*A quick update on this one. I've ordered some 93C66WP eproms. There is some conjecture that this chip does not hold firmware for the air con, it holds data relating to the last user settings and will cause this error if the microcontroller can not read the eprom at power on. The best way to prove this is to fit a new one and see what happens. At that time I will also see if I can read the contents of the original one. These chips are very cheap so it isn't a problem just to order some and see*
I work on large UPS systems for hyperscale data centers, and we don't troubleshoot anything at the component or even board level. We swap entire subsystems to get the equipment back online as soon as possible. It's nice to see someone poke around a board and try to figure out how things work and where the fault might be. I watch your videos for relaxation.
You'd never guess it but I just repaired a similar set of controls for a window air con. Actually there are some similarities. Both have inverter driven compressors (Three phase). The unit I repaired had a failed optocoupler on the main (inverter) board which received the data from the front keypad/display board. Identical opto sending data to the front keypad board. $300 AC and a $1 opto fixed the problem. I could see the data stream driving the LED on the opto with my scope, but clearly nothing was happening on the photo transistor side of the opto. Certainly not a EEPROM issue on the one I fixed. You should OCR the specific model number(s) for the controller/ac unit and post them next. Ethan
I was hoping my favorite electronics guru would start troubleshooting hvac inverter systems. It’s my lucky day! HVAC is my profession but most of us know nothing about repairing the pcb’s. There are very few YT’ers repairing these in English, but from what I have learned from them, that “strange IC” IPM (Intelligent Power Module) is a common problem. Drives the compressor and or fan motor UVW windings. Also Optocouplers are common problem for comm errors. I’m setting up my little electronics lab to repair these inverter systems. Most of mine are newer and still under warranty, but after the warranty runs out, I hope to be a little more knowledgeable with the help of several books and great guys like you on YT. Thank you!
Yes. My IPM exploded and killed the EEPROM (or maybe it was the other way around, I'm not sure). The thing is: the EEPROM must be programmed, and I couldn't find the file to program it anywhere. I hope Richard have better luck!
Thank you. I am no expert on Air Con just approaching it from a purely electroncis point of view. Considering how expensive the PCB are to replace I'm a bit surprised there is not a whole load of techs already fixing these.
I see a lot of comments about it being more cost effective to simply replace the boards... Well one of my first jobs as an electronics tech was to repair some pretty complex microprocessor controlled alarm panels. The boards were eye-wateringly expensive and so we, as the manufacturer, offered a buy-back scheme where we sold a PCB at full cost to the installer, then credited his account with 75% of the cost when he mailed the faulty part in. It made economic sense, especially if muggins here could actually repair them. Most of the time the faults were simple, but sometimes you could spend days tracking an issue down, which definitely had an effect on the economics. However, the fact most boards were fixed in an hour or less (usually because the installer had done something stupid when wiring units up) offset the lengthier repairs. My first exposure to using curve tracers was in that job - loved what I did there, I still see some of those panels in use today, and that job was 45 years ago! Looking forward to the next update on this one.
Thank you for doing this video. I have tried to repair several similar non-working main boards for different mini splits (220/240v), unsuccessfully. There is very limited information online about component level repair & troubleshooting these boards. Schematics aren't available. I am limited in my ability to test where I do my work, as I do not have access to 220v (only 110v - US). I would like to see if there is a method to troubleshoot using voltage injection with a lower voltage since there are areas on the board that do use a lower voltage. My frustrated experiences with these boards lead me to believe they are software related failures after methodically evaluating all components (except controller type IC's) for their integrity, which is an extremely time consuming task. These boards are expensive to replace and it would be fantastic to know if there is a known point of chronic failure.
you must check if there is communication between the inside unit and the outside unit and take a measurement with the oscilloscope in eeprom.sometimes if you read and program eeprom it comes back.
nice video & explanations. Yes you should give a try with the thermal camera on the other side of the board.( & the EEPROM or around , to check if it receives good vcc but no moves on oscilloscope is very suspicious)
if you are not familiar with these, the inputs and outputs can vary a bit from the long term accepted norm. Good to see that even the semi conductor community is being more progressive and inclusive!
Is the -9V good and +15V bad? Did you look at the bottom of the board with the thermal camera? The EEPROM had discolored coating surrounding it. I was thinking either moisture ingress or over-heating.
These serial EEPROMs generally store only the settings and usage counters and stuff like that (this one is just a 4 kbit = 512 byte job), so chances are you could put in a blank one and the unit would behave like straight out of the factory. The channel Buy It Fix It has dealt with issues like these in several videos, have a look. -9/+15 does in fact seem super sus. Maybe that was supposed to be +/-12 and there's a bad ground connection somewhere? Reminds me of that time when I was troubleshooting an Onkyo T-9900 tuner that was intermitttently acting very strange... it ultimately turned out that there were some bad solder joints on one of the copper bus bars used for grounding. Kind of a bear to find.
Yes i do suspect there is a gorund issue (the the +/- supply is not referenced to the same gorund as the 5V). Thanks for the advice towards some research on other YT channels, it is much appreciated
I did wonder that too but can't see it in the resistance readings to ground. TBH I could not even figure out where the +/- 15V? is generated on the pcb yet
Heya, a tutorial about IGBT's would be great as I have an all in 1 hybrid inverter which I shorted beyween PV+ and PV-. at 24:00 minutes in the pin function table I see "serial data clock and input could you explain what that is and/or do I mean serial. yes a shematic would be nice
That is weird maybe the negative voltage cap is faulty or something, or maybe an optocoupler issue becuase the other optocoupler maybe is regulating the negative side.
remove the ic and see if -15V is restored. Also, reading one of the comments below reminded me that i think i saw the position of the optocoupler where there is a slot in the board. This may play a role in regulation.
You can understand why customer facing engineers change the whole board instead of fault-finding, The customer wants it up and running without having to pay for 20 hours diagnostics. Part 50p + labour £1000 vs part £200 + labour £50. If you have many with the same fault, then time invested in understanding the problem can offset repair time / cost.
I can agree with the pont the the customer wants it up and running... but invest some time in repairing all the faulty ones you aquire 'for free' and then you can sell them back to new customers for pretty ,much full cost and make fart more profit than buying and selling a new one? Suerly that is a better business model?
@@LearnElectronicsRepair He he he he, you said "fart" he he he he. That's my childish side coming out. Hope your AC friend can supply you with more boards to build your knowledge base up. Keep up the good work, love your vids.
18:40 - If it hadn’t made me so sick, that might have been funny. I don’t care _what_ those people do behind closed doors: just _don’t_ make me watch! 12:39 - Richard, wouldn’t it be sad if all you had here was a wiring connection issue. This board is throwing the same error as a faulty indoor board when it is not connected. There is as you said earlier, feedback communication from the indoor board on either or both the 6 pin and 8 pin terminals (TB2). There may be nothing wrong with the indoor board.
great video as always richard ... sorry cannot give any info on anything to help you im not technically minded just like watching to gain knowledge ...
Have used split rails with a different negative voltage if an opamp doesn't have the full voltage rating and you don't need a symmetrical swing. Anyway the negative rail still need further investigation. I would expect the main firmware in the the controller and maybe just configuration data in the eprom. Unless the chip is an fpga with an external configuration epom. It seems that nowadays i create a directory and download the datasheets on all the components, and give them a quick once over as this aids the old grey cells. Another reason for just swapping boards is reliability issues. It costs more to revisit customers and they get pissed off with recurring issues. Easer to just replace the board and employ a fitter instead of a engineer. A fitter will never get bored replacing boards, but an engineer will soon get fed up replacing fuses etc. Its just a sad fact that we have been living in a throw away society for decades, i knew of a monitor company who just looked at the condition of the monitor to decide how much the warranty repair would cost, scrap the monitor and send them a new one and change for the repair. No expensive engineer, no reliability issues, quick turnarounds and happy customer's. I rest my csse, it was getting heavy anyway 😂😢
But very practical/profitable/lucrative for anyone buying/scrounging scrap boards for practically nothing and fixing them to resell at say 50-60% of new replacement cost don't you think? Or even more % of replacement cost if they are no longer available to by from the manufacturer? Whay not even sell them back to the 'industry' who think it is not worth fixing them? Surely there must be good profit in this... and from what I am learning so far at something I never tried to fix before it is not so difficult, only tackled three so far but fixed two including the one with the fried lizard on it and I am coming at this as an absolute beginner with Air Con.
now day the repair guide is to replace the PCB it self or throw the entire unit to trash and buy a new one typical technique .you did not get any info about the error code or what it mean .i missed the day when the device have the full schematic and datasheet available on the net ,even in the user manual it have some technical guide. Like what exactly the error code mean.i hate a company treat me like an idiot "unplug the device and plug it again " what a lame thing to explain an error.
@@LearnElectronicsRepair it's more the time the system is down that really makes this the immediate action. Also, many HVAC guys don't really do electronics. So it would have to be a call out or sent to an electronic tech... And, as much as the boards are all the same, they're all different enough to make holding stock unviable. Kind of like laptops etc...
@@joeds3775 How about buying/scrounging all the scrap ones, fixing as many as you can, and selling them back (on ebay or wherever) to the guys who can't be arsed to fix them for say 60% of the cost of a new replacement? Does that sound kinda profitable?
@@LearnElectronicsRepair definitely feasible but expect a letter from manufacturers... I used to work for a manufacturer... Spent a week fixing a room full of boards... Then did a presentation about faults found etc...lead balloon scenario. Even on out of stock items biig contracts were screaming for...It's definitely discouraged. The client swapped all our stuff out for our market leading competitors, we were number 2... I left. Saddened.
I hate these mini-split systems. You can't clean them propperly and the controls are overkill. You can do everything (aside from the inverter part) with simple mechanical relays and cotactors and thermostats. In my opinion, these are more disposable than a window unit.
@@AnnaVannieuwenhuyse I am an HVAC journeyman. I know what they are. I would rather have a traditional split system than a mini split. I only used a window unit as an example of quality and dosposability.
*A quick update on this one. I've ordered some 93C66WP eproms. There is some conjecture that this chip does not hold firmware for the air con, it holds data relating to the last user settings and will cause this error if the microcontroller can not read the eprom at power on. The best way to prove this is to fit a new one and see what happens. At that time I will also see if I can read the contents of the original one. These chips are very cheap so it isn't a problem just to order some and see*
I work on large UPS systems for hyperscale data centers, and we don't troubleshoot anything at the component or even board level. We swap entire subsystems to get the equipment back online as soon as possible. It's nice to see someone poke around a board and try to figure out how things work and where the fault might be. I watch your videos for relaxation.
And even for you it may be really nice to know what could've happened, right? I find it gratifying to find what happened. It's like a treasure hunt.
@@AnnaVannieuwenhuyse Yeah me too
You'd never guess it but I just repaired a similar set of controls for a window air con. Actually there are some similarities. Both have inverter driven compressors (Three phase). The unit I repaired had a failed optocoupler on the main (inverter) board which received the data from the front keypad/display board. Identical opto sending data to the front keypad board. $300 AC and a $1 opto fixed the problem. I could see the data stream driving the LED on the opto with my scope, but clearly nothing was happening on the photo transistor side of the opto. Certainly not a EEPROM issue on the one I fixed. You should OCR the specific model number(s) for the controller/ac unit and post them next. Ethan
Thanks for the help and advice Ethan it is much appreciated
I was hoping my favorite electronics guru would start troubleshooting hvac inverter systems. It’s my lucky day!
HVAC is my profession but most of us know nothing about repairing the pcb’s. There are very few YT’ers repairing these in English, but from what I have learned from them, that “strange IC” IPM (Intelligent Power Module) is a common problem. Drives the compressor and or fan motor UVW windings. Also Optocouplers are common problem for comm errors. I’m setting up my little electronics lab to repair these inverter systems. Most of mine are newer and still under warranty, but after the warranty runs out, I hope to be a little more knowledgeable with the help of several books and great guys like you on YT.
Thank you!
Yes. My IPM exploded and killed the EEPROM (or maybe it was the other way around, I'm not sure).
The thing is: the EEPROM must be programmed, and I couldn't find the file to program it anywhere. I hope Richard have better luck!
@@viniciusvbf22 Well someone should get an EEPROM from a working unit and post it online :)
@@viniciusvbf22 I will see how it goes... there will be a part 2
Thank you. I am no expert on Air Con just approaching it from a purely electroncis point of view. Considering how expensive the PCB are to replace I'm a bit surprised there is not a whole load of techs already fixing these.
Fixed my minisplit a couple of months ago. Internal board had a blown schottky. Cost £0.70 . I couldn't believe i fixed it!
I see a lot of comments about it being more cost effective to simply replace the boards... Well one of my first jobs as an electronics tech was to repair some pretty complex microprocessor controlled alarm panels. The boards were eye-wateringly expensive and so we, as the manufacturer, offered a buy-back scheme where we sold a PCB at full cost to the installer, then credited his account with 75% of the cost when he mailed the faulty part in. It made economic sense, especially if muggins here could actually repair them. Most of the time the faults were simple, but sometimes you could spend days tracking an issue down, which definitely had an effect on the economics. However, the fact most boards were fixed in an hour or less (usually because the installer had done something stupid when wiring units up) offset the lengthier repairs.
My first exposure to using curve tracers was in that job - loved what I did there, I still see some of those panels in use today, and that job was 45 years ago!
Looking forward to the next update on this one.
Thank you for doing this video. I have tried to repair several similar non-working main boards for different mini splits (220/240v), unsuccessfully. There is very limited information online about component level repair & troubleshooting these boards. Schematics aren't available. I am limited in my ability to test where I do my work, as I do not have access to 220v (only 110v - US). I would like to see if there is a method to troubleshoot using voltage injection with a lower voltage since there are areas on the board that do use a lower voltage. My frustrated experiences with these boards lead me to believe they are software related failures after methodically evaluating all components (except controller type IC's) for their integrity, which is an extremely time consuming task. These boards are expensive to replace and it would be fantastic to know if there is a known point of chronic failure.
you must check if there is communication between the inside unit and the outside unit and take a measurement with the oscilloscope in eeprom.sometimes if you read and program eeprom it comes back.
nice video & explanations. Yes you should give a try with the thermal camera on the other side of the board.( & the EEPROM or around , to check if it receives good vcc but no moves on oscilloscope is very suspicious)
An LBGT - love it !! 😂
I'm still looking for the data sheet !
@@we-are-electric1445 Have you tried typing LGBTQ or perkaps even LGBTQ+ :p
@@semiRockethr Doesn't work - must be a one off custom device.
if you are not familiar with these, the inputs and outputs can vary a bit from the long term accepted norm. Good to see that even the semi conductor community is being more progressive and
inclusive!
Love your channel. Will there be a part Two?
very good
Is the -9V good and +15V bad?
Did you look at the bottom of the board with the thermal camera?
The EEPROM had discolored coating surrounding it. I was thinking either moisture ingress or over-heating.
To be honest I only looked at the top of the board with the thermal camera as i whowed in the video. Thanks for the suggestion
Measure the inductance of the coils that supply the opamp. That could give you an indication if the supply is being pulled down?
These serial EEPROMs generally store only the settings and usage counters and stuff like that (this one is just a 4 kbit = 512 byte job), so chances are you could put in a blank one and the unit would behave like straight out of the factory. The channel Buy It Fix It has dealt with issues like these in several videos, have a look.
-9/+15 does in fact seem super sus. Maybe that was supposed to be +/-12 and there's a bad ground connection somewhere? Reminds me of that time when I was troubleshooting an Onkyo T-9900 tuner that was intermitttently acting very strange... it ultimately turned out that there were some bad solder joints on one of the copper bus bars used for grounding. Kind of a bear to find.
Yes i do suspect there is a gorund issue (the the +/- supply is not referenced to the same gorund as the 5V). Thanks for the advice towards some research on other YT channels, it is much appreciated
Is there maybe an extra load on the -15v line caused by a faulty regulator or transistor causing the low volts on it?
I did wonder that too but can't see it in the resistance readings to ground. TBH I could not even figure out where the +/- 15V? is generated on the pcb yet
Heya, a tutorial about IGBT's would be great as I have an all in 1 hybrid inverter which I shorted beyween PV+ and PV-. at 24:00 minutes in the pin function table I see "serial data clock and input could you explain what that is and/or do I mean serial. yes a shematic would be nice
That is weird maybe the negative voltage cap is faulty or something, or maybe an optocoupler issue becuase the other optocoupler maybe is regulating the negative side.
remove the ic and see if -15V is restored. Also, reading one of the comments below reminded me that i think i saw the position of the optocoupler where there is a slot in the board. This may play a role in regulation.
👍
You can understand why customer facing engineers change the whole board instead of fault-finding, The customer wants it up and running without having to pay for 20 hours diagnostics. Part 50p + labour £1000 vs part £200 + labour £50. If you have many with the same fault, then time invested in understanding the problem can offset repair time / cost.
I can agree with the pont the the customer wants it up and running... but invest some time in repairing all the faulty ones you aquire 'for free' and then you can sell them back to new customers for pretty ,much full cost and make fart more profit than buying and selling a new one? Suerly that is a better business model?
@@LearnElectronicsRepair He he he he, you said "fart" he he he he. That's my childish side coming out. Hope your AC friend can supply you with more boards to build your knowledge base up. Keep up the good work, love your vids.
@@LearnElectronicsRepair That is a new 'financial' term I am not familiar with.
Whats the point of fuses --- when so many other components fail ---- without blowing a fuse ??? ;)
Thanks again
18:40 - If it hadn’t made me so sick, that might have been funny. I don’t care _what_ those people do behind closed doors: just _don’t_ make me watch!
12:39 - Richard, wouldn’t it be sad if all you had here was a wiring connection issue. This board is throwing the same error as a faulty indoor board when it is not connected. There is as you said earlier, feedback communication from the indoor board on either or both the 6 pin and 8 pin terminals (TB2). There may be nothing wrong with the indoor board.
I'm not sure what you mean with your latter timed comment. What people, watching what? 😂
great video as always richard ... sorry cannot give any info on anything to help you im not technically minded just like watching to gain knowledge ...
If you like learning about technical things then I think that counts as technically minded!
Did you ever do a video on putting Mr Bleep inside the DMM?
Oops - fixed
video description updated to inlude the link
@@LearnElectronicsRepair thanks. Wow, a year ago, didn’t think you built it that long ago.
This was interesting, pretty odd that its -9V and +15V
Have used split rails with a different negative voltage if an opamp doesn't have the full voltage rating and you don't need a symmetrical swing.
Anyway the negative rail still need further investigation.
I would expect the main firmware in the the controller and maybe just configuration data in the eprom. Unless the chip is an fpga with an external configuration epom.
It seems that nowadays i create a directory and download the datasheets on all the components, and give them a quick once over as this aids the old grey cells.
Another reason for just swapping boards is reliability issues. It costs more to revisit customers and they get pissed off with recurring issues. Easer to just replace the board and employ a fitter instead of a engineer. A fitter will never get bored replacing boards, but an engineer will soon get fed up replacing fuses etc.
Its just a sad fact that we have been living in a throw away society for decades, i knew of a monitor company who just looked at the condition of the monitor to decide how much the warranty repair would cost, scrap the monitor and send them a new one and change for the repair. No expensive engineer, no reliability issues, quick turnarounds and happy customer's. I rest my csse, it was getting heavy anyway 😂😢
Turn off and on again is a reset. Its not to diagnose the fault.
"Strange device", are you for real. First of all it's obviously a three phase inverter, second, didn't you look at the drawing you took a photo of.1
No I didn't ever never seen one of these before therefore to me it is strange. Later it could become familiar.
maybe ask for a working spare from AirCon bloke
Or a broken one with a different fault and swap the EPROM.
I will ask him if he has one I can read the eprom from - assuming of my eprom programmers will do it
Being able to troubleshoot these boards could be lucrative but not practical for the industry. Difficult enough to find a real tech.
But very practical/profitable/lucrative for anyone buying/scrounging scrap boards for practically nothing and fixing them to resell at say 50-60% of new replacement cost don't you think? Or even more % of replacement cost if they are no longer available to by from the manufacturer? Whay not even sell them back to the 'industry' who think it is not worth fixing them? Surely there must be good profit in this... and from what I am learning so far at something I never tried to fix before it is not so difficult, only tackled three so far but fixed two including the one with the fried lizard on it and I am coming at this as an absolute beginner with Air Con.
now day the repair guide is to replace the PCB it self or throw the entire unit to trash and buy a new one typical technique .you did not get any info about the error code or what it mean .i missed the day when the device have the full schematic and datasheet available on the net ,even in the user manual it have some technical guide. Like what exactly the error code mean.i hate a company treat me like an idiot "unplug the device and plug it again " what a lame thing to explain an error.
I totally agree. how much do these things cost and how much is a replacement PCB that makes it the most economical repair option?
@@LearnElectronicsRepair it's more the time the system is down that really makes this the immediate action.
Also, many HVAC guys don't really do electronics. So it would have to be a call out or sent to an electronic tech...
And, as much as the boards are all the same, they're all different enough to make holding stock unviable.
Kind of like laptops etc...
@@joeds3775 How about buying/scrounging all the scrap ones, fixing as many as you can, and selling them back (on ebay or wherever) to the guys who can't be arsed to fix them for say 60% of the cost of a new replacement? Does that sound kinda profitable?
@@LearnElectronicsRepair definitely feasible but expect a letter from manufacturers... I used to work for a manufacturer... Spent a week fixing a room full of boards... Then did a presentation about faults found etc...lead balloon scenario. Even on out of stock items biig contracts were screaming for...It's definitely discouraged.
The client swapped all our stuff out for our market leading competitors, we were number 2...
I left. Saddened.
Has anyone tried bypassing the electronics and applying power to the compressor to make sure it can run. After all it is just a single phase motor
Compressor is not faulty so no point. The fault is the board not talking with the other.
It is a 3 phase motor supplied from the invertor.
I hate these mini-split systems.
You can't clean them propperly and the controls are overkill.
You can do everything (aside from the inverter part) with simple mechanical relays and cotactors and thermostats.
In my opinion, these are more disposable than a window unit.
There's much more to why you'd choose a mini split though! They're different from window units for good reasons.
@@AnnaVannieuwenhuyse I am an HVAC journeyman.
I know what they are. I would rather have a traditional split system than a mini split.
I only used a window unit as an example of quality and dosposability.