Hameg 408-1A

[tautech]

Oops you're right, fatigue has set in in that case my information was all wrong, it was the connection to the 330pf cap that was broken and that wouldn't cause the problem I doubt?

Correct, that shouldn't stop it.

I checked the components around T403, that is R432, D423, R417, T403 and R422, all checked out OK but while checking that leg, I also checked L401 which shows as a 270uH but measures 64uH on my LCR bridge. I checked R433 as well and it is 1M5 so that OK.

Test frequency for L401 could alter measured values.

We need to understand IC405 and how it affects the SMPS IC VCC, it's just a 4093 so cheap to replace.
Gotta go.....later.

[Specmaster]

Oops you're right, fatigue has set in in that case my information was all wrong, it was the connection to the 330pf cap that was broken and that wouldn't cause the problem I doubt?

Correct, that shouldn't stop it.

I checked the components around T403, that is R432, D423, R417, T403 and R422, all checked out OK but while checking that leg, I also checked L401 which shows as a 270uH but measures 64uH on my LCR bridge. I checked R433 as well and it is 1M5 so that OK.

Test frequency for L401 could alter measured values.

We need to understand IC405 and how it affects the SMPS IC VCC, it's just a 4093 so cheap to replace.
Gotta go.....later.

Thing that keeps bugging me is that before I took the psu out of the scope, I was getting dots on the screen and the LED's lit up, now, the LEDS are on for approx 2 secs and go off, nothing on the screen. I suppose it could just be a coincidence that it was going to fail anytime and it chose then to do so....

[tautech]

Thing that keeps bugging me is that before I took the psu out of the scope, I was getting dots on the screen and the LED's lit up, now, the LEDS are on for approx 2 secs and go off, nothing on the screen. I suppose it could just be a coincidence that it was going to fail anytime and it chose then to do so....

Maybe but you've done a bit of work on her so it's time to revisit that and double check you've not made any error/s.

Thinking about this a good part of today and I think we need use the inputs into the SMPS IC to help tell us what's going on so time to study both the schematic and IC datasheet, the parameters that get it going and keep it going.

As it's not wise to connect a scope (elevated voltages) we're stuck with using a DMM, hopefully yours has a Peak Hold function that might capture voltages on the various inputs at power up before it falls flat on its face.

Otherwise this is differential probe territory but you might be able to do something with Maths and 2 channels so that there's no probe reference connection needed and zero risk of ground loops.

Assuming the SMPS circuitry is faultless then the shutdown is normal behaviour for a fault in the circuitry it supplies as the current sense R457 in series with the transformer primary will throw an excessive voltage into the CS pin 11 input. Specs are on P4 of the datasheet. Small mention at foot of P7.
Further explanations and graph showing hysteresis on P12.

Yeah, yeah, I know SMPS is not your strength, not really mine either but that's where we need to go IMO.


Edit
BTW I looked at the EN 408 manual and there seems to be zero in the way of circuit description or theory of operation, these are just User manuals but at least they have schematics and PSU voltages.

[Specmaster]

I was revisiting what I have previously done and thats how I noticed the pad and trace were not joined. I will be tracing my steps back again shortly and also trying to fathom out IC401, 403 and 405 relationship.

What about the opto couplers, could they be playing a part in this?

As you said, the chips are relatively inexpensive so I also ordered up a replacement set while they are still available, the NE5560N is already obsolete.

[tautech]

What about the opto couplers, could they be playing a part in this?

I wondered that too but if they are now, why ?

One is to transfer the Line frequency to the timebase board so that won't have an influence on startup and the other is secondary side voltage feedback but if it's working correctly it should be only for regulation.

You could try feeding only the +5V from a bench supply and check the current drawn seems reasonable.
This would be the only rail I would check like this to minimise the chance of damage to componentry.

[Specmaster]

Thats sounds like a plan, I'll do that after lunch.

[Specmaster]

You could try feeding only the +5V from a bench supply and check the current drawn seems reasonable.
This would be the only rail I would check like this to minimise the chance of damage to componentry.

The results of the 5
+5 current drain test is without the psu being connected is 800mA and the single largest consumer of that is the digital board and not surprising given the large number of IC's on it, over 100 of them.

Testing the psu on the +5v draw is zero which I think is correct as the IC's on the secondary side of it are jfets and those on the primary side are connected to the 5v rail, instead getting their power directly from the incoming mains, via droppers and diodes and or the full wave bridge rectifier.

[tronde]

BTW I looked at the EN 408 manual and there seems to be zero in the way of circuit description or theory of operation, these are just User manuals but at least they have schematics and PSU voltages.

No description in the German manual either.

It says that all DC voltages are electronically stabilised and should be within +/- 2%. If outside it is something wrong in the circuit. It also says that the only adjustable voltage is the +12VDC. Adjustment and test-point for that is on the X/Y PCB.

+12V, +5V, -12V, -5V, +55V, +115V, -2000V and +12kV are listed as DC voltages and 6.3V and 12V are listed as AC-voltages. 6.3VAC is for CRT heater and 12VAC is for screen graticule light and mains triggering.

[Specmaster]

BTW I looked at the EN 408 manual and there seems to be zero in the way of circuit description or theory of operation, these are just User manuals but at least they have schematics and PSU voltages.

No description in the German manual either.

It says that all DC voltages are electronically stabilised and should be within +/- 2%. If outside it is something wrong in the circuit. It also says that the only adjustable voltage is the +12VDC. Adjustment and test-point for that is on the X/Y PCB.

+12V, +5V, -12V, -5V, +55V, +115V, -2000V and +12kV are listed as DC voltages and 6.3V and 12V are listed as AC-voltages. 6.3VAC is for CRT heater and 12VAC is for screen graticule light and mains triggering.

I know, it seems that there must be another manual somewhere for this, I downloaded what was supposed to the service manual, it doesn't give you the info for calibration either unlike the one I have for my Iwatsu scope which gives much more detailed information, about 200 pages and is called "Instruction Manual" and has the following sections within it 1) Specifications, 2) Operating Information, 3) Measuring Procedures, 4) Theory of Operation, 5) Maintenance, 6) Check and Adjustment, 7) Schematic Diagrams, Electrical Parts List and lastly 9) Mechanical Parts List and Illustration. It also is full of photos and screen shots etc. to assist.

For electrical parts it also provides the circuit reference, description and the Iwatsu parts number and it does that circuit by circuit and does the same for the mechanical parts which includes are the all the case parts as well, chalk and cheese eh?

There are other threads on the forum about Hameg scopes and it seems that Oldway has had cause to contact them before, maybe for his large number of scopes as more then a couple are Hamegs, and he recommends people to contact them as they did have in 2014 at least have spares still available. To that end I have also emailed them @ contact.uk@rhodes-schwarz.com to see if they can throw any light on this problem, and also how does this smps operate particularly the interplay between the 3 IC's in the control section. Hopefully I might hear something back soon, who knows, I thought is was worth a try at least. 

All the chips I ordered have been dispatched so should be with me in few days time, if we can find nothing else wrong, it may be an idea just to swop them out for the new chips and give it a go?

I had been toying with the idea of getting one of those cheap Chinese IC testers but I doubt they would be upto testing old obsolete chips as their internal libraries would be full with the more modern chips I think.

[tautech]

You could try feeding only the +5V from a bench supply and check the current drawn seems reasonable.
This would be the only rail I would check like this to minimise the chance of damage to componentry.

Testing the psu on the +5v draw is zero which I think is correct as the IC's on the secondary side of it are jfets and those on the primary side are connected to the 5v rail, instead getting their power directly from the incoming mains, via droppers and diodes and or the full wave bridge rectifier.

The secondary side opamps plainly get their supply from the unregulated +12 and/or -12V SMPS secondaries, nothing to do with the +5V rail.

Yes there will be another manual for the scope but I think with the datasheets and schematics it is viable to fix this.

To understand the operation of the logic gates you need to grab the datasheets and study the Truth tables.
4093 is a Nand Schmitt and 4049 is an inverting buffer.

This website can be handy to visualise the operation of many types of logic functions:
https://tams.informatik.uni-hamburg.de/applets/hades/webdemos/10-gates/00-gates/chapter.html

[Specmaster]

I'm poking around the Nand Schmitt and Inverting Buffer and have discovered that the T403 is infact drawn with its Emitter and Collector reversed. Also I have found that R407 is shown as 475k but measures 400k, would this make enough difference that it could shut down the psu?

[Specmaster]

Had a response from rhodes-schwarz today and they gave me the phone number of their technical support centre 01252 818 900 which goes straight through to the Munich office. I called it and got a recording saying today they have a public holiday, so I'll try them again tomorrow and see if they offer any guidance for us.

[Specmaster]

A quick update for those playing along at home, I forgot to phone their tech support line Friday, so I'll do that in the morning.

All the replacement chips have now been delivered so if all fails, I can try replacing the chips as it is possible that the previous damage could also have taken out a chip or two. I have slowly but surely working my through the schematic for the smps, locating and checking the parts against the specified values and so far nothing has actually lept out as a likely cause of the failure? The mystery rumbles on...

[tautech]

Also I have found that R407 is shown as 475k but measures 400k, would this make enough difference that it could shut down the psu?

Not sure but if it has drifted ~20% from original values it's on the way out.
Remove it and remeasure to be sure and replace if needed. Even crap SMD resistors will meet a 5% tolerance.

[Specmaster]

I took it out to test it, but because it was minimelf type it had been glued in and attempting to get it free from the glue, it pinged across the room and promptly became lost, oh well, I now have 50 on order so I'll have a few spares haha.

[Specmaster]

A quick update of the current position.

Still awaiting the arrival of the 475k minimelf resistors.

Spoken to the UK arm of Rohde-Schwarz who gave me a phone number for the technical support center in Munich, Germany. Had a long chat with a helpful person in Munich who explained that they only do the calibration there but emailed me a copy of their schematics and also gave me the name and contact info for a company who do the repairs for Hameg scopes in Germany.

I have now emailed this company giving them the information that we have arrived at, to see if they could direct us where to look for this particular problem and to check if the resistance values are good enough to continue or are we looking at other issues on the main boards?

I'm also informed that they carry spares so may be a source of replacement knobs, once the PSU is up and running. Worryingly though, they also have a download section on their web site for manuals they but don't list the HM408 in any of its guises??

So at the moment I'm in limbo awaiting resistors and hopefully a reply from the company (SKY Messtechnic GmbH) in Germany.

Here are the resistances to ground taken on the isolated PSU:
-12v     13k
-5v       11.5k
+5v      13M
+12v    5.3k
+115v   40k
+57v     66k

so no short circuits there although the +5v seems to rather high, especially in relation to the others, could this really be as high as 13M or doe we have an issue here?

[tautech]

IMHO there is something wrong with the high +5V rail resistance. Could it be an open joint or faulty connector......you need do some continuity checks.

Why ?
If there are sockets and looms, check the sockets pins for cracked and/or dry joints.

Let's say when it powers up the +5V rail presents little load for the SMPS.......the +5V rail is the rail that the
SMPS feedback is sensed from..........it then signals overvoltage to the feedback opto isolator.
Study the feedback circuit (PSU schematic, btm right) and its feedback path to the opto.

[Specmaster]

It seems that all the resistors connected to the op amps IC408, IC407 & IC406 are reading approx 33% lower (in circuit) so I pulled out the op amps and the readings are the same, tomorrow I'll lift one end of each and measure out of circuit. I didn't order any of the op amps (TL081 and TL802) but I did get just in case a set of IC's for the primary side inside of the screen can plus another opto for the timebase board interface. No sign of any cracked or broken connections at this stage, more might be revealed tomorrow when I start desoldering.

[tautech]

It seems that all the resistors connected to the op amps IC408, IC407 & IC406 are reading approx 33% lower (in circuit) so I pulled out the op amps and the readings are the same, tomorrow I'll lift one end of each and measure out of circuit. I didn't order any of the op amps (TL081 and TL802) but I did get just in case a set of IC's for the primary side inside of the screen can plus another opto for the timebase board interface. No sign of any cracked or broken connections at this stage, more might be revealed tomorrow when I start desoldering.

Let me restate my concerns as to the 13M resistance of the +5V rail.
I don't think this is anywhere normal, it should be MUCH lower. Find out why it is not.

[Specmaster]

Oops, looks like my DMM is misbehaving, possibly due to the caps on that rail? Sometimes it shows M instead of K it seems, I had to drag my AVO out and double check and the results are, depending on which way the probes, + or - to ground I get the following .75k or 22.5k, that seems to be better. I say due to the caps because the readings start low and increase slowly, which I guess is due to the 1000uf caps charging?

[tautech]

Oops, looks like my DMM is misbehaving, possibly due to the caps on that rail? Sometimes it shows M instead of K it seems, I had to drag my AVO out and double check and the results are, depending on which way the probes, + or - to ground I get the following .75k or 22.5k, that seems to be better. I say due to the caps because the readings start low and increase slowly, which I guess is due to the 1000uf caps charging?

It will be plus any others that aren't local to the PSU.

That's a more realistic figure. 

[Specmaster]

It will be plus any others that aren't local to the PSU.

The PSU is not attached to the scope so its just those on the PSU that will be having an effect.

DMM with auto ranging can be handy but they can also be a problem when the digits keep changing, whereas an analogue can be so much easier to read at times.

[Specmaster]

Double checked the resistors associated with the op-amps today which all checked out fine after lifting one leg from the pcb so I decided to remove the transistors for checking, T404 and 406 came out ok and passed testing, but the SMD ones are a different story, after using my hot air gun to remove T407 for testing, I promptly lost it when despite sticking it onto double sided tape to stop it escaping when probing with the test leads, it managed go flying into the ether, new ones on order...

Did not find any broken traces or bad solder joints anywhere as went round the board and checked traces point to point with my meter.

The +5v circuit all appears to be OK, diodes, caps and resistors, traces and sockets all checked OK.

I also tested for hotspots with an infra-red thermometer before removing the smps and all I found was a few IC's in places across the boards were warming up to mid 30C area, which is what I expect them to do anyway.

Had a response from Sky Messtechnic, they will respond with some schematics and tips / hints on what we need to check on Monday

[Specmaster]

The info from Sky is that the problem is likely to be contained within the red circled area which monitors if any circuit is taking to much power and shuts down the psu in that event. Given that they have seen the circuits resistance to ground and these passed with no comment, I think we can take it then there is likely to be a fault here somewhere. I have given everything as far as possible a quick check in circuit in that area and the biggest error so far found is that R462 on the +115v rail (221k) reads approx 78k. This could be partly due to the caps on that rail and the transformer windings? That would mean that D441 was suspect but that has been pulled from circuit and passes the diode test on the DMM?

As the transistors are small and difficult to hold down and test, I have ordered enough to replace them in this section and also of course T407.

They also have, should it be required, new psu's in stock at 102 Euros, this might be more than the complete scope is worth surely seeing as it about 20+ years old?

[tautech]

Great, good pointers from Sky that also point you to the SMPS FB circuitry that I mentioned in reply #77.
R462 out of spec will certainly bugger things up and all those divider resistors need be checked.

Good to know you can get another PSU if needed.