diy solar

diy solar

Quality vs China MPPT?

I would have preferred if Midnite Solar made a Classic that would handle higher voltage so I could save on wiring.
But their philosophy is higher voltages create higher losses because of heat generated.
I think you are mistaken. Higher voltages = Higher losses? That simply does not make sense. Considering total power being transmitted, higher voltage will increase transmission efficiency due to lower current. Did you mistype this sentence?
 
I think you are mistaken. Higher voltages = Higher losses? That simply does not make sense. Considering total power being transmitted, higher voltage will increase transmission efficiency due to lower current. Did you mistype this sentence?
There are losses in transmission like you are talking about and there are losses within the unit itself. Midnite have a graph somewhere showing the efficiency of their controller with panel voltage V battery charge voltage.
 
Can you guys both post your evidence here in the forum. Either a link or a picture of this curve. Thanks.
Ill have to have a look when i have some time. It definitely makes it easier having the ability to run higher voltages and cheaper cable wise.
 
I bought an Epever 4215BN in 2016. 4 275W panels on it, never had a glitch with it since day one. The day I install more panels, I most certainly buy another Epever.

But….. I never tried anything else so I can’t compare.
 
Can you guys both post your evidence here in the forum. Either a link or a picture of this curve. Thanks.
@Will Prowse ,

read this post by myself and the graph directly above this post by weldman of Midnight power curve


I did extensive research using two identical pre-production prototype Kid controller using two identical arrays of solar panels using 2S3P configuration and 3S2P configuration

Repeated the same tests using two identical Midnight Classic 150’s using a variety of panel types each test using identical panels in a 12P configuration vs. a 2S 6P config. using 96 cell 53.5 volt panels as well as 2S6P vs. 3S4P configuration

The only difference was the “voltage transform ratio”

The original tests were posted at forum.midnightsolar

I forget, maybe under KID BETA TEST

Anyway I documented this behavior using the EPSolar AN 150 volt 30 amp units as well is a side by side tests vs. the beta Kid controllers.

I am a pre-production prototype beta tester for midnight solar for the Kid (done)

All current results will be posted at forum.midnightsolar.com


search the term ......”voltage transform ratio” by poster “tecnodave”
 
Last edited:
I never tried anything else so I can’t compare.
And empirically you probably never will be able to compare. Not knocking victron or anything else but the “difference” might not be nearly as significant as perhaps just adding another panel. On the other hand, some victron products aren’t that much more money measured in pizzas versus the percentage difference in cost.
 
The biggest difference between the Victron “blue plastic box” and the EP Solar is the life per dollars spent ratio,
The Victron is a much more elegant design but the sealed blue plastic box is nowhere as good a heat sink verses the finned extruded aluminum case of the EPSolar and maybe the EPEver as well. I had a few years using the EPSolar, but they could not be set to the approiate voltage sets required by my battery set, only 4 generic profiles, none of which could adequately charge true solar batteries

If kept cool the Victron will outlive the EPEver by a mile.....
 
only 4 generic profiles, none of which could adequately charge true solar batteries
Epever has a ‘user’ profile now where one can enter their own parameters.
Generic defaults for flooded, agm, and “street light” options, plus a user defined profile.
 
@Will Prowse ,

read this post by myself and the graph directly above this post by weldman of Midnight power curve


I did extensive research using two identical pre-production prototype Kid controller using two identical arrays of solar panels using 2S3P configuration and 3S2P configuration

Repeated the same tests using two identical Midnight Classic 150’s using a variety of panel types each test using identical panels in a 12P configuration vs. a 2S 6P config. using 96 cell 53.5 volt panels as well as 2S6P vs. 3S4P configuration

The only difference was the “voltage transform ratio”

The original tests were posted at forum.midnightsolar

I forget, maybe under KID BETA TEST

Anyway I documented this behavior using the EPSolar AN 150 volt 30 amp units as well is a side by side tests vs. the beta Kid controllers.

I am a pre-production prototype beta tester for midnight solar for the Kid (done)

All current results will be posted at forum.midnightsolar.com


search the term ......”voltage transform ratio” by poster “tecnodave”
Ohhh yes, converter efficiency. Ok, super basic stuff. I was thrown off by the term "Voltage Transform Ratio". The chart you posted is a very typical efficiency curve dependent on converter input/output Voltage differential. This is why we match array voltage with battery voltage whenever possible. Charging a 48V DC battery with 600VDC array is going to have higher losses than charging a 48V DC battery with a 100VDC array.

You are making valid points. But for most people, the difference in output is not a big deal, really. Pretty small unless you are building a solar farm that is specifically charging a battery bank with DC. I would imagine that most utility size battery systems would be AC coupled.

As shown clearly in Midnight’s graph above My controllers can do 2700 watts at 24 volt battery but if i wired 3S my maximum power would be 2553 watts with the same panels and Controllers.
Ok yes, small gain in output. Very true, and your system components will last longer.

I was thrown off by your terms, but these are all valid points. Agreed.
 
Last edited:
I would love to try this out in an experiment. Charge a 48V battery with a high voltage array to see how bad this loss is. That would be really cool! I did mention these losses in some of my older videos, but no one seemed to care too much because they were so small.

I do like the point about hardware durability that you mention. I did not think of that. I would have to test these and see the difference in heat that you refer to in your posts. Do you have any literature backing that up? I have noticed that most 48V MPPT can easily handle 250-450V input, and seem to handle it fine for years. And it is not a switching voltage regulation method, so matching it is not as crucial. Yes there will be a small loss. But I do not think of it as that big of a deal. Especially considering the cost of copper and how cheap solar panels are. Wouldn't you be better off just adding another panel? I know of many MPPT that run at high voltage differentials from input to output, and run just fine for ages. What components would fail first from the heat? They are pretty simple circuits overall. I suppose the caps do not like the heat, but if you buy the good controllers, they should be outfitted with ceramic capacitors.
 
Last edited:
The loss in effeciency was not the problem, that loss in effeceincy was converted into heet in the controller was the problem, It caused the controller running at the higher voltage to run 50 degrees hotter. I had two identical arrays one right at the power shed, panels to controller 12 feet wired 2S6P at 70 v mpp , the other array mounted remotely, 325 feet away wired 3S4P at 105 v mpp. The first thing that I noticed , the Classic with the longer run at higher voltage , the fans were running all the time, but the controller locally the fans never came on. Looking closer both arrays were producing basically the same net power. The longer wiring at the higher voltage did have less voltage loss (power loss) but the controller had more loss due to the higher conversion loss (power lost to heat in the controller) .

Running at the higher voltage basically resulted in a very small amount of extra power but at what cost.....cooking the power electronics in the controller, 50 degrees F. is huge to power electronics, not just the caps, but the FET banks and inductors as well. Infrared heat maps of both controllers side by side told a big story. It was dramatic.

This was a very closely matched test, both controllers were identical MidNight Classic 150 controllers, all panels were Sharp monocrystalline 200 watt 72 cell panels.

The test was repeated with Midnight pre production Beta Kid controllers with smaller arrays , same results, then again with very early EP Solar AN controllers, 150 volt, 30 amp units, forget exact model but two identical controllers, same results. I have other controllers but these three types were the only controllers that I had two identical controllers that I could put side by side for these tests.

I do have others as well that were tested head on head but not totally identical electronics. The others include BZ 250, BZ 500, Morningstar MPPT 60 amp, 150 volt , Morningstar PWM 60 amp, vs. Trace C-40 etc.

Variation in battery state was eliminated by using Hewlett-packard power source/sink as the load. It will maintain the same exact set voltage wether sourcing power or sinking power. Data logging was done with a 24 channel data logger that monitored all parameters. voltage and current at arrays, at controller input and controller output to power sink, temperature, etc.
 
Last edited:
I think you are mistaken. Higher voltages = Higher losses? That simply does not make sense. Considering total power being transmitted, higher voltage will increase transmission efficiency due to lower current. Did you mistype this sentence?
No, I didn't think it made sense either. I read it on their website. I'll try to find it !
 
The biggest difference between the Victron “blue plastic box” and the EP Solar is the life per dollars spent ratio,
The Victron is a much more elegant design but the sealed blue plastic box is nowhere as good a heat sink verses the finned extruded aluminum case of the EPSolar and maybe the EPEver as well. I had a few years using the EPSolar, but they could not be set to the approiate voltage sets required by my battery set, only 4 generic profiles, none of which could adequately charge true solar batteries

If kept cool the Victron will outlive the EPEver by a mile.....

The entire backside of my Victron 100/50 is one big heat sink. When the controller is cranking out the amps to the battery bank the heat sink is producing a bit of heat. My dinky inverter is nearby and it puts out a heck of a lot more heat than the solar charge controller.
 
Well......boB and Robin Gudgel agree with me on this issue, I think that they have far more experience than anyone on this forum.......founders of Midnight Solar.....Outback Solar......Trace Engineering

You are welcome to do the study that I did
 
Well......boB and Robin Gudgel agree with me on this issue, I think that they have far more experience than anyone on this forum.......founders of Midnight Solar.....Outback Solar......Trace Engineering

You are welcome to do the study that I did

I was just pointing out that the Victron MPPT does have an external heat sink. Your post may have left readers thinking that it did not have a heat sink.
 
I was just pointing out that the Victron MPPT does have an external heat sink. Your post may have left readers thinking that it did not have a heat sink.
I was not replying to your reply, I have never had the 50 amp version, the only one that I saw in operation was a 30 amp, 150 volt version that could not be programmed to properly charge a set of Rolls Surette S-550 L-16’s directly due to the fact that no bluetooth device or Victron programmer was available to program the device. The user had a very high end android device and it as well as several other googliod devices could not link to the Victron. As a service provider It is not my responsibility to provide a google device to program your difficult to program device.
That user trashed a brand new set of Rolls-Surette batteries saving less than $100 dollars buying a controller that was not supported locally nor able to be programmed without some external device, nor was it capable of providing the sustained 40 amp charge current needed to properly charge the Rolls Surette batteries

Not being able to program a device is a deal killer for me, That is a must. I have heard the same issue from other users of the “blue box Victron devices” that are made in China and India by Guangsfaun. The bluetooth link sucks, Victron VE programmers do not make any sense for a 1 device system, they are far too expensive for that.

All my controllers are programmable locally with no external programming device needed.
All of my controllers are made here in America and fully supported by the manufacture.

You should know what you want and need.
 
Will,

I do not think that you were disagreeing once you got my point about the buck converter throwing excess power off as heat. The other poster disagreed about what I said about the low end Victron SCC’s, about a sealed box being able to throw off heat as well as a finned aluminum case., or a rough cast aluminum case as is found in some better brands.
 
Back
Top