Testing a Printed circuit Board

[ke5frf]

I was just about to make the point that Frank has made here.
I didn't notice in this thread if anyone admonished you to check the diodes, triac, and resistor resistance by lifting leads off the board. I noticed you gave diode ohmic measurements and said the diodes read "000" either way. Well, even in circuit this shouldn't be the case but I never trust "in circuit" resistance readings. Been burned too many times by overlooked parallel components!

Honesly, the best way to troubleshoot a circuit like this is on an AC voltmeter point-to-point with voltage applied but I understand if you aren't familiar with electronics that this wouldn't be the time to start learning

Just a hint if you are willing to try measuring the operation of the circuit with power applied, if there is an open circuit, chances are you will read the full AC supply voltage dropped across the faulty component. That would be the voltage at your wall outlet unless there is an external transformer somewhere that we aren't aware of. And all of this assuming the reverse-engineered schematic that someone posted was correct.

 

[ke5frf]

I have a hard time believing all of the bridge diodes are shorted. If these were read in circuit, and the motor leads were still connected the the motor, I'm suspicious that you may have been reading the motor windings some sort of way instead of the diodes. Just a hunch. I don't think I have ever tested a motor circuit with the rectifier right before the motor like that, but I have with a transformer secondary which is basically the same thing when the circuit is deengergized.

 

[ke1000]

Hello All

Thank you fro the latest answers I can see now how Little I know and yes I live in the North West of England in Southport about 15 miles north of LIverpool. The Mill is the Proxxon MF70 and yes it's really small, a table top size. Since my last post I have desoldered all the mains voltage cables from the board, i.e, neutral, negative to the motor, positive to the motor and the supply from the isolating switch, so that I could remove the board from the mill to make is easier to remove the diodes. I found out like you said that the desolder braid is hard to use only being able to get the solder up the first 1/16 of an inch. I have phptographed and sketched everythng, which is why it has taken so long. I have been unable to remove the diodes as at first I ahd nothing small enough to hold the body and pull when I melted the solder. Now I have a pair of tweezers but although I have straightened the ends and melted the solder the first one refuses to come out of the hole.
Do you think I am in over my head and I should contact Proxxon to buy a new board?

Regards and Thanks, I will re-read the latest posts to see if I can get to understand.

Bruce Davis

 

[sheldonstv]

o my- it all comes down to practise when you are using desolder braid i have never had a problem with using the stuff ...........

 

[#1supertech]

ke1000,

I didn't get a notice that you posted a follow-up to the thread so that is why I'm replying kinda late here.

I see you have the PCB out so all you have to do is now re-measure the diodes and they should show that they are indeed all good, or at least I would think so.

No need to desolder the bridge rect diodes if the board is out of the circuit.

Forget that solder wick (braid) crap - GET A SOLDER SUCKER device either from online or at your local electronics store there near you. I'm sure you have one - right?

I'm thinking that your country has that LEAD FREE standard as well and that you have NO LEAD in your solder there??

Also - you need some solder acid flux #30 - such as that like Superior Flux Mfg Co sells. This will aid in your soldering iron tip (hopefully it's cleaned & tinned properly first) being able to flow the solder easier to start with. All solder joints have some form of surface oxidation (impurities) the older the board is, and thus make it even harder to desolder! That or it is LEAD FREE and that makes it harder as well. Here in the USA we use a 60/40 solder and it flows very easily! Hopefully your PCB there used lead in the solder (but I doubt it) otherwise that too will make it harder to desolder!

Do that diode test again, and repost an update for us.

Thx,

Frank

 

[#1supertech]

PS

Forgot to ask you - What kind of DVM are you using to test the diodes, etc.

What brand name and model that is??

Frank

 

[ke1000]

Hello

Thanks for all the help. In answer to a previous post the "bang" came from the area of the PCB and the switch. I won't desolder the diodes and will look to get a solder sucker. Up to now I have onl;y used soft solder for the wiring on Motorbikes and cars.The solder I use contains lead it is 1.6mm cored. The mill is about 5 years old made in Germany, I don't know about the solder on the PCB. The Tester I have is a Caltek CM1505. What results should I be getting, testing the diodes in situ?.

I can't always get access to the computer as my wife is doing a course and uses it for assignments and research on the internet, so please bear with me if I don't answer right away.

Regards and Thanks again

Bruce Davis

 

[#1supertech]

Hello Bruce,

Just read your reply, and I did a retake (relook) at not only your prior posts, and a few others including what (unclejed613) noted earlier (in #4 post) about the BURNT spot on that TRIAC, but I also reloaded the pix you posted, and it really does look like something may have shorted out the components in ? here! Esp that (TRIAC ?) as by the burn mark on the top of the heat sink mounting pad - whatever it's p/n is??

(unclejed613) posted in #4 that your PCB circuit there is similar to, or basically a dimmer circuit feeding a bridge rectifier, which then feeds the motor, and he's pretty much correct in saying that. The PCB motor control being pretty simple in design and function at that.

What probably will be harder for you to do is to desolder all the bad parts, and then successfully replace & resolder the good parts - after you find exact replacements for them.

The TRIAC (TR1), and DIAC (D5) will be the hardest to find no doubt.

I also missed that/the PDF file drawing you made and posted, so I just now got to look at that as well.

Btw - your DVM (or DMM if you like) is a pretty good model from what I noted online and the overall accuracy isn't too bad either.

Seeings you DID say the PCB is now out of the Mill, and there are NO other external connections to it - then it should now be safe to do a accurate test measurement of the components on it.

1st - To check the DIODES you just put the meter select in the DIODE test mode (read your operators manual as it will show you how to do it if you are unsure).

Then in the forward bias config you should get a reading of .600 (volts), or very close to that across each diode in test - if they are indeed good. In some cases the diodes may read a tad lower like a .525 or maybe even lower at say .475, and though they may look bad that may not be a bad diode esp if it's a higher wattage rectifier. Just depends on the Mfr and its specs! If all 4 diodes read pretty equal to each other then chances are that they are all good. If you see all 000's across any 1 diode, or all 4 in ? then you indeed have shorted out (blown) diodes there, and all 4 need to be replaced!!!

Next - Test all 4 diodes in the reverse bias mode, and you should now see only a "1" displayed on the meter across each diode being tested. If you see anything else displayed - including all 000's - then for sure all the diodes are bad as in shorted (blown out), and again - all 4 need to be replaced.

I may have spoken in haste - as to what I said earlier about them being good, as now in looking at that burnt TRIAC there - I'm no longer sure that may be the case, and that what you measured before may indeed be showing the beginning to the end of all the related diode-junctioned components being blown out!! Namely D1 thru D4 & D5 as well as TR1.

As to that TR1 there make sure you carefully remove it as well as there should be a component marking on it - regardless if the PCB Mfr took "blackout" and covered over it just to hide that part. Sometimes they do that just to get you to buy a whole new PCB from them.

If they indeed black painted over the p/n on that TR1 part try and first remove the paint (lacquer) very carefully using mild (non-acetone) finger nail polish remover first. If that doesn't work try using a razor blade flush to the surface and shave off the black paint/lacquer until you can read the p/n. If that fails to work then we can try to match the part to one that is close in SPECS. Usually (hopefully) the motor will have some ratings on it, so that will help narrow it down a bit.

The PCB shows the correct insertion (polarity or bias) of each diode, so make sure when you R&R the bad ones with new ones that you get each one right - as to it's proper junction placement - otherwise you will blow everything out once again!!

Next - In checking the resistors they will have to have at least one pin removed from the board to be measured correctly. The POT will need to be totally removed to be measured properly! If any one resistor is out of spec (including the POT) then replace the bad resistor with the same identical valued resistor AND NO OTHER!! Must be exact tolerance % replacement as well! Wattage can be a bit higher, but by looking at the pix it doesn't appear there was any prior HEAT damage to the PCB so going with the same wattage rating should be just fine!

Be careful when removing that POT - so as not to disturb the preset wiper setting position, and just measure across the 2 outside pins, as that looks like to be 500K ohm pot as by what you had on the PDF drawing? I can't see what's after the 2nd 0 as that C2 CAP is in the way on the pix you posted, but if there is a K then it's a 500,000 ohm POT. R2 should read 56K ohms 2% tolerance, and R1 should read 26K ohms 2% tolerance if I read your color bands posted right?

Once you've established that the POT is indeed good (hopefully is) then measure across the center wiper pin to either outside pin, and next mark down on a piece of paper what that value is - JUST FOR REFERENCE - in case the wiper arm setting should get disturbed. Be careful when desoldering & handling the POT so that doesn't happen!

You may also want to pick up a small pair of used surgical type locking hemostats (if you can find them used) so you can better snap onto the leads of the components when desoldering them, as they will act as not only a heatsink to protect any "good" components, but also will aid in lifting off the pin or lead of whatever you are trying to desolder as well. Tweezers are rather useless for doing this kind of repair work!

As for the 3 CAPS on the board - they too will all need to be taken out and measured for proper value. If they have been compromised - esp C2 & C3 (possibly being shorted out) then those too will have to be exactly matched up and replaced as well. I doubt C1 was hurt - as it is across the incoming AC line, but best to check it as well. Never know as it too may have shorted out as well.

TR1 & D5 will definitely need to be replaced for reasons above! No getting around that part.

Btw - that 1N6007 I was seeing on your drawing of D1 thru D4 comes up as a ZENER DIODE on Microsemi's data sheet - which seems a bit strange unless you got those numbers wrong, and they are indeed a 1N4007 instead??

Please try and double check those numbers again and post the results if different - thx!

Btw - how many hours would you guess you have on this Mill?

Lastly - Is it possible for metal shavings to get into the area where the PCB is located? As related to the TR1 part that is.

Best regards,

Frank

 

[ke1000]

My PCB

Hello

Thanks for the replies. I have printed them off as there is a lot to take in. I will get back on tonight (my wife is at her course) with answers I hope.

Regards and Thanks

Bruce Davis

 

[ke1000]

Hell All

To answer a question I missed before I have used the mill for about 50 hours, but it was second hand when I got it. I have mainly used it for drilling small holes(0.5mm - 2mm) in mild steel, I am making a small traction engine called sweet sixteen.

I have read the responses and have re-tested the diodes so far. On the diode setting of the multimeter, I get 000.6 on all of them, even when I reverse the leads, which I hope is what is meant by the reverse bias test.

I have "tested" the motor using a 12V car battery charger and it turns over, albeit slowly. According to the Proxxon agent in the UK, brushes are not available. If the motor fails you buy a new one (£30+) good way to make money.

Regards and Thanks

Bruce Davis

 

[unclejed613]

i don't know exactly what your rectified voltage should be, but using the voltage marking (63V) on C2 as a guide means it's probably about 50V (48V has become an industry standard for industrial controls). so the motor turning slowly at 12V isn't a surprise, but you're right, it turns, aso there's no immediate reason to suspect it has a problem. you probably had a chip work it's way into the controller box and short between the case and the triac. when you put the unit back together, it might be a good idea to stretch a piece of nylon stocking over any openings or vents to keep metal chips from getting inside, but still provide airflow. back to troubleshooting... the diodes look like 1N4004. you can use any diode from the 1N400x series as long as the last digit is equal or higher to that of the original (i.e. 1N4004, 4005, 4006, or 4007) ,so which ever part is most readily available you can use. the differences between them are maximum voltage (PIV) ratings (4001=50V, 4002=100, 4003=200 4004=400, 4005=600, 4006=800, and 4007=1000), which ideally should be twice the highest voltage expected to be encountered in the circuit. so at 50V or so RMS input= 50*1.414=70.7V peak. 70.7Vp *2= 141.4Vp-p. doubling that we get 282V. the next highest step up from this estimate is the 400V rating of the 1N4004. so if you can't find 1N4004s, but can find 1N4007s, they will work. i would check the diodes out of circuit, as well as testing the capacitors out of circuit. lifting one end of the component is sufficient to get it out of circuit. if the triac is shorted, chances are that the diac is also. diacs can take very short current spikes when they hit breakover, but won't handle extended overcurrent. once you get repairs completed, the best test before connecting the motor would be to connect a resistive load where the motor would go. a 25 watt 120V light bulb would be perfect, since you will see the response when you turn the pot. before you start, mark the position that the pot is at with a permanent marker. since the circuit is very similar to a lamp dimmer, you should be able to control the brightness of the bulb with the pot and see it's response.

solder sucker or wick? i use both. the sucker is good for double sided boards with small traces, and the wick (or larger sucker) is generally better for single sided boards with larger traces. both of them take a bit of practice to use properly. for a sucker on this board i would use a large sucker. the larger the "swept" volume of a solder sucker, the better. also, i usually use the soldering iron (preferrably a temperature controlled one) at a higher temperature than i use for soldering the components in place, since the higher temp reduces "time on target" and makes lifting a trace less likely (it sounds counterintuitive, but it works... too low a temperature makes you hold the iron on the pad longer, and is more likely to melt the glue that holds the trace down). also try to avoid touching the iron with the tip of the sucker. the teflon will erode the surface of the tip. i heat the solder with the iron and get it all melted (it helps to add a little bit of solder to improve heat transfer from the tip to the work), align the sucker tip slightly above the work, and in a quick smooth motion, pull the iron away-plop the sucker tip down on the work-and push the release on the sucker all pretty much at the same time. with practice, it moves pretty quickly and smoothly. also make sure the board is held in a stable position, such as with a PC board vise. it's not easy to desolder a board that is in motion.

 

[#1supertech]

Bruce,

Not sure what your DVM/DMM is set to (other then the Diode check position) seeings you have an autoranging type, but double check the "function" buttons, and make sure you take it OFF the Autoranging feature - IF you can.

That would possibly explain why you are seeing 000.6 across all 4 diodes even with reversed leads, and in seeing that .6 part of that diode test reading - it's not reading a .0 because that is the Accuracy error % built into your meter there. A better quality meter would or should read all 0's if the/a diode is indeed shorted out.

A way to verify that part is to put the meter in that diode test mode, and then short the test leads together and see if you get that same 000.6 display reading. You should, and if you indeed DO then all 4 diodes are definitely shorted.

At that point verifying that I would just totally rebuild the PCB with all new components less the 3 CAPS as well - esp if you have no way to test out the 3 CAPS with a separate CAP meter like I have. CAPS aren't that expensive anyway. Chances are the resistors are alright depending on how much shorted SURGE current went through them when TR1 got shorted like it did. The POT as well.

You'll still need to remove all the components and test each one out, and follow the caution on the POT R&R that I mentioned prior.

The motor sounds like it is OK - hopefully you didn't compromise the braided copper wire contacts on the brushes when TR1 shorted out. Like with all permanent magnet brush type motors - replacement brushes can be gotten a hold of very easily at any automotive type repair shop, or any small power tool repair shop that rebuilds DC motors. Taking one brush out - if you think they are worn out or bad, and matching it to a replacement takes but a few minutes, as I've done it many a times repairing people's Treadmills as well as numerous Medical related equipment when I worked as a BMET in a major hospital system.

Having said that let us know how it works when you totally rebuild the PCB and get it running again.

Frank

 

[ke1000]

My PCB

Hello Again

Thank you for your patience and expertise. I have my weekend sorted out by the look of the responses. I will print them off and get reading.

Regards

Bruce Davis