Oh Blimey I bought a CNC Lathe !!!!

[awemawson]

OK I had a few minutes so escaped briefly to the workshop for more investigation. More close scrutiny of the board revealed an overheated track and a blown capacitor (35v 68uF) with one leg burnt off! How I didn't spot it before I don't know. Bit more tracing shows that the overheated track, although it goes very close to the blown up capacitor, isn't connected to it! In fact this track snakes around ending up on the output of a 7915 voltage regulator a few inches down the board. I decided to take my chances and, having replaced the blown capacitor (with a 35v 100uF as that's the nearest I had to hand), returned the transformer to the board, and linked the originally found blown track, connected up my ancient (probably lethal) open framed Variac, set it to half normal input voltage of 100, and powered up. No smoke, no smells, nothing got hot. Slowly wound it up to 200v and again no dramas. Gingerly started probing around (everything is floating at a high voltage). That 5v Zener diode in the hand drawn picture a few posts ago actually had 5v across it, and from my reading of the circuit that can only happen if the thing is oscillating and driving the transformer. Somewhere down the line there are two 'standard' voltages regulators, a 7815 (+15v) and a 7915 (-15v whose output track had overheated but not blown) - neither are giving outputs so something is still wrong. I need to poke around with my oscilloscope, but need first to work out what I can use as a common earth - a job for tomorrow.

Incidentally, with a 200v input, the large reservoir capacitor is charged up to 295V

Some pictures:

A/ Second burnt track and adjacent blown capacitor
B/ Capacitor removed
C/ Original blown track linked
D/ Lethal Variac - don't try this at home children 

[awemawson]

Looks increasingly as though the hybrid ic in this power supply design is the fault. It is a DK-466 and it seems that they are rare as hens teeth - only found two references to them by Googling, one was a chap looking for one, and the other is this:

http://mtc-technik.de/online-shop/mitsubishi-electric-hybrid-ic-dk-466-power-supply::7255ID.html

I've emailed them but don't hold out much hope.

The oscillator is running at about 20-30 khz which seems reasonable, and there is a power fet in series with one winding of the transformer (lets assume it's the primary,) fet and winding are across the 270v dc input voltage -  but the fet is only turned on for a very short time, so it's not generating much power. Looks as though it is supposed to generate 5v at a significant current, and +/- 15v at somewhat less . Belling out the circuit round the fet and the unknown guts of the DK-466 is very confusing as it's a three layer board some of which is covered in heavy conformal coating but I think that one winding is generating a reference voltage with a zener diode, and another winding feeds the DK-466 as feedback - it's this I assume isn't working so the 'on' time of the fet isn't being tweaked.

[techonehundred]

This may (or may not) not be help, but I found this link also.
http://www.antechenterprise.com/Stock.asp?Page=384

[lordedmond]

Andrew

maybe of help
http://www.aliexpress.com/store/product/Special-promotions-DK-466-module-In-Stock-welcome-to-buy/217335_810050418.html


Stuart

[awemawson]

Wow chaps thanks you've cheered me up no end 

I suspect I'm going to have to pull the transformer AGAIN and also the DK-466 as everything round there is such low resistance that it make tracing the circuit neigh on impossible. I'm glad I got that SOIC unsoldering kit - not used it for SOIC's yet but it makes removing things like the transformer relatively easy with the very low melting point solder. I assume it's bismuth based like woods metal.

Hope to have another session going cross eyed circuit tracing after super ...

[awemawson]

OK NOW I'm confused but in quite a good way 

It seems that the transformer I removed, previously, one coil of which is definitely driven by the FET off the rectified mains, also drives the other two transformers on the board, EACH of which provide four isolated + & - supplies. The first transformer also has one winding giving the 5v supply, another giving about 25 - 0 - 25 which goes to two 'conventional' series regulators, a 7815 (+15v) and a 7915 (-15v) both of which I had removed as at least the -15v one had blown up due to a blown capacitor on it's output. Tonight I decided to load the 5v supply with a 0.5 amp load, expecting it to drop like a stone. Not a bit of it - rock steady at 5v bang on to the nearest 10mV. This lead me to replace the 7815 (as I had one in stock), but to my surprise it only gave 7.8v unloaded, meanwhile the 5v supply was motoring along getting my load across it nice and warm !

So a few tentative conclusions:

A/ The DK-466 may not be faulty after all
B/ Either something down stream of the 7815 is heavily loading it, or there is something odd about the rectification from the transformer on this winding. Tomorrow I'll pull the 7815 out again (lucky I have a drawer full of them!) and load it's input independent of it's output circuitry and see what's amiss.

anyway it's sort of progress

[mattinker]

It would seem you know something about electronics!

Best of luck, keep on doing the incantations and you should get the magic white smoke in the right places!

Regards, Matthew

[vtsteam]

Yes, haven't commented in a bit, but best wishes, still following right along. 

[Fred Bloggs]

Andrew

Just another thought but does the machine have any more of the same pcbs on the other axises, if yes why dont you set another one up on the bench and have a careful probe around with the scope to get an idea of how its working.

Best Regards

Fred

[awemawson]

Fred: yes it does, and that's a good idea, though it's an absolute pig to remove it.

OK results of a bit more poking around: Firstly see the very simple circuit I'm working with:

[Pete.]

I would say that one of the two little springs on the left have broke

[awemawson]

Now apart from the fact this is running at about 25 kHz there's nothing complicated there is there ? The positive output feeds the input to a 7815 regulator (+15v) , the negative output feeds the input to a 7915 regulator (+15). Both via a 'choke' winding on T2. Each rail has an electrolytic capacitor of 68uF and a 1 nF polyester capacitor from rail to common. Now bear in mind BOTH regulators have been removed for testing. Also bear in mind that these regulators are each rated at one amp, and are mounted on small upright heat sinks, so presumably are expected to be delivering at least a 100mA or so.

Right, so there is no load on either of the outputs as the regulator is removed - voltage from common to+ve is about +42v, and to -ve is about -35v so in the ball park of where you'd expect. If I load either of them with a 22k resistor, so only drawing perhaps 2 mA, the voltage drops to 12volts

High resistance transformer winding perhaps? No it measures virtually zero ohms. Dodgy diodes perhaps. Well very possibly. They only bear the marking "4L" with the cathode marked with a blue band. They may be some exotic species - quite possibly schottky  fast recovery ones? I tried replacing them with 1N4007's from stock but only got about 12v out, then I tried 1N5819's which are the only schottky ones I have in stock, with the same result.

So what the heck is a '4L' diode?  Poor things have been in and out of the board four times now.

Here is a picture of the little darlings, out and then back in the board.

By the way I had the second transformer out of the board to trace it's windings and it seems it's being used as a choke, three windings, each in series with a rectified transformer output.

[awemawson]

I should add that I have, with the power off, applied a lab supply to the +ve output of the circuit, and wound it up on minimum current limiting (20 ma) up to 20 volts. Repeated this on the -ve output, and also done it on the outputs from the regulators up to 15v (they are not in circuit), and am able to raise each rail to the set voltage drawing virtually nothing from the lab supply, so nothing is breaking down (other than perhaps me!) and limiting the voltage.

I remain confused, though it may be those diodes if I can find what the 'ell a 4L is   

[lordedmond]

what does the volts say with a scope  instead of the DVM


evan my mk8 played merry hell with the voltage reading that had a say 20khz wave form , bear in mind I used to mess around with 432 mhz and 1.3 gig   1kw of the former and 100w on a good day with 1.3 gig my tuned cavity seem to dislke moisture in the air


had a quick look in my ancient data book no joy I am afraid they are not quite on velum but very near


Stuart

[awemawson]

I've only been measuring DC levels as quoted - my Avo & fluke both run out of steam at these frequencies and the 'Scope of course raises issues of the circuit 'floating' at a few hundred volts.

I did try measuring the ac output of the transformer whilst the diodes were removed, but it gets increasingly hard to interpret the results - I gave up when the 'scope  (Tektronix 465) internal DC fuse blew 

[lordedmond]

I have not said this and is not to be done under any circumstances 

what we used to do when using a scope on floating ccts was to use marigolds ( proper kite marked ones ) and disconnect the earth on the scope 

we did the same thing with our soldering irons ( before we got the isolated weller ones )


if you even think of doing the above take care


I guess from my posts I am of the old school sparks before H&S got into there swing , we had to sort it at the iron works it was 75 tons of molten CI at the bank it was the down time and pride

Stuart

[awemawson]

No doubt that was Stanton & Staveley ?

Yes we used to have the earth wire deliberately and very obviously poking out of the top of the 13a plug top so we could float the chassis of the 'scope when I worked for Ferranti all those years ago. I don't think that this fuse blowing was anything other than old age - it just happened when I turned the 'scope on and it was not connected to anything at the time.

[lordedmond]

 10/10 for your deduction powers

Stuart

[Fred Bloggs]

Andrew
if you are struggling with using the scope with a normal probe due to floating voltages if have got a pico battery powered differetial scope probe if you want to borrow it.
Its alot easier to loan that than my 1kva isloation transformer ( she weighs4or5kilos).
If you do want to borrow it pm me and ill drop in the post on saturday morning .im not using it at the moment, and the post will only be a couple of quid each way, you can have it till the end of may it it will help.

Best regards
Fred

ps trying to type from an android touch screen is not eay sorry about lack of capitals etc

[awemawson]

Fred that is a very generous offer - thank-you. Let me hold that option in reserve - I'm bracing myself to apply your other suggestion of a side by side comparison of the other unit.

[awemawson]

  Well I MAY have been chasing Red Herrings 

I was doing some tests with a load on the 5v rail to see the effect on the power V-Fet that is the main switching element, and sure enough with a load on. the 'On Time' increases (as does the frequency). Then it occurred to me, suppose that unloaded (as I've previously been testing), insufficient current is passing for the + & - 15 volt rails to be established, and the feedback loop is only monitoring the 5v rail? So I put a moderate load on the +15v rail (or rather the feed to the input of the 7815 which is out of circuit), and spun it up still with a load on the 5v rail - wohay - yes - volts! So on a roll I put the 7815 back in circuit, and we are developing 15.1 volts across the +15v rail. Can't do the -15 7915 as the postman hasn't brought them yet.

... this may be positive progress, but I'm keeping my fingers crossed 

[Fred Bloggs]

Andrew

Glad to see that progress is being made 

Switch mode psu's are, and always will be a pain in the rear 

In the mid 90's I used to work for for a major British engineering company that produced power control equipment, One of my jobs their was to look after a production line that built approx 50,000 switch mode psu for for their products, I was responsible for the custom and COT's test equipment and supporting the shop floor staff and I spent upto 20 hours a week looking for obscure faults ( the designers where never interested as it was current product, they only interested in the new toys)

Nearly all of the shop floor staff where ladies - hundreds of them!! but that is another story.

Fred

[DMIOM]

Fred,
your message just brought back a memory.

Aeons ago (before PCs, in fact I think it was before CP/M and WordStar) I was responsible for a dedicated word processing system made by "Xionics". One of the secretaries had endless trouble with her workstation hanging. We gave her several replacement workstations, she crashed them all; workstations that crashed for her worked OK for the other staff; and if she went to someone else's desk she would crash their workstation. 

We joked that it was due to her 'magnetic personality, but one of my colleagues then discovered it was due to a combination of silk and satin 

Dave

[SemiSkilled]

but one of my colleagues then discovered it was due to a combination of silk and satin 

Dave

Ah yes, research is a wonderful thing   


Lee

[awemawson]

  RING THE CHURCH BELLS 
  Let everyone have Monday as a Holiday

Postie brought the 7915 today, I rapidly installed it and - plugging the logic card and option card into the intermediate power board (which had the fault) ...whoohay :  we get "A6" which is "OK I'm here, now talk to me"

Pictures:

A/ Logic card loaded on, powered up showing "A6"
B/ Overview of same
C/ Re-assembled on high power chassis and still "A6"