PSU - Power Supply Unit - TDS 500 700 - Failure while color adjustment

[Tantratron]

While I was doing the brightness and contrast adjustment routine on my TDS794D, suddenly the oscilloscope stop working then I check the fuse was blown (see attached). The scope was working ok so far since I've bought it last december on eBay. In fact I did attempt a brightness and contrast adjustment a few days ago, everything was OK so no idea what could have happened now 

The PSU model: ZYTECH 229430040 Rev G - Date 9831

What could be in general typical root causes of a direct fuse blow up with PSU of these TDSxxx serie C and serie D ?

I do have a spare ZYTECH 22943040 Rev E - Date 9730 from my TDS540C so is it better to swap both PSU then see if my TDS794D works again with its A10 A11 and A30 boards ?

The fuse has written on it BUSS AGC 6 - 250V so I guess it is 6A, probably fast fuse but what replacement model would you source nowadays ?

More generally, do you know if the color TDS will use much more energy than the mono TDS model ?

Could we consider that while doing the brightness then contrast adjustment, the power consumption goes much more high than when operating the oscilloscope to display 4 curves, some maths and high sampling ?

Thank you, Albert

[Tantratron]

This week-end, I've deconstructed the good PSU from my TDS540C and the failed PSU from my TDS794D. Please see attached picture showing both of them, on the left (TDS794D rev G) and on the right (TDS540C rev E) where it seems only very minor design differences.

I'm quite afraid of this repair because a quick check using my DMM shows all transistors, diodes and the FWBR from the enclosure of the big electrolytic to be shorted. These are all inside the metal housing where it is written DANGER, I think this the PFC stage.

Later I'll check other transistors and diodes in the A16 board but what could have made such a destructive avalanche of so many transistors (Q5 Q6 Q7 and Q8), so many diodes (CR4 CR14 CR16) and the FWBR (Full Wave Bridge Rectifier) while the oscilloscope was going through the routine to adjust the CRT brightness and contrast ?

Before I attempt to install the good PSU from my TDS540C into the TDS794D, what do you recommend that I test on other boards (A11, A10, A30) of the TDS794D to make sure there were the root cause of the PSU failure ?

I mean here how to make sure no risk that connecting a good PSU would not destroy it then no more spares to get going.

Thank you, Albert

[CaptDon]

I have no idea what you did to cause the failures to such a great degree.  The adjustment tool should have been plastic to avoid shorting items inside the protective metal housing around the CRT boards where the adjustments are located. I have dead shorted the 5vdc supply several times when servicing a TDS644A and never had a hard failure. The supply shuts down and you have to recycle power. I can't imagine how you managed to blow it up so bad. The power supply has several protective circuits in it, but if you shorted the 21vdc to ground the PWM circuit may have gone to maximum pulse width and just wiped out everything in the supply and possibly beyond the power supply itself. OUCH!!!!

[Tantratron]

I have no idea what you did to cause the failures to such a great degree.  The adjustment tool should have been plastic to avoid shorting items inside the protective metal housing around the CRT boards where the adjustments are located. I have dead shorted the 5vdc supply several times when servicing a TDS644A and never had a hard failure. The supply shuts down and you have to recycle power. I can't imagine how you managed to blow it up so bad. The power supply has several protective circuits in it, but if you shorted the 21vdc to ground the PWM circuit may have gone to maximum pulse width and just wiped out everything in the supply and possibly beyond the power supply itself. OUCH!!!!

I always use plastic adjustment tools wether these TDS or any other test equipment plus frankly, the two potentiometer R403 and R404 are fully plastic, same for their screw and they're accessed from the outside once we floppy disk is removed !

Furthermore the failure or power off happened when I've pushed the center push button to go after the contrast routine following the brightness routine per the service manual.

On a side note, the J5 connector brings +25V and +5V to the A30 color board.

[Tantratron]

Ok here is an update where I've de-soldered all semiconductors from the PFC section and the DC-DC converter section (see attached pictures):

• Partial good news, the PFC section is after all OK where Q5 Q6 CR5 (Full Wave Bridge Rectifier) and CR6 are healthy except the PFC-Boost diode CR4 which is dead
• However bad news, the DC-DC converter is KO where Q7 and Q8 are dead but CR14 and CR16 are healthy
  

So I'm still not sure what really happened while the oscilloscope was running fine for few weeks then during the brightness-contrast adjustment under EXECUTE routine the PWM controller just KILLED the DC-DC converter. I guess the problem must be in the A18 board which commands the gates of Q7 and Q8. As a reminder the PSU failed when I did choose the White Field push button while entering the Contrast Adjustment routine.

The Q5 Q6 Q7 and Q8 are all IRFP450 mosfet, the CR4 is HFA08TB60 diode so I'll try to order but I'm still worried how to understand, solve and check what major failure triggered such fatal destroy where there are so many safety sensors in the A18.

As you see in the picture, I've removed the C11 and C12 for checking as well to ease the checking health state of the gate drives in the A17 board.

[Jwalling]

Might be a long-shot, but was the aluminum sheet metal cover under the CPU board screwed into the heatsinks on the PSU when you were running the unit? I think that the top sheet metal is used as part of the heat-sink strategy...

When I run them without a case, I have a fan forcing air into the right side of the unit as well.

[Tantratron]

Since I've had a spare PSU from a partial failed TDS540C, I've decided to take the risk of install this working PSU from C serie into my TDS794D. On a close view visual inspection, it seems that the serie C and serie D circuit are almost the same, maybe the same BOM except two local parts of the PCB which went some local re-routing (see attached pictures of these 2 sub-parts).

Anyway after re-installing the C serie PSU, my TDS794D works again. I guess there must have been a problem in its prior PSU to destroy both Q7 and Q8 of the main DC-DC converter, the root cause would not be the A30 display board or A11 logic board or A10 acquisition board.

For whatever reason, all the safety controls (over voltage, over current) and sensors governed by the A18 board (PWM control and fault signal conditioning) has failed to shut down the PSU. Once the Q7-Q8 were continuous conductive, it is logical as a consequence to destroy as well the CR4 (boost-diode) where I'm quite lucky the Q4-Q5 were not destroyed due to this power electronics avalanche.

I'll order soon the IRFP450 mosfet and the HFA08TB60 diode, re-solder the C11 and C12 capacitors then see if I can test un-loaded the PSU plus how to simulate the On/Standby push button effect.

Might be a long-shot, but was the aluminum sheet metal cover under the CPU board screwed into the heatsinks on the PSU when you were running the unit? I think that the top sheet metal is used as part of the heat-sink strategy...

When I run them without a case, I have a fan forcing air into the right side of the unit as well.

I do not think this to be a problem for the PSU because the specific metal enclosure where Q4 Q5 Q7 Q8 and CR4 are passively heat sinked contains a thermistor RT7 which serves the monitoring of the temperature. By the way, when I was doing these color screen adjustment on both my TDS784C and TDS794D, they never run long time but you're right, it is not good idea to have these Oscope run without the blue cabinet to compress the fan air flow. I think the major heat generation comes from the acquisition board A10 versus the logic board and CRT board.

[gmilliorn]

Interesting that the newer, but failed board, does not have the bodge resistor over the LM339.
From the pictures alone I would have guessed the opposite.

[Tantratron]

Interesting that the newer, but failed board, does not have the bodge resistor over the LM339.
From the pictures alone I would have guessed the opposite.

Well actually the 10K bodge resistor you mention (R185) which crosses the LM339 on C serie is still there except the PCB now has correct trace, through hole to host R185. You can see this very same 10K resistor on the top of both attached pictures if you count the number of resistors.

In fact this design seems similar to the B series when looking at the component manuel of the TDS520D so this PSU design could be the same for B series, C series then D series which in a way is good news so we gave the schematics.

[Tantratron]

OK good news, I've finally purchased few mosfets and diodes then replaced the tow mosfet DC-DC converter and one PFC diode, the PSU works again fine. Since the equivalent mosfets are now IRFP450PBF, I decided to replace as well two PFC mosfets to make sure same parts so in total, I've installed 4 switching mosfets and one PFC diode.

No idea if there is a big difference between legacy IRFP450 and IRFP450PBF.

Same with PFC diode which was HFA08TB60 which I've replaced by HFA08TB60PBF.