involve some added electronics or tricking the charge controller. I'll have to think aboout how that might be done.
Thanks, I'm excited to give your ideas a try! These panels were never in service so it would be a shame to dumpster them. I'll keep you updated and i'm sure l'll have more questions. Thanks, Bob
 
There is a weird breed of solar panel made by an defunct Minnesota company called Ten K Solar. They operate very differently from the standard solar panel. Every now and then I see a post on the internet asking about these, so I'd like to share what I know about them...
Peterator,

Thank you for this write up! Would have loved to have seen this 3 years ago when I started to go down this rabbit hole. FWIW, I have eight of the Apex 500 modules attached to eight YC500-K micro inverters connected conventionally (Each panel connected directly to the micro-inverter) and have been running them for over a year now without issues. I have many more of the modules in storage. I have voltage displays on 3 of the panels so that I can see what is going on. I tried to get these panels to work with the YC600 inverters but failed. I too found that I could "trick" the panels into turning on when I introduced 300uF across the outputs of the panel. It seems that at about 14VDC the panels would turn on. The problem I found was that the YC600 inverter would pull down the voltage until the panel shuts off again. I did not try increasing the capacitance to 1000uF or more, that may have made the difference with these inverters. When running the YC500-K inverters (That were specifically made for these Ten-K panels) I note that when power is first applied and before the inverter allows power to flow back to the grid, the panels are static about 59VDC. When the inverter allows power to flow again, the panel voltage is fairly steady at about 51VDC.

As to your assertion that the panel could directly charge batteries...
While I have not tested this myself, it seems that if the old lead-acid batteries were used, four 12V batteries connected in series for a total of 48VDC might be perfect for these panels.

One question I do have is: Do you know anything regarding the rumor that the later Apex 500 panels were internally limited to 350-375 watts? Early experiments indicated that is about what I was getting out of these.

Thank you again for your time in sharing with us your expertise!
 
Welcome to the forum Sealtee. I hope the information helps.

The problem I found was that the YC600 inverter would pull down the voltage until the panel shuts off again.
That's not unusual for conventional inverters. Do you think it was an issue of the TenK panels not having time to get their output voltage up, or just the inverter MPPT failing to optimize the output power and driving the voltage down? I've been racking my brain for a way to get TenK modules to play nice with conventional inverters but it's a tough problem. I don't think more capacitance would help - the fact that you were able to get them to start with 300uF means I'm probably mis-remembering the 1000uF requirement.

While I have not tested this myself, it seems that if the old lead-acid batteries were used, four 12V batteries connected in series for a total of 48VDC might be perfect for these panels.
Yes, I think the voltage limits line up pretty well for lead-acid, and TenK has sold a system or two with that exact configuration.

Do you know anything regarding the rumor that the later Apex 500 panels were internally limited to 350-375 watts? Early experiments indicated that is about what I was getting out of these.
I don't know of any modules being limited that low, although it's not unreasonable. The standard limit is around 10A for a 500W module. It may have been lowered on some systems to better fit the capacity of the attached inverters, wiring, and number of modules.
Then again, 500W is peak power in ideal conditions, and doesn't include inverter loss. If the power output is actually being internally limited, you'd see a flat peak on your daily solar power graph.

If your modules have been limited, they can be reprogrammed to the full 500W. You'd need access to one of their custom programming tools, which might be a challenge to track down.
 
Welcome to the forum Sealtee. I hope the information helps.


That's not unusual for conventional inverters. Do you think it was an issue of the TenK panels not having time to get their output voltage up, or just the inverter MPPT failing to optimize the output power and driving the voltage down? I've been racking my brain for a way to get TenK modules to play nice with conventional inverters but it's a tough problem. I don't think more capacitance would help - the fact that you were able to get them to start with 300uF means I'm probably mis-remembering the 1000uF requirement.
Thanks Peterator.

To your question...

What I did was I set up a panel on the ground with a voltmeter display connected to the panel. Obviously it tells me when the panel is on and what voltage is present. What I did initially was to connect 100uF to the panel when nothing else was connected. The result was the panel was unstable and did not want to stay on. 200uF was much better. 300uF looked good, the panel took a couple of seconds to "charge" the caps to about 14VDC then the output went to nearly 60VDC and was steady. I connected the YC600 along with the 300uF cap and powered it on. I watched the panel voltage get dragged down to the point where the panel turned off again. I even went so far as to rig up a small transformer (16VDC 600mA) and connected that to the panel. The panel would turn on but when the inverter was connected it once again pulled the voltage down turning off the panel and significantly over driving the small transformer!

What I THINK is happening, (Not necessary what IS HAPPENING but only what I think) is that when the YC 600 inverter first comes on it "looks and tests" to see what the short circuit voltage is. I say this because that is what it looks like it is doing, shorting the panel.

I will be doing more testing with the one of the YC600 inverters that I have left (12 others are connected to 24 330W conventional solar panels and running without issue) to see if I can somehow get these two to play well together. I will be trying 1000 to 2000+uF to see if this is enough reservoir of current to satisfy the inverter during start up. I have reason to figure this out as I have 24 of these panels that are worthless for grid tie unless a suitable micro inverter can be found.
 
Last edited:
What I THINK is happening, (Not necessary what IS HAPPENING but only what I think) is that when the YC 600 inverter first comes on it "looks and tests" to see what the short circuit voltage is. I say this because that is what it looks like it is doing, shorting the panel.
That's plausible, and behavior like this is why it's a difficult problem to get TenK panels to work with conventional inverters. I'm not confident that more capacitance will work but I guess it's worth a try. There are more fundamental problems: TenK panels don't provide a power/voltage peak for the MPPT to latch on to, and the power electronics don't respond to the MPPT P&O algorithm the way the inverter expects.

Regarding your unusable panels, I do have a contact at the local installer company that owns a bunch of active TenK arrays, and they mentioned getting APS to do a special run of constant-voltage inverters to replace some that had failed. I could investigate and see if they would get extra to sell.

Otherwise, in order to get conventional inverters to work, I think there'd need to be a major change in the TenK firmware, change in the inverter firmware, or a specially designed adapter circuit to make the TenK panel "look" like a conventional panel. Each of these is a pretty major project that I could potentially do given a lot of time to work on it.
 
That's plausible, and behavior like this is why it's a difficult problem to get TenK panels to work with conventional inverters. I'm not confident that more capacitance will work but I guess it's worth a try. There are more fundamental problems: TenK panels don't provide a power/voltage peak for the MPPT to latch on to, and the power electronics don't respond to the MPPT P&O algorithm the way the inverter expects.
I hear what you are saying and I tend to agree with you. That said, now that the seed is planted, I will not know the answer for certain until I try!

As to some of the other panels I have...

I am actually in the process of building a small shop. On the shop I plan on using 10 of these Apex panels. I have 6 - 1400W Lead Solar inverters (Works with 2 of these Ten K panels) that while they turn the panels on (Even when not energized) they do not transfer power to the grid. It seems they are pre-configured for the 277VAC range and the voltage at my home is just out of that range. Tomorrow I will be experimenting with a buck-boost transformer to see if they can be made to work. Other wise I only have 5 more of the YC500-K inverters if I cannot get the YC600 inverters to work.

I would be interested in purchasing inverters if your local installer were to be able to convince AP Systems to do a limited run on these.
 
Checked with my contact - the developer is Ideal Energies, though they're probably not too interested in being an inverter reseller. He suggested contacting APS directly, since it seems like they are capable and willing to build TenK-compatible inverters. My understanding is that the hardware is the same, they just load in different firmware, so theoretically it's not a lot of extra work for them.

You can wire 3 TenK modules to a 1400W inverter - in the rare (perhaps impossible) case the 3 modules make more than 1400W total the inverter will limit itself. Or do it the TenK way and wire all 10 modules and 3 inverters in one bus. 277V was the most common inverter voltage - meant for commercial connections. A transformer should make that work for a 240V grid assuming the power rating of the transformer is sufficient.
 
Peterator: I think it is great that you took the time to post, because there is always someone who can benefit from it. Even if there wasn't, it is interesting history. I'm interested in the concept of just putting all power direct to batteries.

Carry on!
 
Checked with my contact - the developer is Ideal Energies, though they're probably not too interested in being an inverter reseller. He suggested contacting APS directly, since it seems like they are capable and willing to build TenK-compatible inverters. My understanding is that the hardware is the same, they just load in different firmware, so theoretically it's not a lot of extra work for them.

You can wire 3 TenK modules to a 1400W inverter - in the rare (perhaps impossible) case the 3 modules make more than 1400W total the inverter will limit itself. Or do it the TenK way and wire all 10 modules and 3 inverters in one bus. 277V was the most common inverter voltage - meant for commercial connections. A transformer should make that work for a 240V grid assuming the power rating of the transformer is sufficient.
Peter, I talked with APS. They are definitely not interested in selling their microinverters other than with their current non-compatible design. They claim that TenK bought their APS inverters and modified them. I got the impression that more than firmware was involved. APS referred me to another company to find someone willing to supply inverters but that company has no interest either.

I also talked with Ideal Energies. When TenK went out of business, Ideal bought a lot of remaining spare parts, but it's unlikely that they still have any.

I talked to a number of other solar and non-solar inverter suppliers.

Not really any good solutions, I think.
 
Thanks for making these inquiries, Dave. I guess there just aren't any inverters out there that will work.
I'm not quite ready to let everybody's panels go to a scrap heap yet. I have a few ideas, they're all going to take some effort for me implement but maybe I can get that back in selling what I come up with.
  • Reverse-engineer the TenK control circuit, and develop my own firmware that emulates a solar panel. I made some progress reverse-engineering the programming tool a couple years ago, but without access to the source code, new firmware has to be done from scratch. There's a chance this won't work; if the output doesn't act fast enough due to hardware limitations the inverter still won't be able to optimize, but if it does work we won't need any extra hardware and you can use any inverter.
  • Build an adapter circuit, something that bolts on to the TenK modules and makes their output work with microinverters. This wouldn't be particularly difficult and allows people to use their existing hardware, but it adds the cost of a third piece of hardware that would need to be added to each module.
  • Design a new microinverter that works with TenK modules. Probably the hardest option; I don't doubt I could get something to work but getting the safety and reliabilty up to standards may be beyond my resources. The benefit is that this could be a drop-in replacement for failed TenK-compatible inverters which would be easy for existing array owners.
  • Adapt an existing inverter. In particular, something with programmable/output limiting like a GTIL could probably be made to work, perhaps with a small added control circuit. Inverter choice would be limited but at least there would be one option, and I think this would be easy relative to the other options.
At this point I'm not sure which path to go down. Let me know if you have any thoughts.
 
Back
Top