Jump to content

Axpert MKS 5kv Inverter


Instinct

Recommended Posts

Could somebody please explain most common reasons for the above inverter to fail completely. Been working perfectly for six months. Heard slight noise like breaker switch and inverter was completely dead. Tried reset but no luck. Sent back to supplier. Would like to know what to do to prevent new one doing same.

Link to comment
Share on other sites

1 hour ago, Instinct said:

Could somebody please explain most common reasons for the above inverter to fail completely. 

The most common complete failure mode is a bunch of MOSFETs blowing up, often with a pair of IGBTs as well. If the inverter powers up at all, you'll see fault code 09 (bus soft start failure). The bus won't charge because of shoot-through by the IGBTs. Often, the capacitors protecting the MOSFETs (rated for 2000 h of use at 105°C) dry out, allowing higher and higher voltage spikes across the MOSFETs, which eventually break down. The resultant surge may be what takes out the IGBTs. Usually several gate drive components fail as well (resistors, diodes, zener diodes, transistors).

This is why I preemptively replaced my MOSFETs with higher voltage parts, and the capacitors with longer life parts. It's a pain to do unless you're set up for it, but it eliminates the most common cause of failure. Other than that, keeping the unit cool (out of direct sunlight for example) is about all you can do.

Link to comment
Share on other sites

1 hour ago, Coulomb said:

 

This is why I preemptively replaced my MOSFETs with higher voltage parts, and the capacitors with longer life parts. It's a pain to do unless you're set up for it, but it eliminates the most common cause of failure. Other than that, keeping the unit cool (out of direct sunlight for example) is about all you can do.

The 5000 Watt Mecer produced March 2018 was fitted with CSD19505KCS MOSFETs 80 volt . The capacitors is 2 Jamicon 80 volt 3300 uf and 2 Jainghai CD 294.  80 volt 2200 uf . The capacitors is now 22 mm diameter not 18 mm . Would you still recommend United chemi - con EKZN 800 ELL 182 mm 405  1800 uf 10 000 hrs .Thanks for always sharing your knowledge with us .

 

 

 

Link to comment
Share on other sites

2 hours ago, Coulomb said:

The most common complete failure mode is a bunch of MOSFETs blowing up, often with a pair of IGBTs as well. If the inverter powers up at all, you'll see fault code 09 (bus soft start failure). The bus won't charge because of shoot-through by the IGBTs. Often, the capacitors protecting the MOSFETs (rated for 2000 h of use at 105°C) dry out, allowing higher and higher voltage spikes across the MOSFETs, which eventually break down. The resultant surge may be what takes out the IGBTs. Usually several gate drive components fail as well (resistors, diodes, zener diodes, transistors).

This is why I preemptively replaced my MOSFETs with higher voltage parts, and the capacitors with longer life parts. It's a pain to do unless you're set up for it, but it eliminates the most common cause of failure. Other than that, keeping the unit cool (out of direct sunlight for example) is about all you can do.

Thank you so much Coulomb for giving good and expert advice.Wont be able to replace components as unit still under warranty.Will try and keep unit as cool as possible.

Link to comment
Share on other sites

8 hours ago, Chris Louw said:

The 5000 Watt Mecer produced March 2018 was fitted with CSD19505KCS MOSFETs 80 volt .

Those are probably not too bad if you have a lowish battery voltage, e.g. Pylontechs with sane battery voltage settings. But if you need to go near 60 V battery at times, then 80 V is only 7% better than the original 75 V, and 64 V (the max) is 10% higher than 58.4 V (the old max battery voltage).

8 hours ago, Chris Louw said:

The capacitors is 2 Jamicon 80 volt 3300 uf and 2 Jainghai CD 294.  80 volt 2200 uf . The capacitors is now 22 mm diameter not 18 mm . Would you still recommend United chemi - con EKZN 800 ELL 182 mm 405  1800 uf 10 000 hrs .Thanks for always sharing your knowledge with us .

Fascinating that they still use two different brands, when they are all electrically in parallel. It really depends on the lifetime rating of the capacitors. I doubt that they are using 10,000 h parts; I don't have time to check just now.

Link to comment
Share on other sites

  • 1 month later...
On 2020/07/13 at 6:34 PM, Instinct said:

Thank you so much Coulomb for giving good and expert advice.Wont be able to replace components as unit still under warranty.Will try and keep unit as cool as possible.

Morning Coulomb

Took your advice on replacement unit by keeping it as cool as possible. Unit failed again(2nd) yesterday with quite a loud crack sound.Returning again for warranty replacement. This is so frustrating. Is there nothing else one could do on settings that would make a difference?

Link to comment
Share on other sites

  • 4 weeks later...
On 2020/07/14 at 2:46 AM, Coulomb said:

Those are probably not too bad if you have a lowish battery voltage, e.g. Pylontechs with sane battery voltage settings. But if you need to go near 60 V battery at times, then 80 V is only 7% better than the original 75 V, and 64 V (the max) is 10% higher than 58.4 V (the old max battery voltage).

Fascinating that they still use two different brands, when they are all electrically in parallel. It really depends on the lifetime rating of the capacitors. I doubt that they are using 10,000 h parts; I don't have time to check just now.

Morning Coulomb

I am getting rather desperate as 3rd unit popped yesterday. Quite a loud cracking noise. Unit completely dead and wont turn on again. Tried everything. Temp just before blowout was only 37 deg C ? Load was running approx 500w. Added a 30A voltage protector unit. Warranty most probably replace or repair. Worried same thing will happen. Must I rather look at another unit ?

Instinct

Link to comment
Share on other sites

19 hours ago, Coulomb said:

Yikes.

What battery are you using, and what values for settings 26 and 27?

What panels are you using, and what configuration (how many in series, how many strings)?

Hi Coulomb

The settings for both 26&27 is 54V.The battery is a 48V Lithium 100ah.Panels are 9x 405w Canadian, placed 3xseries in 3 strings. Panels deliver a average of 137V.

Link to comment
Share on other sites

49 minutes ago, Instinct said:

The settings for both 26&27 is 54V. The battery is a 48V Lithium 100ah.

Panels are 9x 405w Canadian, placed 3xseries in 3 strings. Panels deliver a average of 137V.

I'm sorry to say that both of these could be contributing to the failures. I assume that the battery is a 15S type; 54 V is not high enough to properly charge a 16S type. But 54 V for a 15S LFP battery is 3.60 VPC average; usually this is the never-exceed / alarm / disconnect immediately voltage. That means that the BMS would be disconnecting the battery regularly, including under heavy loads. Axperts are renowned for over- (and under-) shooting their setpoints. I don't know exactly how bad this is, but I'd certainly not be comfortable with it. I can readily imagine huge spikes of voltage on the battery bus, chipping away at the MOSFET breakdown voltage safety margin until they fail.

137 V is above the maximum MPPT voltage, and it's even above the derating voltage of 130 V. So at 137 V, the solar charge controller is backing off the power by 7/15 = 47% of rated power, to protect itself. As the voltage nears 145 V, that power reduces to zero. Above 145 V, even with no power throughput, the solar charge controller can still be damaged, and the hardware has no way of protecting itself (it can't open a relay to isolate the panels, for example). A failure in the solar charge controller could conceivably damage other parts of the inverter-charger, or it might be the source of the loud bang all on its own (it's a 4000 W power converter, after all).

If they let you have a fourth unit, please reduce settings 26 and 27 to the forum-accepted values of 52.5 V and 51.8 V, and rewire your panels to 2S (either 4P with one unused panel, or 5P and add another). You'll likely get more power from 8 panels wired 2S4P than nine panels wired 3S3P anyway.

Link to comment
Share on other sites

Hi Coulomb

Thank you once again for your kind help and advice. The battery is a Averge Solar LFeLi 48100TB. Equalized Chrg Volt=54-56V.      Floating Chrg Volt = 54V.      Disharge End Volt = 40.5. (don't know if this will help) I will definitely do settings 26 & 27 according to your recommendation. Will also rewire panels to 2S4P.

Thank you

Instinct

 

 

 

On 2020/09/26 at 3:28 PM, Coulomb said:

Yikes.

What battery are you using, and what values for settings 26 and 27?

What panels are you using, and what configuration (how many in series, how many strings)?

Hi Coulomb

The settings for both 26&27 is 54V.The battery is a Lithium 100ah.Panels are 9x 405w Canadian, placed 3xseries in 3 strings. Panels deliver a average of 137V.

Link to comment
Share on other sites

I got exactly the same system as Instinct. I just now 1 hour after sun's zenit measured the voltage at the MPPT PV input: 100V at almost max power: 2.8kW.

The PV panels are in 3s3p configuration. The max OCV I measured is approx 126V. The max PV current is said to be 9A, that totals to max 27A at the MPPT input. I wired with 6mm2 wire section.

I now changed 26 to Coulomb's recommendation, 27 was already there.

Link to comment
Share on other sites

22 hours ago, Instinct said:

Panels deliver a average of 137V.

So the specifications of the device states 60-115v and you are running it on average at 137v and you are wondering why it is blowing up...

Edited by Gnome
Link to comment
Share on other sites

22 hours ago, Gnome said:

So the specifications of the device states 60-115v and you are running it on average at 137v and you are wondering why it is blowing up...

Not very logic. The Axpert MKS 5K hybrid system consists of 3 distinct components: The inverter, the 230V charger and the MPPT. If PV overvoltage harms it would damage the MPPT, not the inverter. I suspect the MPPT has its own controller with the U2 firmware. I noticed that even with the inverter turned off at the toggle switch the MPPT continues to charge the batteries.

Link to comment
Share on other sites

2 hours ago, Beat said:

Not very logic. The Axpert MKS 5K hybrid system consists of 3 distinct components: The inverter, the 230V charger and the MPPT.

Why is it not logical?  Where do they state they individual components? The inverter is basically a boost converter with a rectifier followed by an AC output stage.  When the inverter isn't in use the boost converter can be used as a buck converter to charge the battery.  The charger and the inverter are the same component.

The MPPT isn't isolated.  This isn't even unique to Axpert.  If you over volt any kind of electrical component that isn't isolated by a transformer, Y class capacitor or opto-isolator, you'll blow right through the MOSFET and fry anything that either isn't isolated or not rated for the voltage.

Only grid-tied micro inverters (the kind that directly outputs AC from a few panels) would stand a chance of not blowing up your inverter because the devices sit between transformers.

2 hours ago, Beat said:

I suspect the MPPT has its own controller with the U2 firmware

Still doesn't change the fact that there is no isolation...

Fact remains, if the device has a rating and you don't follow those ratings bad things happen.  Only a non-engineer would say things like, but I thought they were separate.  That just isn't how engineering works, you don't build for every possible eventuality, and when you explicitly put something in a datasheet as an absolute limit, you design around those. 

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...