Tektronix Scope TDS 524A LV PSU Repair Advice

[vectra_1098]

Hello,

I am working on a faulty Tek Scope TDS 524A LV PSU, the scope will not power up!
I have removed the LV power supply from the scope placed it on the bench.
Powering the PSU and monitoring the low voltage outputs at J5 & J6,
also as per the manual linking pins 3+5 of J5 together to allow start-up
The no-load outputs at J5&J6 are all zero ?
I have checked the ac input supply and filters (and fuse ) and Bridge(CR5) output and there is a
 335V dc across  C12 the Bulk capacitor.

I suspected the Standby circuit , I removed Q9 and transient suppressors  VR4 & VR5.
I have obtained new components for VR4 &VR5 ok.
Q9 in this power supply is a MJE 1320 Power switching transistor,
But they are very difficult to get hold off and other similar power supplies use the
Bu508A as Q9. I don’t think there parts are interchangeable unless the stand-by circuit is modified to take the BU508A part in this power supply ?
Is there an equivalent part for the MJE 1320 ??

**Referring to TDS524A/TDS520B service manuals for TDS 524A Scope

Advice would be nice to have.

Thanks
T

[pbarton]

Can you identify the source of the manuals you are using, as it often helps if we are all referencing the same document?

[vectra_1098]

Hi,
In my opion  Tek did not want anybody to repair these scope power supplies
and so to  find the correct schematics /pcb layouts is made difficult.
Ref Docs:
The Manuals reference are -
1.TDS 520B Mod Cm
Digitising Oscilloscope
Component Service Manual
070-9710-03   

2.) TDS 520A,524A,540A & 544A
Digitising Oscilloscope
070-8713-03
3.) A schematic drawing from an  EE blog, Post” Help with my TDS 520B showing a standby Supply”.
My Notes:
Ref 1:  Manual does not give component level detail although its for the TDS 524A scope.
Ref 2: Does provide schematics  which are similar but they are for the TDS 520B not the TDS 524A
Ref3:EE blog post  Standby Supply,shows a bipolar transistor Q9 type MLE 1320 and transorbs VR4/VR5.

My Power supply/custom rectifier LR40071
Zytec part number:22904701 rev g
300w,3.8A,1phase
Thanks for the reply

[Jay_Diddy_B]

Hi.

There are at least two standby power supplies used in these TDS5xx and TDS7xx scope.
There is some useful information in this thread:

https://www.eevblog.com/forum/testgear/fixing-tek-tds684-psu/msg453682/#msg453682


This is one:



This is the other:



Regards,

Jay_Diddy_B

[Ordinaryman1971]

I have identical problem with TDS744A, what was your replacement for VR4 and VR5... would they be the same in both power supplies?

[vectra_1098]

Hi,

=======================================================
Jay_Diddy_B Wrote Earlier :-

There are at least two standby power supplies used in these TDS5xx and TDS7xx scope.
There is some useful information in this thread:

https://www.eevblog.com/forum/testgear/fixing-tek-tds684-psu/msg453682/#msg453682
==========================================================


I can see the Stand-up supply schematic version with the Bipolar transistor  Q9 and VR4/5.
I removed these components from the PSU and checked them with a Multimeter
and other test equipment and noted their part numbers.

My test equipment does not like Q9 and VR4/5

I have obtained the Transorbs via Ebay
TVS Diode VR4=part number   1.5KE220CA St Microelectronics
TVS Diode VR5=part number   1.5KE250CA St Microelectronics

As I mentioned in an earlier post My Transistor(Q9) is a MLE 1320 which I read is obsolete ??. What Q9 do you have in your PSU ?

I am going to reverse engineer my pcb around the Standby supply area to see how close it is with the schematic
and possibly do a spice simulation for he circuit

Thanks






I can identify my bip

[Ordinaryman1971]

I will look it up today and take some pictures.

[Ordinaryman1971]

Took few days but... here are some pictures.
Q9... I have BU508A in mine.
I've just installed the suppression diodes hoping that that's the only damage, but q9 is shorted as well. Don't have those transistors so that's the next purchase.

I hope it's good news about the replacement transistor. The BU508A are about $2 a piece.

[radioguy123]

So what was the outcome of this repair ? Was the power supply fixed ?
I have a similar problem with a 620-0063-04 psu.
Mine has the Q9 BU508A but it test's good.
The TVS diodes were bad but that didn't fix it.
All components on either side of the standby circuit test good now.
But still want's to overheat the TVS devices and still no standby output ??

[chronos42]

All components on either side of the standby circuit test good now.
But still want's to overheat the TVS devices and not standby output ??

Check C17 (47µF) again for ESR or replace it with a very high quality, low ESR Capacitor.

[radioguy123]

Hello chronos42
Have you actually seen a case when C17 was a cause for this problem ?
I did check the capacitance of C17 and it looked good.
I have a new capacitor on order.
Also I'm getting an ESR meter.
Thanks for your reply

[madao]

Old Standby PSU is very sensitive  by bad  ESR of C17 eletrolytics.
If ESR is  too high = more spikes, which it is absorbing  by TVS-diode -> overheated !   If TVS diode is dead, BU508A play kamikaze.

I have similar  failure by TDS540A, but  C17 has good ESR, only  Transistor is dead. I have not MJE8501, it is replaced by TIPL760 (TIPL760 has much less limit voltage, not recommed!) BU508A is in europe  cheap, but i have MJE8501 w TO220 case in PSU. 

[radioguy123]

Hi madao
I think you are right. This version of psu is very sensitive.
And if you leave your oscilloscope plugged in all the time, that makes matters worse.
The stand by supply is running continuously.
I'm just about at my wits end on this one. I've removed nearly every part and tested it out of circuit.
Of course not the transformer, it has continuity but still could be bad.
Probably a lot of these psu's have hit the dumpster for just this reason.

[Ordinaryman1971]

I've had to put this project on the shelf for a while.
The outcome of changing the suppression diodes and the transistor was none. The power supply is still dead.
I was just looking at it the other day. I've ordered working power supply few weeks ago so as soon as I have some time I will have it changed and see if the scope is working ok with the "supposedly" working power supply.
Then I will go back to fixing the not working one .... that's the plan.
I didn't really test it outside of the scope.
Let me look at the way to actually have it tested by linking the pins on the connectors.
Didn't change any caps...  that will be the next on the list.

[rf+tech]

Hello radioguy123, Ordinaryman1971,

By disconnecting Q9 base, an in-circuit static DC bias test can reveal useful information about multiple components.



With Q9 base connected, what DC voltages are measured on Q9 base and collector in a non-functioning standby supply?

It is expected that T3 pin 4 produces a negative-going pulse to start oscillation, when Q9 base becomes sufficiently forward biased to cause collector-emitter conduction, as the DC bulk line rises following application of AC mains/line voltage. Scope this point, with triggering derived from Line Trigger on J5 pin 1.

RF+ Tech

Edited to resize image

[radioguy123]

So there should be a +0.6v positive voltage on the negative terminal of C17 ?
And this is with the Q9 base still disconnected ? right ?

[rf+tech]

radioguy123,

So there should be a +0.6v positive voltage on the negative terminal of C17 ?

Circuit analysis reveals this to be true. If this tiny reverse voltage is a concern, disconnect C17.

And this is with the Q9 base still disconnected ? right ?

Attempting to diagnose this non-oscillating condition with Q9 base connected is a bit problematic. RT5, RT4, T29 and R25 provide forward bias to Q9 base-emitter junction. At the same time, Q9 collector-emitter junction begins to conduct, lowering the voltage at RT4/R29 junction. In the non-oscillating condition, where feedback is not present, Q9 base forward bias and Q9 collector voltage settle to a point of static equilibrium where collector voltage limits base forward bias.

How is one to get around this obstacle and better evaluate components in the feedback path?

The objective of this seemingly unconventional test is to check the complete series circuit of RT5, RT4, R29, R25, CR11, VR1, CR10 and T3 secondary pin 4 to 3. If Q9 base is not disconnected, the base-emitter junction acts as a DC bypass around CR11, VR1, CR10 and T3 secondary pins 4 to 3. Again, circuit analysis shows Q9 base must be disconnected, as a prerequisite for this test. The voltages measured can confirm these components to be okay, or provide valuable clues that help pinpoint a fault.

Now for another less than obvious potential cause of failure to oscillate. A shorted rectifier or bypass capacitor on one of the T3 secondaries could reduce feedback amplitude to the point where the circuit cannot oscillate.

For the record, it would be advantageous to know what resistances RT5 and RT4 exhibit at room temperature.

Hope this insight and detail helps understanding,

RF+ Tech

[chronos42]

Hello chronos42
Have you actually seen a case when C17 was a cause for this problem ?

Hi,

yes, i had one broken PSU with a C17 with too high ESR. The complete circuit fails in this case, because the bias voltage is not clean (spikes) and can trigger Q9 in a wrong way. Also the bias can be too low, therefore the circuit is not working as intendet. In this case the TVS diodes were overloaded, too. From my experience C17 is a very critical part in this circuit.
I would also follow the recommendations from rf + tech to exclude other parts as failure source.

[radioguy123]

Wow a lot of useful information here. Thanks guys. Now were really getting somewhere.

[radioguy123]

I replaced C17 and there is no change in the operation. A few seconds and the TVS diodes are about to melt their own solder.

Ok so I got a chance to measure the static voltage points that rf+tech has shown.
This is with the base of Q9 disconnected and board powered up.

The voltage at the disconnected base of Q9 connection is: +6.55 volts
I measure +5.96 volts between the Zener VR1 and diode CR11
And at the negative terminal of C17\anode of VR1 I see +0.43 volts.

rf+tech do you make anything of my measurements ? I think they are pretty close to yours. Still isn't working.

[rf+tech]

radioguy123,

Yes, these voltage measurements are good. Did the TVS heat up with Q9 base disconnected?

The total withstand voltage of VR4 and VR5 should be a minimum of 470 Volts. If the peak to peak voltage across VR4 and VR5 exceeds some 500 Volts, then they will begin to conduct and heat. Are these original or have they been replaced? If replaced, what part number was chosen?

I’ve been under the impression that this circuit is not oscillating.
So this raises the question, have you actually scoped the signals at Q9 base and collector?

What DC voltage is measured at VR1 anode? What is the DC Bulk voltage?

RF+ Tech

[radioguy123]

No with the base of Q9 disconnected the TVS did not get warm at all.

The TVS devices I have in the circuit right now are 1.5KE250CA

Well I guess it's oscillating. But in a very wrong way. There must be tremendous voltage spikes to heat the TVS that much.

No I have not yet scoped the points you point out. I'll have to get back to you on that.

I show the voltage at VR1 anode above. It is +0.43 volts.

I'm running on a mains voltage of 123vac. The bulk voltage I measured is about 169vdc.

I'm wondering if the BU508A could be bad even if it tests good ? This could make it operate strangely. I think I will change it just because.

[rf+tech]

169 Vdc is expected, before the Boost Converter takes over. The boost converter needs +12 V Bias to operate. That comes from T3 secondary.

I understood +0.43 Volts at CR1 anode to be with Q9 base disconnected. What is VR1 anode voltage with Q9 base connected? Is it anywhere close to -6.08 Vdc? If so, then yes, the circuit is oscillating.

CR7, R21 and C15 form a snubber that dampens the flyback pulse on T3 primary. In normal operation, VR4 and VR5 would never conduct enough to get hot. A failure in the snubber network could cause the waveform amplitude to exceed the breakdown voltage of VR4 and VR5 and explain their overheating.

The symptoms suggest Q9 could be struggling to overcome excessive load on T3 secondary. This is why it is important to make lots of additional measurements. Effective troubleshooting skills are gained by measurement and analysis, not replacing parts just because.

What voltages are measured at J4-7, J4-5 and J2-7? If any of these are significantly out of range, try disconnecting one end of the associated rectifier (CR17, CR15 or CR8) to isolate that particular load.

RF+ Tech

[chronos42]

Hi Radioguy,

maybe a dumb question, because I assumed that you did check this first, but: Are you really sure that there is no short or overload in the +12V and +15V output rails?

[radioguy123]

Hi chronos42
Yes I did check if those were shorted. I went one step further and totally disconnected them. Still no go.

But I think I have found the problem. I tried another BU508A transistor and got no oscillations at all.
Since I wasn't getting any oscillations I measured the voltage on the BU508A collector. 8 volts ?!!?!?
It seems that one of those (unobtainium) thermistors was just going open when under load.
It measures fine when no current is being drawn.
They both measure around 30 ohms. I just don't have any information about these devices.
There is a single marking on them " SL " ? Does anyone have a part number or any more info on what these are ??
thanks