Jump to content

Axpert 5000-48 fault code 02


ProjectGKR

Recommended Posts

Hello

When switching on inverter, it continuously beeps with fault led on with code 02 and then after short while switches off.

As it has been in use for 2 years now, I did blow out dust first but did not solve fault.

Looking at the service manual, I looked at the NTC section for fault code 02. All resistors are in the expected ranges. The ntc on cn4 & cn8 unplugged is above 14.2 & 14.8, not sure if must be exactly 15 as per manual? The ntc1 on the main board value is 7, but i haven't removed it from the board yet. Must I also like with cn4 & 8 remove it in order to get correct reading?

Any thing else I can check?

Link to comment
Share on other sites

1 hour ago, ProjectGKR said:

Any thing else I can check?

If it's really an Axpert inverter (Microcare doesn't seem to be an Axpert, at least the first few searches I did), then you could check what it thinks the temperature is, using either your monitoring software or the Q1 command. This will tell you which of up to 4 sensors is causing the problem: the "battery side" heatsink, the AC side heatsink, the transformer, or possibly the solar charge controller. The Q1 command is undocumented, and seems to change its output, so my page may not match your inverter.

Link to comment
Share on other sites

1 hour ago, Coulomb said:

using either your monitoring software or the Q1 command.

Thanks @Coulomb, will give the Q1 Command a go.

 

1 hour ago, Coulomb said:

If it's really an Axpert inverter (Microcare doesn't seem to be an Axpert, at least the first few searches I did)

Sorry, should have mentioned it is for another setup and not the one as per my signature.

Link to comment
Share on other sites

2 hours ago, ProjectGKR said:

Looking at the service manual, I looked at the NTC section for fault code 02. All resistors are in the expected ranges. The ntc on cn4 & cn8 unplugged is above 14.2 & 14.8, not sure if must be exactly 15 as per manual?

One of mine measured around 13 kΩ. I don't think that's the problem.

Quote

The ntc1 on the main board value is 7,

On the board I have here, it's around 7.5 kΩ with the control board disconnected. The only load on the NTC will be from the control board. One end is +5V, the other end connects directly to the control board (CN10 pin 1), nothing else that I know of.

So that's 3 out of 4. Does the fault go away if you isolate the PV input? If not, it sounds like a control board fault. You may be able to trouble shoot that, though I don't know of any schematics. The circuit has to be pretty simple, just a few resistors and an op-amp or comparator straight into an ADC input of the processor, I would think.

Buying a new control board through the supplier could be a slow process. If you go that route, you'd need to supply a photo of the label on the side. Make it abundantly clear exactly what model it is (e.g. is it a 64 V model, may affect the control board), or you'll get the wrong one. It may also come with a crazy serial number, and getting that fixed is a hassle too. So check for bad connectors, shorts, and components on the control board first. Also the two 750Ω resistors near CN4 and CN8. Pin 1 of CN4 and CN8 are +5V, the 750Ω resistors go from each pin 2 to ground.

Link to comment
Share on other sites

Thanks @Coulomb I removed the mppt completely and tried again but fault remained. I also as per service manual left cn4 & 8 disconnected and fault remained.

2 hours ago, Coulomb said:

control board disconnected

In order for me to measure the resistance of ntc1 at the transformer I had to completely disassemble the inverter, so the control board was disconnected when taking measurement.

I tried the Q1 command via serial port using auto send in Access port software set at 8000ms and at 1000ms but no response from inverter and it shuts down after 30 seconds.  I believe as mentioned that the control board the issue here and will check the components on it but I think it is end of the road  for the inverter, unless I can find a repair centre in Capetown.

 

 

 

Link to comment
Share on other sites

On 2020/12/28 at 5:20 PM, ProjectGKR said:

Using QPIGS command I get a value 0102 for the Inverter heatsink temp. 

That'll do it. But which one is causing the problem? You might have a model that doesn't implement the Q1 command.

On 2020/12/28 at 5:20 PM, ProjectGKR said:

Will ntc on mppt as test work on inverter side?

I don't know. my guess is yes, they would use the same NTCs to save inventory. But that's just a guess.

Link to comment
Share on other sites

8 hours ago, Coulomb said:

You might have a model that doesn't implement the Q1 command.

It is possible that newer firmware version implements this?

What resistance value is expected for the NTC1 when removed from the main board? One I removed is 30.5k ohm. 

When taking a resistance measurement on the main board on the contacts where the NTC1 is removed I get a reading of 6k. All the boards are removed from the main board.

Link to comment
Share on other sites

14 hours ago, ProjectGKR said:

It is possible that newer firmware version implements this?

Yes. Actually, what firmware (display and main) are you running there? All the firmwares I checked recently have it.

Edit: or at least, what model inverter is it?

 

Edited by Coulomb
Link to comment
Share on other sites

9 hours ago, Coulomb said:

Actually, what firmware (display and main) are you running there?

U1 74 10 is what the display shows

9 hours ago, Coulomb said:

what model inverter is it

The model no on the label on the side is Axpert 5000-48. It also notes that it is made for Solar MD PTY Ltd

Edited by ProjectGKR
Spelling
Link to comment
Share on other sites

7 hours ago, ProjectGKR said:

U1 74 10 is what the display shows

Ok, thanks. I only have 74.40 in the 74.XX series, and it does have the Q1 command. There is no removable display firmware to confound matters with this model.

7 hours ago, ProjectGKR said:

The model no on the label on the side is Axpert 5000-48.

Ah, the so-helpful (not!) model numbers some resellers have. (Not your fault, obviously.) That description fits about half a dozen different models (PF0.8 and PF1, low and high voltage SCC, 64 V or not, with and without removable display, parallelable and not, single/dual/triple SCC, MPPT or PWM, and double conversion or not). All of them 5000 VA, 48 V models.

In case you care, yours is PF1, 145 V max SCC, 58.4 V max battery voltage, no removable display, parallelable, single MPPT SCC, and not double conversion.

Link to comment
Share on other sites

Q1 command is working. (Typo in the Hex for the commands I created and saved to text file for easier use. Sorry @Coulomb my bad)

So this is the findings using the Q1 command

initial
(00000 00000 00 00 00 000 097 030 030 00 00 000 0100 0000 0000 50.00 10 1 060 015 060 005 58.40 000 600 0 0258—
(00000 00000 00 00 00 000 097 030 030 00 00 000 0100 0000 0000 50.00 10 1 060 015 060 005 58.40 000 600 0 0258—
(00000 00000 00 00 00 000 094 030 030 00 00 000 0100 0000 0000 50.00 10 1 060 015 060 005 58.40 000 600 0 0258õÈ
(00000 00000 00 00 00 000 092 030 031 00 00 000 0100 0000 0000 50.00 10 1 060 015 060 005 58.40 000 600 0 0258Ä[
(00000 00000 00 00 00 000 090 030 031 00 00 000 0100 0000 0000 50.00 10 1 060 015 060 005 58.40 000 600 0 0258t1

cn4 disconnect
(00000 00000 00 00 00 000 091 000 028 00 00 000 0100 0000 0000 00.00 00 1 060 015 060 005 58.40 000 600 0 0258Ú&
(00000 00000 00 00 00 000 097 000 028 01 00 000 0100 0000 0000 50.00 10 1 060 015 060 005 58.40 000 600 0 0258àb
(00000 00000 00 00 00 000 099 000 030 00 00 000 0100 0000 0000 50.00 10 1 060 015 060 005 58.40 000 600 0 0258Bœ
(00000 00000 00 00 00 000 097 000 031 00 00 000 0100 0000 0000 50.00 10 1 060 015 060 005 58.40 000 600 0 0258’å
(00000 00000 00 00 00 000 096 000 031 00 00 000 0100 0000 0000 50.00 10 1 060 015 060 005 58.40 000 600 0 0258ÊÐ
(00000 00000 00 00 00 000 094 000 031 00 00 000 0100 0000 0000 50.00 10 1 060 015 060 005 58.40 000 600 0 0258zº
(00000 00000 00 00 00 000 092 000 031 00 00 000 0100 0000 0000 50.00 10 1 060 015 060 005 58.40 000 600 0 0258º%

cn8 disconnect
(00000 00000 00 00 00 000 097 030 031 00 00 000 0100 0000 0000 49.99 10 1 060 015 060 005 58.40 000 600 0 0258Èš
(00000 00000 00 00 00 000 099 030 031 00 00 000 0100 0000 0000 50.00 10 1 060 015 060 005 58.40 000 600 0 0258Íî
(00000 00000 00 00 00 000 097 030 031 00 00 000 0100 0000 0000 50.00 10 1 060 015 060 005 58.40 000 600 0 0258ì›
(00000 00000 00 00 00 000 094 031 031 00 00 000 0100 0000 0000 50.00 10 1 060 015 060 005 58.40 000 600 0 0258˜\
(00000 00000 00 00 00 000 087 031 031 00 00 000 0100 0000 0000 50.00 10 1 060 015 060 005 58.40 000 600 0 0258žþ
(00000 00000 00 00 00 000 087 031 031 00 00 000 0100 0000 0000 49.99 10 1 060 015 060 005 58.40 000 600 0 0258ºÿ

cn4 & cn8 disconnect
(00000 00000 00 00 00 000 096 000 031 00 00 000 0100 0000 0000 50.00 10 1 060 015 060 005 58.40 000 600 0 0258ÊÐ
(00000 00000 00 00 00 000 096 000 031 00 00 000 0100 0000 0000 50.00 10 1 060 015 060 005 58.40 000 600 0 0258ÊÐ
(00000 00000 00 00 00 000 096 000 031 00 00 000 0100 0000 0000 50.00 10 1 060 015 060 005 58.40 000 600 0 0258ÊÐ
(00000 00000 00 00 00 000 093 000 031 00 00 000 0100 0000 0000 50.00 10 1 060 015 060 005 58.40 000 600 0 0258â
(00000 00000 00 00 00 000 091 000 031 00 00 000 0100 0000 0000 49.99 10 1 060 015 060 005 58.40 000 600 0 0258v{
(00000 00000 00 00 00 000 088 000 031 00 00 000 0100 0000 0000 50.00 10 1 060 015 060 005 58.40 000 600 0 0258X

 

What is interesting is that disconnecting cn8 had no effect on the output

 

Link to comment
Share on other sites

So with the Q1 command now sorted, I decided to add a additional fan using the fan connection at the top of the main board and have it blowing directly on the NTC1. The Temp actually started dropping each time I switch on the inverter, and after about 5x switch the inverter on after auto shutdown the fault error and beep actually stopped at round 67. But it still switched off after 30 seconds. I then changed the battery off voltage as my test battery bank is a bit on the low side, thinking it is due to low voltage, but had no effect as it still does the auto shutdown. I then when it switched on, performed the factory reset via Watch Power. During this process I could hear a relay click and the inverter after its reboot still had no fault and auto shutdown solved. Using the Q1 command again, the temp for all sensors stayed at between 33 - 36. I then decided to move the extra fan to the exit vents on the control board and put all together. And again when switching on, no fault and the Q1 command stayed in the 33-36 temp range.Will let it run for a while and check again later.

 

It is possible the NTC1 is starting to fail. Does anyone maybe have the detail on what exact NTC I need to use to replace the old one?

Link to comment
Share on other sites

On 2021/01/03 at 3:34 AM, ProjectGKR said:

cn8 disconnect
(00000 00000 00 00 00 000 097 030 031 00 00 000 0100 0000 0000 49.99 10 1 060 015 060 005 58.40 000 600 0 0258Èš

 

On 2021/01/03 at 3:34 AM, ProjectGKR said:

What is interesting is that disconnecting cn8 had no effect on the output

Yes. That's the problem. The second set of 4-digit number is always in the 90s or high 80s; this is the "inverter" temperature, from the sensor connected to CN4. Are you sure you got CN4 and CN8 right in your post?

But it hardly matters: one of the readings is way too high, even with the sensor disconnected. So the problem has to be either the wires to the control board, or the op-amps or comparators on the control board, or components very close to those. Fortunately, you have a good channel to compare with.

The attached diagram (which I drew up for a fellow forumite who is getting a complete schematic ready for publication) may help. [ Edit: That came out wrong. 99% of the work is his, not mine. I just added the purple parts. ]

Unfortunately, I don't know of any schematic traces for the control board. But that part of the circuit is pretty easy to follow, apart from the traces that go to the actual ADC inputs of the DSP. It seems like it should be a fairly easy fix.

 

Near CN4, CN8 mod.png

Edited by Coulomb
Modified the diagram slightly
Link to comment
Share on other sites

On 2021/01/03 at 4:43 AM, Coulomb said:

Are you sure you got CN4 and CN8 right in your post?

Yes, double checked. the high 90s was reported with CN4 and CN8 disconnected. only the "Battery Temperature" value dropped to 000

 

On 2021/01/03 at 4:43 AM, Coulomb said:

The attached diagram (which I drew up for a fellow forumite who is getting a complete schematic ready for publication) may help

Thanks for the diagram. This assisted in getting a replacement NTC to replace the Transformer NTC. Alos got some NTC replacement probes for the heatsinks. Will be dismanteling and replacing the NTC tommorrow and give feedback.

Edited by ProjectGKR
Additional info
Link to comment
Share on other sites

On 2021/01/03 at 12:43 PM, Coulomb said:

The attached diagram (which I drew up for a fellow forumite who is getting a complete schematic ready for publication) may help.

That came out wrong. 99% of the work is his, not mine. I just added the purple parts, so he could correct his schematic.

Link to comment
Share on other sites

On 2021/01/04 at 8:50 PM, ProjectGKR said:

Will be dismanteling and replacing the NTC tommorrow and give feedback

Replaced all the NTC's and fault remained.

Using the Q1 command the 2nd set of 3 digits remained on 102. CN4 when disconnected reflects on the 3rd set of 3 digits. Disconnnecting it it goes to 000 and I also tested that the probe temp changes on CN4 when sticking the probe in ice water. Removing NTC1 reflects on the 4th set of 3 digits. With it removed also showed 000. CN8 when removed did not reflect.

Using the schematic for the ntc on the main board, all resistance values are as expected on cn10 pin 1-3 when disassembled.

With inverter switched on, I get a resistance reading from pin1 to Ground and pin 2 to ground. Pin 3 to ground getting no reading.

I did manage with battery bank voltage above 50v, that the fault after a few restarts disappears, but as soon as it drops under 50v, the fault returns.

 

Link to comment
Share on other sites

23 hours ago, ProjectGKR said:

With inverter switched on, I get a resistance reading from pin1 to Ground and pin 2 to ground. Pin 3 to ground getting no reading.

That sounds like when you plug in the control card, CN8 gets shorted.

I had a look at some photos of control boards (I don't have one handy, unfortunately). It looks like there is no op-amp or comparator between the DSP and the temperature sensors; it's basically just an RC filter and a double-diode clamp. My guess is that the lower diode of the diode clamp is shorted. Either that, or the DSP's ADC input is shorted. Let's hope it's the dual diode. Can you read a marking code on it, or one of its neighbours?

Edit: your control board may not look exactly the same, but I suspect it will be close. [ Edit: I see it's not all that close. ]

Temp sensor ADCs.jpg

Edited by Coulomb
Link to comment
Share on other sites

3 minutes ago, Coulomb said:

My guess is that the lower diode of the diode clamp is shorted.

Now that I think about it, you should measure 1.0 kΩ, not zero ohms, from CN10 pins 3 to GND (e.g. CN10 pins 8-11). BTW, GND on the diagram I posted a few posts back is digital ground, not necessarily the same as mains earth. I can't recall, but I think that digital ground is essentially AC-in neutral.

You should not be making resistance measurements with the inverter powered up.

Can you carefully measure the resistance of pins 1, 2, and 3 to digital ground on the control board when it's not powered up? I suspect pin 3 will be different to the other two; it might be around 1 kΩ when the others are much higher. If so, you could try removing the diode clamp whose middle pin connects to CN10 pin 3; the circuit should work for testing purposes without it. Obviously, it's an important protection device, so replace it as soon as you can if removing it fixes the problem.

I can't imagine why battery voltage would affect whether it comes good or not; perhaps it's just intermittent and the "connection" is a coincidence. Battery voltage is over towards the left of the control board, and I suspect it goes into one of the ADC A inputs (pins 16-23, around the corner of the DSP chip).

Link to comment
Share on other sites

Thanks @Coulombfor the guidance.

 

5 hours ago, Coulomb said:

your control board may not look exactly the same, but I suspect it will be close.

I have attached photos of the controlboard.

IMG-20210107-WA0006.jpg

IMG-20210107-WA0007.jpg

 

4 hours ago, Coulomb said:

You should not be making resistance measurements with the inverter powered up.

Moment of weakness, hoping for a AHA moment

Edited by ProjectGKR
Images under correct quote
Link to comment
Share on other sites

14 hours ago, ProjectGKR said:

I have attached photos of the controlboard.

Ah, you have the other kind. It still looks like pin 3 should connect to one of the circled diode midpoints, probably via a 1kΩ resistor. My guess is it's more likely one of the the red ones, else one of the orange ones.

It will still be very worthwhile measuring the resistance from pin 3 to one of the pins 8-11 on CN10, and comparing it with the resistance from pin 1 or 2 to 8-11.

Temp sensor ADCs 2.jpg

Link to comment
Share on other sites

On 2021/01/08 at 8:10 AM, Coulomb said:

It will still be very worthwhile measuring the resistance from pin 3 to one of the pins 8-11 on CN10

With the board removed from the inverter, the resistance from pin 3 to 8-11 is 7.7k

 

On 2021/01/08 at 8:10 AM, Coulomb said:

and comparing it with the resistance from pin 1 or 2 to 8-11

With the board removed from the inverter, the resistance from pin 2 to 8-11 is 7.6k and from from pin 1 to 8-11 is 7.7k

Link to comment
Share on other sites

9 hours ago, ProjectGKR said:

With the board removed from the inverter, the resistance from pin 3 to 8-11 is 7.7k

Ok, that pretty much shoots down my diode theory.

Sadly, that basically leaves the DSP chip itself. It's more work than replacing just the chip, too. I'm still working with Adriaan Scheffers on how to restore the boot loader.

So a control board replacement would seem the only reasonable option.

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...