Tek 2235 power supply

[bico959]

Hello,

I've got the 12khz chirp problem but followed the posts here and still can't
seem to get it going. I would greatly appreciate any help.

I find nothing that others have done will restore operation, even partially.
I have replaced all the electrolytics in the power supply... The 510k's all
measure fine, no obvious burned parts or bad solder....

Help?
Thanks
RB

 

[Bill Dukenfield]

Hello,

I've got the 12khz chirp problem but followed the posts here and still can't
seem to get it going. I would greatly appreciate any help.

I find nothing that others have done will restore operation, even partially.
I have replaced all the electrolytics in the power supply... The 510k's all
measure fine, no obvious burned parts or bad solder....

Help?
Thanks
RB

I've checked everything in the 2235 power supply on my bench and
everything looks good.

The only parts left to check are the SMPS transformers. If these are the
problem I'm going to float test the scope!

JAM

 

[bico959 via ElectronicsKB.com]

I get about 53V at the 46V test point. Driver tranisistors on the heat sink
test ok, other trannys in the circuit test ok. HELP?

~RB

 

[bico959 via ElectronicsKB.com]

I don't suppose you have a scan of the power supply you'd be willing to share?
;-)

 

[Jim Yanik]

It sounds like the pre-reg is current limiting,meaning the power oscillator
is pulling too much current.Meaning something is loading it.

I've checked everything in the 2235 power supply on my bench and
everything looks good.

The only parts left to check are the SMPS transformers. If these are
the problem I'm going to float test the scope!

JAM

have you disconnected the HV multiplier?
Does the pre-reg output ~43 VDC at about 0.8-1A load? IIRC,for the 2235's
it limits at ~1.3A.
You can test that separately if you build a dummy load.
You can supply the power oscillator with an external 43VDC PS to test it
and the mainframe circuits.IIRC,there's also a 6.2V zener in the power
osc control circuit that goes bad.I've even seen where one of the two power
osc xstrs is open and the other is -trying- to work,but can't do it alone.

It's also possible to have decoupling caps in the various circuits break
down and load the supply.Those are real fun to track down.....

It's VERY rare that the XFMRs go bad.
secondary electrolytics #1(ESR),HV multiplier #2.

and ordinary non low-esr electrolytic caps don't work for long,if at all.

 

[Jim Yanik]

I get about 53V at the 46V test point. Driver tranisistors on the heat
sink test ok, other trannys in the circuit test ok. HELP?

~RB

TOO HIGH.
you are triggering the overvoltage crowbar SCR.that throws the pre-reg into
current limit.
BTW,the reference zener for the crowbar circuit can be too low in value.

READ the circuit description in the service manual.It's worthwhile.

the pre-reg should output 41-44 VDC at normal load.

 

[Jim Yanik]

I don't suppose you have a scan of the power supply you'd be willing
to share? ;-)

http://bama.sbc.edu/tektroni.htm

the 2230 and 2236 are the same PS,I believe. Or at least close enough.

 

[bico959 via ElectronicsKB.com]

Hey thanks for the help!

Unfortunately I don't have the service manual or schematics. I'd probably
toss this thing except it was in a clean room and never saw much use (like
new).

I did remove the wire going to the multiplier=no change. I also removed the
+/- 8v and 5v jumpers across to the rest of the board, also no change...
seems under loaded to me... not sure of the current.

I'll have to grab a power supply from work and try that. Also, should I try a
variac?

Thanks!!
RB

 

[Bill Dukenfield]

I don't suppose you have a scan of the power supply you'd be willing to share?
;-)

Got email?

I have to set up a scanner. I have the service manual.

JAM

 

[Jim Yanik]

Got email?

I have to set up a scanner. I have the service manual.

JAM

the BAMA 2236 pdf is 67 MB. Having a 48K dialup,I declined to DL it. ;-(

Maybe later tonight.....like at 3AM.

 

[bico959 via ElectronicsKB.com]

Hey thanks for the link. Luckily I have good ol cable! Ok, now things are
coming more clear. And yes, the circuit description definitely helped.
Unfortunately too busy with my [stupid day] job to do any further
troubleshooting for a bit...

Thanks again for all the help.

~RB

 

[Fred Abse]

the BAMA 2236 pdf is 67 MB. Having a 48K dialup,I declined to DL it. ;-(

A lot of their .pdfs are enormous. Somebody scanned them as gray scale,
rather than 1-bit. Silly that, 'cos all the photos are screened, anyway.
They reproduce fine as 600 dpi B&W.

 

[Jim Yanik]

Hey thanks for the link. Luckily I have good ol cable! Ok, now things are
coming more clear. And yes, the circuit description definitely helped.
Unfortunately too busy with my [stupid day] job to do any further
troubleshooting for a bit...

Thanks again for all the help.

~RB

hey,can you email me the PS schematic?
I did have a 2236.pdf,but it was corrupted from PC crashes.(I need a
battery backup)

Then I may be able to give you further help.

 

[bico959 via ElectronicsKB.com]

I put it here: http:www.computersolutionsgv.com/tek-2236.pdf

My BW is small, but it'll get there!

Cheers!

~RB

 

[bico959 via ElectronicsKB.com]

Hi Jim, did you manage to get the download? Here is my latest findings. I'm
not over volting. When I measured on the correct ground point, test point 950,
test poin 940 never quite reaches 40 volts. This would tell me that there is
an over current problem however I have lifted all the jumpers (+/-8.6, +5,
+30, +100) and I removed the voltage multiplier completely from the board and
there is no change. So is the over current on the primary side then? I went
ahead and removed the GP caps and put in hi quality Nichicon low ESR's in
their place. I have a good sawtooth wave at pin 5 of U930, oscillation is
visible through Q908 and the FET. The SCR doesn't have anything on the gate.
I can see the voltage drop at R926 (it measures 150k) when it 'chirps'. There
is a short duration 5V pulse from the IC at pin 14 (the ref output) and I can
see a small increase in the voltage at pin 16 also timed with the chirp. R907
measures good (.2 ohm). On the other side of the transformer, each rail tries
to come up gaining about 3/4ths their normal values and then falling again.
My probe hit and killed Q930 and now I have to order a replacement for it. I
see 1 volt at pin 12, no chirping and Q930 is resistive now...

Any help is greatly appreciated. I will paypal you a beer!

~RB

 

[bico959 via ElectronicsKB.com]

Here is a MUCH smaller file (200k), just the power supply:

http://www.computersolutionsgv.com/tek-2236ps.pdf

Thanks in advance!

~RB

 

[Jim Yanik]

Hi Jim, did you manage to get the download?

Per your last post,I downloaded the PS schematic;I was NOT going to try
downloading 70Megs of pdf on a 48K dialup.

Here is my latest
findings. I'm not over volting. When I measured on the correct ground
point, test point 950, test poin 940 never quite reaches 40 volts.
This would tell me that there is an over current problem however I
have lifted all the jumpers (+/-8.6, +5, +30, +100) and I removed the
voltage multiplier completely from the board and there is no change.
So is the over current on the primary side then? I went ahead and
removed the GP caps and put in hi quality Nichicon low ESR's in their
place. I have a good sawtooth wave at pin 5 of U930, oscillation is
visible through Q908 and the FET. The SCR doesn't have anything on the
gate. I can see the voltage drop at R926 (it measures 150k) when it
'chirps'. There is a short duration 5V pulse from the IC at pin 14
(the ref output) and I can see a small increase in the voltage at pin
16 also timed with the chirp. R907 measures good (.2 ohm). On the
other side of the transformer, each rail tries to come up gaining
about 3/4ths their normal values and then falling again. My probe hit
and killed Q930 and now I have to order a replacement for it. I see 1
volt at pin 12, no chirping and Q930 is resistive now...

Any help is greatly appreciated. I will paypal you a beer!

~RB

have you replaced C925? 100uf/25v,IIRC,Low-ESR.
that is the housekeeping supply cap for the TL594.
it charges thru R926 to about 14-16V,which allows the IC to start,then a
winding of that inductor takes over supplying power to the IC,thru
CR920.Otherwise,the IC quickly draws down the charge on C925,and the IC
stops.

 

[bico959 via ElectronicsKB.com]

Thanks Jim. I did replace it, but with a GP 100. Do you think that might be
it? Should it be low ESR to work? Also, do you know of a common cross for
Q930 and Q928?

Thanks... BTW I also have a tek 2215 with power problems (go figure).

Cheers!
~RB

 

[Jim Yanik]

Thanks Jim. I did replace it, but with a GP 100. Do you think that
might be it? Should it be low ESR to work?

it definitely should be low ESR.
It gets supplied by the HF switched current from the inductor T906.

Also, do you know of a
common cross for Q930 and Q928?

http://www.sphere.bc.ca/test/tekequiv.html didn't list any for those.
I suspect gen purpose xstrs would work for them.

Thanks... BTW I also have a tek 2215 with power problems (go figure).

BTW,the 2215 PS is essentially the same circuit as the 2235/36.
the part designations are different however,and the 2215 pre-reg limits at
~0.9A instead of 1.3A.

you can isolate the pre-reg by lifting the .51 ohm res R949,disabling the
inverter section.
You should cook up some load R for the pre-reg.
I don't know what value a minimum load would be.
I used a bunch of parallel 7W W-W resistors I had on bench stock. 45-
100ohms should work,but pay attention to their wattage.

 

[Danny_40]

Hi bico959
I have the maint manual and I can share it.
I think that the U930 (TL594CN) don't work fine, seem that reach the end of the sathow ramp at pin 5 instead of turn off Q9070 (IRF820 for 110 Vac line) before, or the divider resistor R921 (30k 1/4W) R916 (510K 1/4W) don't feed the correct voltage reference, or the Q908 PNP transistor (2N2907A) don't drive off Q9070
Danny_40