Repairing my MPPsolar 2424MSE(No PV input)

Chris

New member
Joined
Oct 9, 2016
Messages
13
First off I like my MPPsolar inverter even though it died due to something that doesnt happen a lot in the UK (TO MUCH SUN)

2 years of constant service without any issues for the price of it I really can't complain too much but when it did have a problem I couldn't find any information on how to fix the problem. I trained on heavy duty 3 phase so when it comes to DC and single phase what I know is what i've learned on the interweb as well as ouch that hurt I won't do that again :)

So the other day I go to check on the system and I realise my batteries are rock bottom and there's a definite smell of magic white smoke and on further investigation the capacitor on the pv input side looked a little poorley

image_whggmz.jpg




so I took the inverter off the wall and dismantled it here is how I fixed it (This worked for me but I do not guarantee it will work for you)
Also on further inspection I found that I had a bad mosfet on the pv input.

1-Remove screws holding on cover (even the one under warranty seal) and pull front up from bottom

image_hqvvqx.jpg

unplug the 2 wires for the front panel.

2- Remove the switch as the pcb lifts out and is in the way

image_vrklsa.jpg


3- Remove the plastic air shroud usually 2 cable ties and white plug

image_eqolim.jpg


4- Remove the 8 fan screws

image_svdfbx.jpg


5- Remove the 2 nuts on the pcb that has the DC choke

image_xtmlxy.jpg


6- Remove the 2 screws from the other side

image_bclbak.jpg


7- Unscrew the blocks for the AC in/out (the screws stay in the block)

image_pdrrgi.jpg


8- Unplug the fans

image_zxsxad.jpg


9- Remove the Earth nut

image_pijfov.jpg


10- Remove the nut from the breaker

image_ubtnem.jpg


11- Remove the PCB screws (2 were missing from mine)

image_qvqvof.jpg


12- On the side near the PV input there is a bracket to hold in a daughter board

image_dfvaho.jpg


13- If you havent already take off the DC choke and the daughter boardand put them somewhere safeas its really easy to damage these parts

image_yauabg.jpg


14- Even though there was only one bad mosfet I decided to replace all 3 on the PV charge side as well as getting anichicon branded replacement capacitor

image_vmorxo.jpg

also I noticed that the screws holding the mosfets were very loose when I removed them, The mosfets are supposed to be torqued down just get the datasheet to find out what this should be I used the rule of turn slowly until I decided yep thats enough.

15- after id soldered the mosfets and capacitorthis was the result

image_qpegle.jpg

This was taken a 7pm on shaded panels the fact that it showed anything was great and the next day it was firing on all cylinders again.

The main reason for this post is that I couldnt find any info anywhere so I hope this helps anyone else that has the same issue.
 
Awesome write up! Great pictures. Thanks for the walkthrough and steps :)

It's great when an expensive piece of equipment can be fixed with a few dollars worth of parts and a couple hours of time.
 
Korishan said:
Awesome write up! Great pictures. Thanks for the walkthrough and steps :)

It's great when an expensive piece of equipment can be fixed with a few dollars worth of parts and a couple hours of time.

Thanx this was my first post except for the introduction post I will be doing one on my budget ebike soon.
 
Dam that'll help someone else for sure - good to see you got so long out of the inverter before it failed - do you know how it happened? Too much solar input, cloud edge pushing amps too high, or just age?
 
hbpowerwall said:
Dam that'll help someone else for sure - good to see you got so long out of the inverter before it failed - do you know how it happened? Too much solar input, cloud edge pushing amps too high, or just age?
It's a rare occurrence so rare in fact I dont remember it happening in my lifetime but we had too much sun.
The panel's I have on the pip is right on the border line 1000w and the pip max is 1000w but just before it died it recorded 1200w something I thought impossible maybe it was a glitch as the MOSFET crapped itself ive taken the panel's down to 800w and I'm looking for another seperate charge controller that's within my budget that at this moment in time is under 100
 
Interesting lesson for all right there - thanks for taking the time to share !!!
 
Hahah, "Too much sun" No such thing! Well, unless your charger can't handle it :p
 
Korishan said:
Hahah, "Too much sun" No such thing! Well, unless your charger can't handle it :p

I recognise your user name from some of Petes live youtube sessions I started out with free in need of repair panels all had cracked glass so I repaired them with uv stable crystal poxyvarnish and it worked ive had them on a frame for years and they are still producing connected to my grid tie, I recently bought 4 new not broken panels all 275w if you do the maths on that thats 1100w and the pip 2424mse is maxed out at 1000w and it seems after nearly 2 weeks of perfect sun that little extra was too much and it decided NOPE im done.

Im currantly looking for another charge controller under 100 and I think ive found one it has a weird name and its quite simple its called a 'make skyblue' ive watched a few youtube reviews and I think it will do

image_tgdogq.jpg


until I save up for a midnite classic the one that looks like an old payphone

image_blvdhu.jpg
 
I have 3 of the "MakeSkyBlue" style (there are copies) and have two originals and a copy, the copy does not have the MakeSkyBlue stamped on the case and the PCB inside is also missing the labelling. However, there appears to be bo difference in the devices and the performance is identical... 100% copy, including firmware, or just a batch without names. The onyl bug with the MakeSkyBlue is they occasionally track the voltage down and hod it lower than the actual MPP where you have say 3 panels and one is shaded it will drop the voltage as if one panel is being cut out but will not track back up once unshaded. A quick off and on for the DC side resets back up to the actual MPP. When they are working they give the same output as the Midnight as one youtube video shows, so they are cheap and work, but may not give 100% matching power output compared to the midnight all of the hours in the day... they are howerver a good cheap tracker to have as a backup.. Nothing else under 100 seems reasonable and is actually an MPPT as a lot of them are labeled as MPPT and actually just PWM under the hood...
 
completelycharged said:
I have 3 of the "MakeSkyBlue" style (there are copies) and have two originals and a copy, the copy does not have the MakeSkyBlue stamped on the case and the PCB inside is also missing the labelling. However, there appears to be bo difference in the devices and the performance is identical... 100% copy, including firmware, or just a batch without names. The onyl bug with the MakeSkyBlue is they occasionally track the voltage down and hod it lower than the actual MPP where you have say 3 panels and one is shaded it will drop the voltage as if one panel is being cut out but will not track back up once unshaded. A quick off and on for the DC side resets back up to the actual MPP. When they are working they give the same output as the Midnight as one youtube video shows, so they are cheap and work, but may not give 100% matching power output compared to the midnight all of the hours in the day... they are howerver a good cheap tracker to have as a backup.. Nothing else under 100 seems reasonable and is actually an MPPT as a lot of them are labeled as MPPT and actually just PWM under the hood...

Thanx thats good to know im currently looking at a 30a one it has the makeskyblue on the front on the ebay listing and its only 64 with an ebay promo code ive got so win win.

When I first started I did get stung by the 10 mppt nonsense hopefully because of this site and everyone on youtube there will be less people getting ripped off not that the pwm chargers dont have there uses I use theone I bought to keep my caravans battery topped up when im on holiday.
 
Forgot to mention, electrically thay all appear to be identical internally and I think they may just be firmware limited or missing a FET in the back, that's all..

I have 40A units and the unnamed one is 60A, the top board is identical, I could not check the FET's underneath, but suspect the only difference is either some FETS missing, firmware or a very slight difference with a resistor/pin pacing on the board. I did send them a message to ask how I tell the difference and strangely enough they did not respond.


They are cheaper via Aliexpress.....
 
completelycharged said:
Forgot to mention, electrically thay all appear to be identical internally and I think they may just be firmware limited or missing a FET in the back, that's all..

I have 40A units and the unnamed one is 60A, the top board is identical, I could not check the FET's underneath, but suspect the only difference is either some FETS missing, firmware or a very slight difference with a resistor/pin pacing on the board. I did send them a message to ask how I tell the difference and strangely enough they did not respond.


They are cheaper via Aliexpress.....



Would be interesting to find out the difference especially if its just the mosfets and ive not bought anything off aliexpress ive been wary of buying from abroad since I bought a phone from gearbest and got stung for customs charges.
 
Will have a look when I install the next unit. I think it was the FET's in the back as they seem a pain to get to as they use the aluminium case for cooling so if you mount the case on a sheet of aluminium/steel it helps with the cooling... don't fix them to wood.. Think I have some pictures,



I still have a dispute in progress with Aliexpress - current case of the mysteries of the Chinese postal service and how parcels dissapear or take a very very long time to re-appear.

The parcel was shipped to me and it took just over 24 hours from collection to being on an aeroplane out of China to me, return trip, 7 days missing in a sorting office, on to the next sorting office, 3 days, etc... woder if the return post should have been marked "not from the USA" "No Trump Tariffs" and it might have been given special treatment !!


This is the 60A non MakeSkyBlue PCB

image_shlfmf.jpg



And this is the new one (I ordered a 40A unit) and it is stamped 60A.. I have not tired it yet so it could be 60A or a firmware programmed 40A. If it is a firmware 40A with identical components then it is a very bad sales approach..


image_dxzalr.jpg


Ironically the no-name board has higher voltage rated capacitors, otherwise the board is identical


Appologies about the colours, flourescent mixed with LED mixed with the phone, mixed with the TV screen... the actual colours are the same as the first image..
 
I was wondering about the colour scheme there :p Odd how that works out.

I would say that the cost difference is there for a good reason. Most people would by the 40A. There isn't much of a difference from using 40A components to the 60A version. This makes not on inventory less overhead, but also the tooling to work with the different designs. Then, when someone wants the 60A version, bump the price up and they aren't the wiser. They paid a higher price for a number change. Cha Ching! It happens a lot here in the US.
This is noted that the "fake" one has higher rated caps. Why is that? Because they are basically the same price as the slightly lower rated caps. Only 1 bin of parts. Plus, you don't have to worry about the person who's putting these parts in grabbing the wrong part and slapping it in there. Then blowing the unit up and having to replace under defect for free. Even more Cha Ching in the end.
 
From the surface components they are identical (appart from the 1 bin slection - nice one..), underneath there are two sets of parts, larger switching FET's for the coils and 3 other smaller FETS. Can't really see the part numbers as I would need to take the boards out... the only thing I can think of is that they exclude some FETs in the back, which is picked up by the firmware and then limits the power level, so adding more FET's back in might uprate them..


This one on youtube says 40A and inside on the PCB..... 60A.


This is a full write up and shows the FET's in the back..
https://syonyk.blogspot.com/2018/05/whats-difference-between-150-and-600.html


Wow... interesting post at the end... someone has done thier own firmware update...
 
So you buy the cheaper 40A version, and do the mods to make it a 60A :)




Read the Blog posting. Interesting overall. Initially, I'd be wary of using such a system, and probably not recommend it, especially after reading his issues. The weird "MPPT" tracking going all wonky every so often.
But then there's the guy who is writing a firmware for it. If he gets that working correctly, and the MPPT functions as expected, and several units can be tied together, this would be a great little MPPT charge controller for a wonderful price.

Will be interesting to see how that project progresses.
 
Back
Top