Cheap Inverter Checker

[Latte]

Evening All,

I Was in HRP Today (www.hrponline.co.uk) and picked up a leaflet. They are now selling a cheap inverter checker, Part no 253792 at £36 + Gordons bit

I Think the leads are extra, but the last time i enquired about these Daikin wanted £75 + Vat

May be of use to someone who wants one

Regards

Raymond

[nike123]

Is this something like this!
http://tinyurl.com/yntf3e

[The Viking]

Yepp, that's the one.

Got it from HRP last week, tried it once and it seems to be a nice little gizmo.
(before I attached it I already condemned the inverter the old fashioned way but so did the new checker, so at least we agreed)

[nike123]

Yepp, that's the one.

Got it from HRP last week, tried it once and it seems to be a nice little gizmo.
(before I attached it I already condemned the inverter the old fashioned way but so did the new checker, so at least we agreed)

Could you please explain your "old fashioned way".
Maybe that way could be of some help to me and others.

[The Viking]

Disconnect the compressor and measure the voltage from each "phase" supplied by the inverter down to earth/neutral.
The readings you get should be the same, or at least within 5-10% depending on which manufacturer you ask.
When you test a working inverter the readings could be something like 135-134-136 but if its faulty it's likely that one phase is far out like 135-134-71.

[nike123]

This is my way to. Sometimes, when I am unsure, I connect 3 light bulbs.

[paul_h]

I use diode test mode on my multimeter. The inverter has postive and negative in, and three phases out.
With the unit powered off and multimeter in diode test mode, put one lead on one of the inputs (solder point on the pcb, or flying lead into IPM chip) and put the other lead on each of the outputs. (again solder point on PCB, or flying lead to compressor plug).
Basically in one direction you'll get between 0.3-0.6v, the other direction you'll get open circuit.
Then you reverse your postive and negative multimeter leads. This proves all the diodes are OK, and thats how inverters make the three phases, diodes.
attached is some pictures to show what I mean. Shown is testing of the diodes starting with multimeter red on input terminal "P", checking the inverter output to the compressor with the multimeter black.
Then you reverse test, ie test with black on "P" terminal/plug, and multimeter red lead on all the outputs.

Not shown is: Then you put your red multimeter probe on the "n" input terminal, then run your multimeter black over each of the outputs, then reverse again. ie multimeter black on the 'n' terminal, and test each of the output pins with your red multimeter lead.
What you're looking for is any shorts, anything that reads 0v is stuffed.

[nike123]

Ok, that is test for diodes (actually diode part of Dual-In-Line Intelligent Power Module) , but inconclusive on inverter circuit functioning properly.

My point is that you need positive prove that that inverter PCB can do its job by lighting bulbs and with appropriate bulb power be sure that he can power up compressor.

If you do only diode test, it's still plenty other components in module or in PCB that can be faulty and you can't tell with 100% certainty that inverter is OK.

There is more than diodes in Power Module!

[paul_h]

Ok, that is test for diodes (actually diode part of Dual-In-Line Intelligent Power Module) , but inconclusive on inverter circuit functioning properly.

My point is that you need positive prove that that inverter PCB can do its job by lighting bulbs and with appropriate bulb power be sure that he can power up compressor.

If you do only diode test, it's still plenty other components in module or in PCB that can be faulty and you can't tell with 100% certainty that inverter is OK.

There is more than diodes in Power Module!

That may be true, but if you find a shorted diode, you don't need to go any further do you? You can say 100% that the IPM is bad. Also many units are a all one one PCB these days, if the IPM diodes are bad, you're going to have to replace everything anyway. The only thing seperate is the AFM and possibly main diode bridge, and they're just as simple to test without an inverter checker.

I've never needed to go any further than testing diodes, AFM and meggering the compressor.
There may be a easier better solution, but inverter checkers aren't available as far as I know in australia, and maybe other countries also.
Just putting the info out there for anyone without a checker like me.

edit: But I agree that all your testing with a diode test is the power circuit, not the controller circuit.
But I've been repairing these things since before these checkers came out anyway, haven't been wrong yet since I leant what to test and how.

I'd just thought I'd pass on some info for people not able to get a checker.
People are seriously better off buying a megger if they don't have one yet rather than buying a inverter tester. No point finding out the inverter is stuffed 100% with a checker, and then finding your PCB replacement stuffs up straight away because the compressor has a short which can't be picked up with a multimeter.

[Thermatech]

Following 15 years working on VRF systems in the uk I tend to aggree with Viking. The acid test for any inverter is if it can supply a ballanced voltage on each phase to the compressor.
So if the inverter is operatedwith the compressor disconnected, then between earth & each phase should be exact the same voltage. Any inballance will indicate some sort of inverter problem.
The next easiest step is to carry out the diode check described by Paul H.
This check can be carried out on TRM,s IPM,s , diode stack / rectifiers.
99 time out of 100 this will identify which components have failed because they most often fail open or closed circuit. Although the optimum tests should be carried out with analog meter in ohms mode the diode check with a digital multi meter will normally result in a correct diagnosis.
On a small split system with all components on one board you would normally at this point opt to replace the board.
On VRF systems alot of the components can be individually checked & replaced so you then need to check the capacitors , resistors to be sure they are ok.
The problem item can be the inverter control board & IPM driver board.
You will find that a number of the VRV/ VRF manufacturers recomend that the inverter control board should always be replaced when any major component like TRM or IMP has failed as a precuation because very often the inverter board will also have suffered some damage.
Manytimes over the years I have been to site when the engineer has replaced the IPM but still could not get compressor going becuse the inverter board was also damaged. And also many times when the inverter board had been replaced by engineer did not realise that the IPM was damaged.
Sometimes the inverter board is ok & can be used again but the the final test of this is to carry out the ballanced power supply test again before testing with the compressor reconnected.
The use of any inverter checker is therefore very limited beause it can only indicate that the inverter operation is correct or not correct. If the engineer has a multi meter then he can make that judgement without any inverter checker. If the enginner wants to identify which commonents in the circuit need to be replaced then the inverter checker is of no help.
Some VRF manufactures are making very good profit for inverter compenents & reconditioned inverters simply because mainy service engineers are unable to carry out the checking procedures to identify the failed compenents & therefore opt to replace the complete inverter, new or reconditioned.
Finally I have to agree with Paul H about megga meter. Too many engineers still do not have a megga & so cannot accurately test compressor windings.
The most common cause of inverter failure is compressor windings down to ground & 9 times out of 10 if the compressor has failed with windings ground fault then the inverter will also be damaged.

[nike123]

I'd just thought I'd pass on some info for people not able to get a checker.
People are seriously better off buying a megger if they don't have one yet rather than buying a inverter tester. No point finding out the inverter is stuffed 100% with a checker, and then finding your PCB replacement stuffs up straight away because the compressor has a short which can't be picked up with a multimeter.

Completely agree with you, only wanted to say that exists simple and cheap way for the testing of the complete faultlessness of inverter assembly.

[ozairman]

Let me throw my devalued $0.02 in here as someone who spends day in and day out talking to random techs on the phone helping them to fix inverter units. What Paulh says has a lot of merit as many times when it is a failure of the actual drive section of the inverter or whatever you call it means it has to be replaced. The kicker is that you also need to be savvy enough to make sure the compressor is not down to earth and has caused the failure or "poof" there goes your new inverter assembly when you have just fitted it and then work out the compressor is burnt out. Unfortunately most "technicians / engineers" are not that and are just parts changers. That is also why a lot of manufacturers say to change both, it eliminates the "poof" factor and an unhappy consumer

A combination of checking the diodes/transistors within the drive module and then checking the output voltages will do the same as an inverter checker. Except Viking I disagree on testing to earth or neutral as at that point in the system you are checking a quasi AC output fed from a high voltage DC supply so neutral or earth has very little to do with it, instead you should be checking between the phases of the inverter and looking at the min max values you get at each phase pair as the deciding factor. I believe Hendrag the LG guy recommends this as well

All of this can be done without an edc ipcm checker or daikins jobby which looks like a highschool electronics project.

But then there are a lot of people out there who don't even know what the diode test function on their multimeter is or does, I have actually asked people what model their meter is googled for a user manual and thens said turn the dial to the .... position You DO have Diode test on your meter !!! So for those techs an inverter checker box with flashing lights is a good idea.

As far as the scenario of a good drive module (IPM whatever you call it etc) with no shorted diodes/transistors but giving bad switching sequences to the compressor these are by far and above in the minority but search youtube for "inverter checker" as it does happen from time to time. This was an isolated case but good to show what it looks like compared to a good one.

Another point that nobody seems to take a lot of notice of and as very few people have a meter capable of accurately reading it is compressor winding resistances. Maximum resistance imbalance is about 5% but try measuring 5% of 0.188ohms to any degree of accuracy without a milliohm meter

And for any aussie fridgies the edc-ipcm units are available in australia!

PS I second the recommendation on a Megger why with the advent of digital multimeters does everyone seem to think that a high voltage insulation resistance tester is no longer needed???? buy an old windup jobby off ebay if you need they are not that dear and least you don't have to worry about flat batteries.

[TRASH101]

The megger is good for checking apprentices too

[The Viking]

Or waking old farts up.

[nike123]

What bulbs are you using? watts??? i think your idea is very good as the bulbs are something safe and act as a fuse .No damage can happen to the pcb.

Depends on compressor power. I try to match power of bulbs with that of compressor. If compressor electrical power is , let say, 1500 Wats then I use 3 bulbs of 500Wat.