Super unreliable tape drive PSU

[pentium]

Ever since I got my desktop Cipher 9-track tape drive it's never worked. The thong would normally just sit emitting a low buzz and eventually the front panel would light up but diagnostics would fail. If you remove the PSU out of the unit you can hear the fan runs faster and the buzz gets a bit less angry but if you leave even the power supply alone for long periods and try to use it again, it will be a dead brick, only emitting the odd volt or two and buzzing from the primary transformer when you flip the switch. It will also of course spring to life if you leave it on for a while.

Here is the unit itself.






The only physical damage I found was browning fo the PCB around the joints on four resistors and diodes. The components by the way checked out as OK when removed and checked. All but two caps (and the two primaries) were in stock at the local electronics shop so it's also been mostly recapped too.
I'm out of ideas on what else might be at fault and I can't find a schematic or even a pinout. I have been wanting to get this drive working for almost a full year now.

[kaboom]

Re: Super unreliable tape drive PSU

When you say "buzzing from the primary transformer when you flip the switch," I assume you mean the SMPS transformer on the PWM board.

What kind of noise(s) does the boost inductor make?

I see there are three wires from the boosted supply to the PWM board. If there were only two, I'd tell you to check the boosted supply for the usual 360 - 420 V DC. But those three wires...

Are those two big caps connected in series? Of the three wires (red, blue, yellow), does one go to the top of the series connected group, one to the bottom, and the last to the center point of the two caps?

And is there any surface mount stuff on the back of the board?

When does that relay click? Right when you hit the switch or only after it suddenly comes up?

[b700029]

Re: Super unreliable tape drive PSU

It looks like a pretty common UC3844-based design with a few rails (and an odd +38V one.) The primary including switchers is on the lower half while the main transformer and secondary is on the upper half. From the pictures it looks like the 3 thick wires go to the primary winding on the transformer in the upper half.

[pentium]

Re: Super unreliable tape drive PSU

When you say "buzzing from the primary transformer when you flip the switch," I assume you mean the SMPS transformer on the PWM board.

The one closest to the mains input with the yellow tape wrap one coils.
It gives off a low and faint hum.
One of the two large transformers on the other board (you called that the PWM board, right?) however will give off a rather angry buzz when the PSU is connected to the tape drive and under load.
The yellow wire on the three wire harness connecting the the two halves is also the only one giving a voltage. Only 23.6v.There are no surface mount parts on the flipside. Those two major caps are mos likely the filter caps.
The relay never clicks when the main switch is flipped. At least, I never hear it.
It does however give a faint click in and out a few seconds AFTER you switch the PSU off.

[kaboom]

Re: Super unreliable tape drive PSU

The one closest to the mains input with the yellow tape wrap one coils.
It gives off a low and faint hum.

If you mean LF2, that may just be a common mode choke. The AC line should pass right through it. Trace it if you want.

One of the two large transformers on the other board (you called that the PWM board, right?) however will give off a rather angry buzz when the PSU is connected to the tape drive and under load.

T1 is the main transformer, L3 is the output inductor.

I made a mistake- I ASSumed there was a PFC booster. Clearly, there is not. Sorry for the confusion. Looks like that first board is the AC input/filtering, PWM/forward converter commutation, and control. Second board is TX/output inductor, secondary rectifiers, filter caps/chokes, maybe some 78xx regulators, dummy loads, and part of the feedback.

The yellow wire on the three wire harness connecting the the two halves is also the only one giving a voltage. Only 23.6v.

23.6V to where? Where did you connect your other test lead? That looks like the primary of the transformer- high freq. AC that your meter may not read correctly. Look around/trace near those MOSFETs- you should be able to figure out how/where the actual PWM gets its raw DC. If it is in fact a single-forward converter, the three leads are:

End 1 of primary
End 2 of primary
End 2 of "flux reset" coil
The other end of the "reset" coil will be shared with one of the primary leads.

If there's no "reset" coil, the third lead may just be an electrostatic shield in the transformer.

There are no surface mount parts on the flipside. Those two major caps are mos likely the filter caps.
The relay never clicks when the main switch is flipped. At least, I never hear it.
It does however give a faint click in and out a few seconds AFTER you switch the PSU off.

Check those two big thermistors for continuity. Check for equal voltages across the two big caps- you'll have to run it with the boards out. If the rectifier is good, each big cap will have the same voltage across it. The voltage across the two in series will be the sum of both.

Many years ago, I got what must've been an ancient external QIC or Ditto drive. It must've been junked because it no longer had power. I "played" with its power supply, an open frame Eltek, 5/12V, 4A each. I never did get it to work, but noticed that only one of the big electrolytics ever charged up. At the time, it stumpted me, since I thought it should do something with one cap charging. But now I know that its LVCO was probably locking out. After unplugging, one cap would give a satisfying bang when shorted. The other one, zip. If I still had it, the first thing I'd try would be a new rectifier.

There appears to be a seperate small power supply, too. Are you sure the three MOSFETs on the board with the big caps are OK?

[pentium]

Re: Super unreliable tape drive PSU

23.6V to where? Where did you connect your other test lead?

To chassis ground.

Check those two big thermistors for continuity.

Which ones? There's a few.

Check for equal voltages across the two big caps- you'll have to run it with the boards out. If the rectifier is good, each big cap will have the same voltage across it. The voltage across the two in series will be the sum of both.

So just poke a DMM at it and flip the scale to 200v? Should I do the test while the board has power or just after power is removed?

[pentium]

Re: Super unreliable tape drive PSU

BuMP.

[kaboom]

Re: Super unreliable tape drive PSU

Which ones? There's a few.

RT1 & RT2, they are black with nylon "bases."

So just poke a DMM at it and flip the scale to 200v? Should I do the test while the board has power or just after power is removed?

With the power on, across each cap on the 200V scale. You'll have to remove and support the board(s) somehow. Try some of those polyethylene/polypropylene cutting boards. Even tupperware will do!

Since you'll be floating your meter, either use a good one or do the following:
Solder short heavy wires to the cap terminals; use alligator clips to connect test leads to wires from the caps; keep hands off meter (if it's an 830.)

With 120V going in, you should have 150-170V across each cap, 300-340 across both in series.

[pentium]

Re: Super unreliable tape drive PSU

Sorry for the wait.

RT1 and RT2 both read about 6 ohms in circuit.
The caps on the other hand....give me 79.8v each and 159v in series. Something is very, very wrong.

[kaboom]

Re: Super unreliable tape drive PSU

The caps on the other hand....give me 79.8v each and 159v in series. Something is very, very wrong.

Sounds like you have only rectification taking place. If it were doubling, you'd have ~320/~160 (series/singly, respectively) across those caps.

On that relay, can you tell if there are a set of contacts that close, between the midpoint of the two caps and one side of the line?

Look carefully at the "120/240" switch in the diagram. It's (supposed to be ) open on 240 and closed for 120. If that relay takes the place of the 120/240 switch on your supply, closing when a comparator drives a transistor supplying the relay coil, and it never closes, your bus voltage will be one-half of 'normal.'

At this point, I don't think the bridge rectifier is suspect. Both caps charge, and "equally." Equally in quotes since you don't really have a load on it, but we can ASSume the bridge is good- one open diode and one of the caps wouldn't charge, but one shorted diode would blow the fuse (AC into an electro.)

If I had it here, I'd trace those contacts from the relay out, and if no 120/240 switch, hard-wire it for 'doubler' action. It would then, obviously, be set for 120 only.

-Paul

[pentium]

Re: Super unreliable tape drive PSU

There is no physical switch or jumper that sets the voltage range between 100-120v and 200-240v. The relay itself (RKS-5DG-48) I can't find a datasheet for but there's an indication that it's ON/OFF not EITHER/OR on the side of the unit.

[kaboom]

Re: Super unreliable tape drive PSU

There is no physical switch or jumper that sets the voltage range between 100-120v and 200-240v.

The relay contacts close and set turn the rectifier up as a doubler for 120V.

The relay itself (RKS-5DG-48) I can't find a datasheet for

You don't need one! There are probably four terminals/traces from the relay- two small ones for the coil and two larger ones for the point and (probably) normally open contact. Trace it carefully and I'll bet one goes to the midpoint of the two caps and the other goes to either hot or neutral, after your EMI chokes.

but there's an indication that it's ON/OFF not EITHER/OR on the side of the unit.

Either/or meaning 120/240V?

Edit- some good pix of the back of the board with the caps and relay would help me trace it out if you are stuck.

[pentium]

Re: Super unreliable tape drive PSU

Either/or meaning 120/240V?

As in either the relay contacts are connecting A to B or they are connecting A to C.

Anyways, pictures are helpful.

[kaboom]

Re: Super unreliable tape drive PSU

I felt like I was missing something. All along, R19 (lower left corner) has blown and R43/R44, maybe R45, look like they were running hot. Check that the 5W 3.6K wirewound is good, and check Q5 for shorts.

Also check that big red R5 (upper left corner) for resistance. Check any components above hotspots on the back of the board.

I know that sounds a bit vague, but if you took some macro pix of the areas of interest, we could see the color codes better. R19 is (was) a yellow-purple-something- a 47x.

Check any of those diodes near that opamp. Is that an LM356 or 358? Just use continuity check- if the diodes have shorted, they're shorted.

If it's just a single opamp (LM356), it may sense the voltage on the caps and drive a small TO-92 transistor to pull that relay in. IOW, if that opamp ckt does not work properly due to bad parts, the relay will never close and fail to set the recto up as a doubler. Hence your low bus voltage. First determine if it's a 356 or 358, then get its pinout/datasheet and make sure none of the inputs or outputs are 'stuck' at its supply voltage. If they are, the opamp is bad and also needs to be replaced.


In the attached image, check the following:

If both red points have continuity to each other

and

The yellow point to either of the blue ones also has continuity, then the relay takes place of the 120/240 switch; ultimately pulled in by way of the opamp/comparator.

In your OP, you said that any suspect parts checked ok and were put back. Just be aware of any carbon residue under those hot resistors- any leakage will upset the bias on, and change the setpoint of, that comparator. And if the element is exposed on R19 and touching nearby traces, components, or leads, then fix that, before banging your head in frustration.

[pentium]

Re: Super unreliable tape drive PSU

R19 has already been pulled from circuit and tested and still reads good. R43 and R44 read 74.4 and 75.2k each. Their bands are red, brown, yellow, black, black, gold (or brown. The light I'm working in is awful). The 5W is good. Q5 is one-way from S to D and S to G. R5 and R4 also read their proper values. That is also an LM358N. R19 was yellow, purple (or brown), (something), black (or black-ish). It currently reads 4.73k. All the diodes I poke at seem to be working one way so they should be fine. Q1 was also tested and seemed okay.

[pentium]

Re: Super unreliable tape drive PSU

Going to BuMP this because two years later I have only just returned to it.

So I traced down the bridge rectifier which is a KBU8G and that I can see it's properly converting 117v AC to 117v DC. I can see that either DC polarity passes through a CL-40 current limiting resistor before going to each of the two primaries. The other sides of the capacitors travel to P2 and to the secondary side of the PSU as well as to R4 and R5. It also travels to the RKS-5DG relay which is normally open, otherwise it would travel to one side of the AC input of the KBU8G. It seems to be powered by something up in the region by U2. (see previous attachments)
In other words I still can't say why We're only seeing almost 80v on the primaries, assuming that's not what we should be seeing at all.
I have not yet contacted Tectrol regarding schematics.

[pentium]

Re: Super unreliable tape drive PSU

I'm not entirely sure what I have done but it now fires up regardless of it being attached to the drive or not. After that the drive now immediately reports logic errors. Poking at the rails finds that +5.1, -12 and +12 are within tolerances by about .1 or .2V but +38 reads +33.5 and +5 reads +4.5.
There's evidence it was recapped before but a few caps (50v 1500uf, 40v 220uf, 50v 47uf and two 25v 4.7uf's) were not disturbed. Those are some oddball values I can't locally find for cheap either.

[pentium]

Re: Super unreliable tape drive PSU

Also, to add to the symptoms and answer a previous question, the PSU is a true autoranging device so you don't need to change anything when powering it from 240v. That being said, while the front power button does not work on 120v line power (pressing it does nothing), it will work under 240 line power. (pressing it turns the tape drive AND the PSU on or off)

[ben7]

Re: Super unreliable tape drive PSU

Going to BuMP this because two years later I have only just returned to it.

So I traced down the bridge rectifier which is a KBU8G and that I can see it's properly converting 117v AC to 117v DC. I can see that either DC polarity passes through a CL-40 current limiting resistor before going to each of the two primaries. The other sides of the capacitors travel to P2 and to the secondary side of the PSU as well as to R4 and R5. It also travels to the RKS-5DG relay which is normally open, otherwise it would travel to one side of the AC input of the KBU8G. It seems to be powered by something up in the region by U2. (see previous attachments)
In other words I still can't say why We're only seeing almost 80v on the primaries, assuming that's not what we should be seeing at all.
I have not yet contacted Tectrol regarding schematics.

What voltage ratings are the main caps? If they are 200V, then they must be in series, so the PSU can work on higher mains voltages. Then, with 120VAC input, you'd see about 80V on each capacitor

[pentium]

Re: Super unreliable tape drive PSU

Yeah, they're both rated for 200v.
So if you say 80v each then we are in the neighborhood.