Jump to content

Sanity check please? (Behaviour of the MPPT controller in the Axpert MKS 3KVA-48V)


moshulu

Recommended Posts

It's late here, I'm tired and it's been a rough week... now I've just had a thought that perhaps I have misunderstood the spec sheet/behaviour of the MPPT in the Axpert MKS 3KVA-48V and was wondering if someone can check I'm not missing something before I hit the go button on the rest of my system please?

The PV Array MPPT Voltage Range spec of this particular Axpert is 60-88V and the maximum PV Array Voc is 102V.  Can anyone just reassure me that my initial understanding is correct (I hope!) and that this means that in a given situation (eg very sunny, hot day or conversely on a freezing, cloudy day) if the array Vmp occurs below 60V (or above 88V) that the Axpert MPPT controller will simply hold the array at 60V (or 88V) and take whatever sub-optimal current the array can supply at that voltage?

I'm hoping I'm just having a moment and my sudden doubt is unfounded?  Please do correct me if I am wrong here...

Many thanks in advance guys...
 

 

Link to comment
Share on other sites

4 hours ago, moshulu said:

... if the array Vmp occurs below 60V (or above 88V) that the Axpert MPPT controller will simply hold the array at 60V (or 88V) and take whatever sub-optimal current the array can supply at that voltage?

I've long wondered about this. I think it's wise not to design a system that puts typical panel voltages outside the MPPT range, but I don't see a safety issue as long as the upper limit (often called the Voc limit) is not exceeded.

I don't believe that it will clamp the panel voltage at 88 V, if the MPPT voltage happened to be say 91 V. It will likely still provide sub-optimal charge current, but I think there is a small chance it might spit the dummy and provide no current at all.

So I'd be interested to hear others' thoughts and experience on this.

Link to comment
Share on other sites

6 hours ago, moshulu said:

if the array Vmp occurs below 60V (or above 88V) that the Axpert MPPT controller will simply hold the array at 60V (or 88V) and take whatever sub-optimal current the array can supply at that voltage?

The MPPT will not clamp the voltage either on the lower as well as the upper limits. It will only take what is produced and try to use it optimally.

However, From experience on my Invini V's as well as Infini 4kw super (Although not Axpert, its also Voltronic), the MPPT only starts providing Current when PV voltage exceed the lower limit. On the Invini V it was more visible/audable than with the Super, but early morning and late afternoon i could here click as the MPPT switched on and of, so early morning, the volts will Exceed lower limit, MPPT starts up, and i could here a click, the volts drop a bit after connection, falls below the limit and MPPT stops, another click. Volts exceed , MPPT on and this pattern repeats it self till the Input is strong enough to remain above the lower limit. Same in the late afternoon in low light conditions i could here the on off clicks for a period until the volts remained below the lower limit. The same was experienced on the Super early mornings, although no clicks inside (Or I cant remember hearing it as with the Infini V's), I watched the MPPT connect and disconnect as the voltage went above the lower limit and fall below it again. 

On the upper limit, I have never experienced any disconnecting of the MPPT  on either of the 2 Inverter. I have occasionally Experienced voltage above the upper limit of my MPPT rating (But still below max VOC) on the Infini V. The MPPT remains connected, and continues supplying current to the system. I suspect that the MPPT will only disconnects from the system (switch off) at voltages above max VOC.  

 

Link to comment
Share on other sites

17 minutes ago, Jaco de Jongh said:

I suspect that the MPPT will only disconnects from the system (switch off) at voltages above max VOC. 

Actually, on the Axperts there is no mechanism to disconnect from the PV input. The clicking is from two paralleled relays (at least with the higher power versions) that connect or not to the battery. I guess they are concerned about leakage possibly running down the battery at night.

Thanks for your recollections; in particular, I'm pleased to see that when the panel voltage is above the top end of the MPPT range, that charge current still flows.

Link to comment
Share on other sites

Oh dear - thanks guys... it appears I was mistaken in my understanding.  Ok so if what you guys are saying is right then this inverter has an MPPT controller that is only useful for 72 cell modules in strings of two, however this slightly contradicts information in an earlier version of the product brochure/manual I've found:

1) Max PV input power it can accept is 900W

2) MPPT Voltage Range is 60-88V

3) Max PV Array Voc is 102V

4) This old brochure/manual (see screen shot attached) recommended an array of six 250W panels which can only have meant 3 strings of 2 modules in series (any other configuration would put such an array permanently outside of either Max Voc or the MPPT Voltage Range spec.419076530_RecommendedPVarray.jpg.bf65db2d9ef8d4b677be72d616fef74c.jpg

5) I'm looking at a 2x2 array rather than a 3x2 array as in the screen grab, but Vmp and Voc are obviously the same between the two configurations.

6) BUT a 60 cell module has a Vmp of 30-31V, which puts a string of two of these at the absolute bottom end of the MPPT Voltage Range spec for the controller.  At NOCT we're already way below (say 28-29V region per module).  In fact here are the ambient temperature points above which the Vmp drops below 30V for one particular fairly typical 60 cell module (Phonosolar PS255P-20/U):

Insolation (mW/m^2)......... Air Temp (deg C)

20..........20

30..........19

40..........16

50..........13

60..........10

70..........7

80..........4

(sorry about the formatting - can't be bothered embedding the actual spreadsheet, but let me know if this doesn't make sense and I'll make the effort/explain my calculations)

So all the above together and the feedback you guys have given means that on a clear sunny day (80mW/m^2), unless the air temp is below FOUR DEGREES CELSIUS the MPPT controller won't accept any current and will spit the dummy???

This seems like an MPPT controller with a ridiculously narrow band of usefulness if this is all true?

- It's not feasible to use two 60 cell modules in series (see above)

- It's not feasible to use three 60 cell modules in series (Voc is well over 102V)

- It's not feasible to use 60 cell modules in parallel (Vmp way below spec)

- It's not feasible to use 72 cell modules in parallel (Vmp way below spec)

- It ONLY appears feasible to use strings of two 72 cell modules in series (Vmp 75-80V at STC and Voc of 92-97V)

Am I missing anything here?  Your guidance is much appreciated...

Link to comment
Share on other sites

@Jaco de Jongh "However, From experience on my Invini V's as well as Infini 4kw super (Although not Axpert, its also Voltronic), the MPPT only starts providing Current when PV voltage exceed the lower limit."

Which PV voltage do you mean here?  Surely if the array is isolated due to being outside the MPPT voltage spec then it is open circuit, thus Voc is what you're talking about the controller sensing here?  But as soon as any current starts flowing V will drop as current increases until either the floor of the MPPT voltage spec is reached or the Vmp is reached (if within MPPT controller spec)?

If Vmp is below the minimum for the MPPT controller, why doesn't the controller just switch the array in/out of circuit at a rate that keeps the voltage from dropping below spec (and the current from rising to Imp where Vmp is below spec)?

I might go re-read all the primers about PV cell behaviour again, perhaps I'm misunderstanding how they behave?  I thought since they attempt to be constant-current sources the voltage of the cells/array will do whatever is necessary (within their specs) to attempt to maintain a constant current flow, rather than there being an arbitrary voltage in a given set of conditions that a cell/array will want to sit at?

Link to comment
Share on other sites

 

42 minutes ago, moshulu said:

1) Max PV input power it can accept is 900W

From what I see above its 1500Watt.

44 minutes ago, moshulu said:

- It ONLY appears feasible to use strings of two 72 cell modules in series (Vmp 75-80V at STC and Voc of 92-97V)

This looks like the safest option right now. 

Link to comment
Share on other sites

1 minute ago, Jaco de Jongh said:

 

From what I see above its 1500Watt.

This looks like the safest option right now. 

Thanks Jaco - it does look like it... I'd really like to get to the bottom of how these MPPT buck converters behave though too.  There's clearly a hole in my knowledge here...

And as for the max PV input power the controller can accept, see the second line of the table in the screen grab: "20 Amp, 900W".  They removed this recommendation for over-sizing/speccing the array in later versions of the brochure/manual.... not sure whether this level of oversize was causing problems.  I don't have roof space for six panels anyway so pretty sure four panels for ~1300W (400W over-spec in ideal sunlight conditions) should be ok, given I'm in a climate where this sort of nice weather doesn't last for long periods... :-) I was going to get 4x 60 cell modules but clearly 72 cell modules looks much more sensible, even if it means significant redesign of the roof area and mounting system to accommodate them.

Link to comment
Share on other sites

7 minutes ago, moshulu said:

Which PV voltage do you mean here?  Surely if the array is isolated due to being outside the MPPT voltage spec then it is open circuit, thus Voc is what you're talking about the controller sensing here? 

That was my opinion as well at the time. 

8 minutes ago, moshulu said:

But as soon as any current starts flowing V will drop as current increases until either the floor of the MPPT voltage spec is reached or the Vmp is reached (if within MPPT controller spec)?

I agree. Also part of my perception. 

8 minutes ago, moshulu said:

If Vmp is below the minimum for the MPPT controller, why doesn't the controller just switch the array in/out of circuit at a rate that keeps the voltage from dropping below spec (and the current from rising to Imp where Vmp is below spec)?

There was a time delay between switching, that decreased in duration as "more sun" became available. My perception was that they did not want to switch to often and gave some time delay to let the Voltage stabilize before allowing the next switch. Why they chose to do it this way, I am not sure. 

Link to comment
Share on other sites

Just noticed the over-size recommendation for PV array size is relatively unique to this 48V 3kVA model - the others in the table either have no over-size array recommendation or a much smaller one... I wonder if it was a typo and that's why they later removed it from their documentation?

Link to comment
Share on other sites

3 minutes ago, moshulu said:

but clearly 72 cell modules looks much more sensible,

Agree. I remember these to be some of the first models available, those years panels were still 150 and 180 Watt panels with Voc = 22volts and Vmp =18 volts. Easier to fit those into the MPPT band. But with the new bigger panels it changes the options a bit. 

Link to comment
Share on other sites

1 minute ago, Jaco de Jongh said:

Currently on their site: 900Watt

 

Yep - only found the old version with the over-size array recommendation by accident one day whilst searching for the service manual... :-)  

"I remember these to be some of the first models available, those years panels were still 150 and 180 Watt panels with Voc = 22volts and Vmp =18 volts. Easier to fit those into the MPPT band. But with the new bigger panels it changes the options a bit. "

Ah this could explain the design specs then... I thought I must have been misunderstanding something because it didn't seem to fit with the majority of (currently) commonly available panels.

Are there any electrical design engineers out there who can help me better understand the characteristics of an MPPT controller behaviour when Vmp reaches the edges of the controller design spec?  I can see myself wasting days rabbit-holing on the internet attempting to learn technical details that won't actually help get my system built otherwise! :-)

Link to comment
Share on other sites

1 minute ago, moshulu said:

Are there any electrical design engineers out there who can help me better understand the characteristics of an MPPT controller behaviour when Vmp reaches the edges of the controller design spec?

@plonkster is the best guy to speak to. He knows so much about MPPT's, I am almost certain he designed them...

Link to comment
Share on other sites

3 minutes ago, moshulu said:

when Vmp reaches the edges of the controller design spec?

Well, the point you want to avoid is when Voc reaches that limit, so when Vmp (which is lower) reaches it you are already in trouble.

Let's talk about why there is a limit in any case. It comes down to the electronic components used to do the switching. Two decades ago we had bipolar transistors, and those were wonderful things except they had a +- 0.6V drop across the junction, and if you put 10A through there you had 6W of heat to deal with, just scale that up linearly. The same was true for diodes.

But then technology moved on and gave us the FET, and the MOSFET (a type of FET), devices with an internal resistance (called RDSon, resistance drain to source when on) that creates a lot less heat and power loss. Similarly for diodes, we now have Schottky diodes which has a much lower drop. The only problem: These new devices generally have much lower breakdown voltages, making ones with high breakdown voltages that also has a low resistance and good power capability costs a lot of money.

For that reason the MPPT will be limited, because there is a MOSFET (usually at least two, but could be more for interleaved designs) that can't take more.

So what happens if you put in too much volts? You blow the FETs, that is what.

However, MPPTs will sometimes have some protection built in, depending on how generous the designer felt that day. Some have a small window at the top, just before things get out of hand, and in that window it shuts things down. It might even turn on both FETs to short the panel input down to zero volts, and if that gets out of hand it will blow a fuse.

So what that means is that maybe... depending on design... you will not blow something. But then again maybe... you will.

Do you feel lucky?

Well do you!?

:-P

Link to comment
Share on other sites

4 minutes ago, plonkster said:

Let's talk about why there is a limit in any case.

Ok, that's for the upper limit; all agreed. Why is there a separate upper limit on the MPPT voltage range? For example: MPPT range 60-88 V, max PV (often called Voc) 102 V. What happens at 88 V? You start with the open circuit panel voltage, and go down. What happens if the power has already started to go down when the panel voltage is say 92 V? 92 V is below the MOSFET blow-up voltage of 102 V, so why can't it MPPT all the way up there? Is it just to give a decent distance from the must-not-exceed 102 V point?

Link to comment
Share on other sites

17 minutes ago, plonkster said:

Well, the point you want to avoid is when Voc reaches that limit, so when Vmp (which is lower) reaches it you are already in trouble.

But then technology moved on[...]

Thanks @plonkster, that all makes sense to me (my EE studies were two decades ago when IGBTs and power MOSFETs were just starting to become mainstream).

The bit that I don't get is your first sentence though.  Clearly Voc is a hard limit you really must avoid reaching unless the design engineer was clever/had a big budget that day.  But that's not what I'm curious about here.  Voc can be well within spec whilst Vmp can be outside spec for this particular MPPT controller.

I think I need to go read up (again) on PWM vs Buck Converters.  In the scenario outlined in this thread EXCEEDING the Vmp spec of the controller is not an issue, it's dropping BELOW the minimum Vmp spec for the controller that will happen routinely with the 60 cell modules.

@Coulomb wrote: "Ok, that's for the upper limit; all agreed. Why is there a separate upper limit on the MPPT voltage range? For example: MPPT range 60-88 V, max PV (often called Voc) 102 V. What happens at 88 V? You start with the open circuit panel voltage, and go down. What happens if the power has already started to go down when the panel voltage is say 92 V? 92 V is below the MOSFET blow-up voltage of 102 V, so why can't it MPPT all the way up there? Is it just to give a decent distance from the must-not-exceed 102 V point?"

This I understand I think - I suspect it's to do with the DC switching POWER rating of the FETs.  At OC there is no current therefore no power to dissipate (at 102V).  At 88V there could be ~10A to switch, ie 880W.  Even high efficiency MOSFETs have some switching losses which as a %, going above 880W switching power may exceed the TDP of the packages/heatsinks.

Edited by moshulu
clarification
Link to comment
Share on other sites

10 minutes ago, Coulomb said:

Why is there a separate upper limit on the MPPT voltage range?

Here I am a bit out of my depth. But I believe this has to do with SMPS design parameters. In classic buck converter design you have three things you can play with: The size of the inductor, the frequency at which you're going to switch, and the size of the capacitor on the output side. We can ignore the cap on the output for now, because that's usually determined by how low you want the output voltage ripple to be.

The inductor effectively serves as a temporary magnetic storage place for energy, and the higher the input voltage, the shorter the time it takes to fill the inductor with energy. Since the PWM usually only has a number of discrete power levels (depending on how many bits), it inherently means that at some voltage level it simply hasn't got enough granularity to control the amount of energy properly, and then your accuracy and throughput starts to suffer.

That's what I THINK in any case.

There's also the matter of running the DC/DC converter in continuous vs discontinuous mode. The former is more efficient. At higher voltages I suspect you move into discontinuous mode. But I may be completely off base here.

Edit: No, I think I am wrong about discontinuous mode. That happens at low power levels. Although high voltage input with low output might well be discontinuous.

Edited by plonkster
Link to comment
Share on other sites

4 minutes ago, plonkster said:

There's also the matter of running the DC/DC converter in continuous vs discontinuous mode. The former is more efficient. At higher voltages I suspect you move into discontinuous mode. But I may be completely off base here.

I need to re-study this so could easily be wrong too, but my hazy understanding dredging the depths of my memory is that continuous mode is when input and output voltages are the same/similar.  Clearly we need an EE guru here! :-)  There used to be a guy named Lee Hart on the US EV discussion list many years ago and I never saw an EE question he couldn't answer! ;-)

Link to comment
Share on other sites

15 minutes ago, moshulu said:

continuous mode is when input and output voltages are the same/similar

Continuous mode is when your switch on the input side is turned back on before the inductor has "emptied" completely, so current is continuously supplied to the capacitor on the output side. That's if I remember this correctly! :-)

There's something about this in "The Art of Electronics".

aoe.thumb.jpg.ad0a047979cf7e344e07678f9901ba94.jpg

Link to comment
Share on other sites

2 minutes ago, plonkster said:

Continuous mode is when your switch on the input side is turned back on before the inductor has "emptied" completely, so current is continuously supplied to the capacitor on the output side. That's if I remember this correctly! :-)

There's something about this in "The Art of Electronics".

aoe.thumb.jpg.ad0a047979cf7e344e07678f9901ba94.jpg

Oh god - I wish I'd never sold my copy of Horowitz and Hill in my poor student days now!!! :-)  Yes - you are most correct!  That's where I was going with the "similar voltages" bit... I was just struggling to remember how similar they had to be. That obviously depends on the inductor and I suppose doesn't really need to be that similar, if the inductor is correctly specced.

Link to comment
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

×
×
  • Create New...