UK Vintage Radio Repair and Restoration Powered By Google Custom Search Vintage Radio and TV Service Data

Go Back   UK Vintage Radio Repair and Restoration Discussion Forum > Specific Vintage Equipment > Vintage Test Gear and Workshop Equipment

Notices

Vintage Test Gear and Workshop Equipment For discussions about vintage test gear and workshop equipment such as coil winders.

Closed Thread
 
Thread Tools
Old 3rd May 2020, 1:05 pm   #41
WME_bill
Octode
 
Join Date: Oct 2007
Location: Exeter, Devon, UK.
Posts: 1,553
Default Re: Cossor CDU150 rescue

Cossor CDU150 / CT531 rescue.
These dual capacitors failing comes up again. The solution proposed by Chris55000 is that which I have used.
I had this fault some years ago, see my posting under CDU110 of June 2011.
The CDU150 is a developed version of the CDU110, and uses the same power supply circuit.
I wasted hours until I discovered that it was a permanent short between the +500 and +50 elements, one was smoothing a positive rail and one the negative rail. So it gave a short across both rails. Reforming is the easy solution, and does seem to work, but for how long?
I replaced with a single 500uf in the can, and a separate 50uf tucked in a clip under the chassis.
And do this almost automatically for any CDU110 or CDU150 I see.
Very difficult fault to identify, as you have to look in the parts list to identify the multiple capacitors in the same can, and then the circuit diagram.
Well done MelJon66.
wme_bill
WME_bill is offline  
Old 21st May 2020, 8:46 am   #42
MelJon66
Pentode
 
Join Date: Apr 2020
Location: Somerset, UK.
Posts: 173
Default Re: Cossor CDU150 rescue

Greetings All,

Thanks to the generosity of Chris55000 my Cossor CDU150 lives again, without relying on my bench power supply to provide one voltage rail. It's taken me some time to get the latest repairs done as the good weather has diverted me onto my campervan project and gardening duties. It’s great to have the old girl back in service as I have grown very fond of her in the short time I have owned her, despite her quirks. She’s a bit like a classic British car; unreliable, requires lots of maintenance, can’t be trusted on long journeys but has an irresistible vintage charm.

As suggested by Chris I fitted new bridge rectifiers and PSU output caps before replacing the transformer. The 50uF sections of the dual can electrolytics have now been by-passed. Hopefully this will pre-empt any problems caused by shorts between the sections as documented by WME_Bill.

The donated transformer appears to be a better quality component than the original. Photos attached. The original has no maker’s name, just a QC stamp, but the new one is clearly marked as a Gresham unit. The Gresham has thicker, better insulated lead out wires and the secondary windings appear to be more evenly wound so it looks a far better product all round. The date on the Gresham is 1969 but the ‘scope appears to be early 70’s. Perhaps Cossor carried out a cost saving exercise at the end of the sixties and sub-contracted the transformer manufacture to a cheaper supplier or started winding them in-house.

There are still a few repairs to do, mostly on the EHT circuit to fix the fading trace issue and the focus control having to be set at the top of its range. Also, the trace sometimes flickers badly for no apparent reason. I am hoping that the EHT work will fix all this. In the meantime it is perfectly usable. I will report back as work progresses.

Thanks again to Chris.

Cheers
Mel
Attached Thumbnails
Click image for larger version

Name:	Original.jpg
Views:	106
Size:	43.1 KB
ID:	206356   Click image for larger version

Name:	Original_Windings.jpg
Views:	107
Size:	49.0 KB
ID:	206357   Click image for larger version

Name:	Gresham.jpg
Views:	94
Size:	38.3 KB
ID:	206358   Click image for larger version

Name:	Gresham_Windings.jpg
Views:	111
Size:	66.9 KB
ID:	206359  
MelJon66 is offline  
Old 29th May 2020, 8:26 pm   #43
MelJon66
Pentode
 
Join Date: Apr 2020
Location: Somerset, UK.
Posts: 173
Default Re: Cossor CDU150 rescue

Hi All,

I have been trying to understand the cause of the flickering trace I am getting much of the time on the CDU150. It’s hard to describe the symptoms as they vary a lot each time I switch on and it is very inconsistent.

These are the most repeatable symptoms:-

Both channels are affected but Ch2 seems worse. If there is no signal at the input sockets the flicker is likely to happen on any channel input selection (ground, a.c. or d.c.). When there is an alternating signal at the input sockets it only happens when the channel input is on the ground setting (Cascade on Ch1). Sine waves are steady on a.c. or d.c. input but on ground setting the flat line trace flickers badly. The flicker is like the input is in Alt mode, which it is not. Chop mode shows two steady traces so does not seem to be affected. The flickering does settle down after a long warm up (30 mins or so) but occurs intermittently after that.

When the flickering starts switching the trigger slope from +ve to –ve or the other way sometimes stops it. Similarly, switching from internal trigger source to external or line often stops the flickering.

I suspect a bad soldered joint, switch contact or grounding fault somewhere, possibly in the internal trigger circuits. Looking at the schematics for the trigger circuit reference C there are only a few components that are unique to the internal source selection, around TR12 and TR23 in the bottom left corner. This section takes take inputs from the trigger amplifier via circuit G pin 6, which itself takes input from circuit H2 pin 25 to circuit G pin 7. But, these circuits seem to work fine when an alternating signal is present so why are they struggling with a flat ground line?

I have been probing these circuits with my Iwatsu ‘scope. At pins 3 and 4 on circuit C I get clean sine traces on the Iwatsu but at ground input setting the Iwatsu trace is a steady line and the flickering on the Cossor CRT reduces. When I touch the Iwatsu probe to circuit G pin 7 or circuit H2 pin 25 the flickering on the Cossor stops completely and the Iwatsu shows a steady line. Even if I disconnect the probe from the Iwatsu and touch it against these pins it stops the flickering. Touching the probe ground lead to the pins stops the flickering but touching a screwdriver does not. Is this because the probe is introducing some capacitance at the pin?

Anyway, as a result I can’t fault find properly because probing immediately stops the flickering and I get no useful information.

Can anyone offer any suggestions on how I should proceed?

Cheers
Mel
MelJon66 is offline  
Old 30th May 2020, 8:27 am   #44
MelJon66
Pentode
 
Join Date: Apr 2020
Location: Somerset, UK.
Posts: 173
Default Re: Cossor CDU150 rescue

A bit more information to muddy the waters still further.

I know there is an issue with the focus on this 'scope which I plan to address when I do the EHT circuits. There are a couple of resistors B/R35 and B/R36 that have gone high so I will change them when I do the EHT work. In the meantime I get a sharp trace with Focus set to maximum and Astig set to about 40% rotation. The trace stays sharp when switching between a.c., d.c. or ground input coupling.

While fiddling with the controls to see what might stop the flickering problem with a grounded input I see that setting Astig to minimum and Focus to about 60% gets a sharp trace AND stops the flickering. Unfortunately when I then switch to a.c. or d.c. input with a sine wave signal the trace is really fuzzy so this is not a setting I would want to use normally.

Maybe I have been on the wrong track so far and the issue is with the geometry and/or focus circuits on PCB B. But, it puzzles me why different settings are required to get a sharp steady trace for a flat ground line or a sine wave?

Any thoughts?

Cheers
Mel
MelJon66 is offline  
Old 5th Jun 2020, 12:51 pm   #45
MelJon66
Pentode
 
Join Date: Apr 2020
Location: Somerset, UK.
Posts: 173
Default Re: Cossor CDU150 rescue

Hi All,

Success at last!

I have completed more work on the CDU150 and it is now running very well. Here is the most recent work and the results achieved at each stage.

1) Replaced the rest of the 500pf/8kV capacitors in the high voltage multiplier. I had already replaced one cracked one and there was nothing to indicate that the others were faulty but they were discoloured and looked suspect so the job needed to be done. Result – no change, but happy that future problems have been avoided.

2) Replaced the 500pf/8kV caps C2, C3 (Bright-up section) and C25 on board B. These are the same as those in the multiplier and were similarly discoloured so I replaced all of them. Result – improved trace brightness and less trace fade but still not as it should be.

3) Replaced 4700pf/3kV caps C12, C13 and C26 on board B (cathode supply?). Result – trace fade across the screen much improved. The first 1cm of trace at the start is still brighter than the rest but it’s even across the rest of the trace.

4) Replaced Zener diodes D13 (68V) & D14 (82V) on board B. Voltage drop across D13 - 65V before and 66.7V after. Voltage drop across D14 - 43V before and 43V after. Results - the voltage drops didn’t change much but the trace fade has gone . When the brightness control is increased the whole trace brightness increases together and not just gradually from left to right. Also, the trace brightness is much more even at all timebase settings. Trace brightness varied a lot before when I adjusted the timebase so this was an unexpected improvement.

5) Replaced R35 and R36 on board B (focus resistors). These should both be 680k but when I first measured them a few weeks back R35 was 915k and R36 was 970k (in circuit). A few days ago R35 was 917k and R36 was 2.1 Meg (in circuit). When taken out of circuit while replacing them R35 measured 1Meg and R36 measured 2.8Meg. This ‘scope had been unused for many years so R36 must have been failing fast now that it was back in service. After the resistors were replaced the voltage drops across the Zeners were 66.4V for D13 and 78.7V for D14 so changing the resistors has brought things back to spec. The beam current can now be set correctly as per the manual. Results – focus is now sharp in the centre of focus pot range not at the extreme end.

None of this work has resolved the trace flickering with DC only or grounded inputs but I think time spent using the machine has revealed the likely cause. With the covers off for testing and repairs the CDU150 seems prone to interference. With the rear (EHT) cover off saw tooth interference is clearly visible on the trace, which should be a flat line (see photo). This isn’t visible when the covers are fitted but I suspect there is enough interference remaining for the triggering system to detect some rising or falling edges. Changing the trigger slope +/- stops the flickering but which way to go depends on how hot the ‘scope is. Selecting AC LF reject trigger coupling usually stop stops it and DC trigger coupling always stops it, which I think is the giveaway. Maybe I am expecting too much and this is normal behaviour on these old Cossor ‘s but DC traces don’t flicker on my Iwatsu when trigger coupling is left in AC mode. Anyway, it’s no big deal and it’s easy to deal with when you know why it’s happening.

The great news is that the trace on the Cossor is probably slightly better than my Iwatsu SS-5711 which is an excellent ‘scope itself. I think the Cossor’s dark green screen background helps as there is more contrast and the focus appears sharper. Overall it is a nice machine to use and the time I have spent on it gives me a strong sense of ownership. The vintage military look is great and it looks right in my work space next to my aged hi-fi gear. It’ll keep me warm on winter nights too as this thing does get pretty hot after an hours use.

The only outstanding work is to finish off calibration, replace one very stiff vertical sensitivity pot and replace the selenium rectifiers (no rush).

So, I think I can safely say that this is another Cossor CDU150 successfully rescued from the skip!

Many thanks to Chris55000 in particular for his expert knowledge and generosity in supplying parts and to other Cossor enthusiasts such as WME_bill, whose posts put me in the right direction on many occasions.

My next project is yet to be decided. I have a HP1743A ‘scope, Heathkit 10-12U valve ‘scope, Tektronix 543 valve ‘scope, Hitachi HA5300 amplifier, Leak TL12+ valve amplifier, Rotel RA-412 amplifier and a Texan T20 amplifier all jostling for first place. Oh, and don’t forget the camper van, new garden fence, garden gazebo………….. I never did believe all those people who warned of boredom in retirement. I wish I could have retired at 21.

Cheers
Mel
Attached Thumbnails
Click image for larger version

Name:	Trace_Interference.jpg
Views:	120
Size:	59.6 KB
ID:	207732  
MelJon66 is offline  
Old 5th Jun 2020, 7:56 pm   #46
MotorBikeLes
Nonode
 
Join Date: Dec 2012
Location: Kirk Michael, Isle of Man
Posts: 2,346
Default Re: Cossor CDU150 rescue

I don't know what the fine details of the CDU150 are, but that wave forn on the screen looks just like the waveform seen when setting up the trigger circuit on some TQ scopes, where there is no hold off period. There should be the sweep period, the flyback, then a hold off period before the next scan.
It may be nothing to do with your problem, but it is a picture I recognise in the TQ context.
Les.
MotorBikeLes is offline  
Old 6th Jun 2020, 10:52 am   #47
MelJon66
Pentode
 
Join Date: Apr 2020
Location: Somerset, UK.
Posts: 173
Default Re: Cossor CDU150 rescue

Thanks for the input Les,

The timebase and trigger circuits are an area I haven't got into yet so I will deal with the more obvious stuff first.

Another piece of the jigsaw worth mentioning is that these 'scopes have a cascade input selection on Channel 1 which then works in series with Channel 2 to give more vertical sensitivity. In this mode I can see interference hash thickening the DC trace at most vertical sensitivities except a few that are multiples of 5. So, 5 mv/cm, .5 v/cm and 5v/cm display no hash but the rest do. The 5 mv/cm sensitivity is a straight wire connection through the attentuator so I can understand why that isn't picking anything up. I suspect there may be one or more attenuator resistance/capacitance circuits that are picking up the interference and they could be circuits that the .5 v/cm and 5 v/cm selections are not routed through. I am not going to strip down the attenuators again just for this but I will tweak the attentuator capacitors (accessible from the outside) to see if it has any effect. If it does I might dive in and have a closer look.

Cheers
Mel
MelJon66 is offline  
Old 7th Jun 2020, 11:03 am   #48
Chris55000
Nonode
 
Join Date: Aug 2007
Location: Walsall Wood, Aldridge, Walsall, UK.
Posts: 2,853
Default Re: Cossor CDU150 rescue

Hi!

That reminds me – I forgot about sending the dual pot I promised you Mel – my apologies and I'll try and remember to get it sorted this week!

Chris Williams
__________________
It's an enigma, that's what it is! This thing's not fixed because it doesn't want to be fixed!
Chris55000 is offline  
Old 7th Jun 2020, 3:14 pm   #49
WME_bill
Octode
 
Join Date: Oct 2007
Location: Exeter, Devon, UK.
Posts: 1,553
Default Re: Cossor CDU150 rescue

CDU150
Mel, Just a point about the hash interference you are getting. Have you checked the frequency. I had a dry joint on the screening to cables from the EHT oscillator, which resulted in a hash at that frequency appearing everywhere. The fault was at points 9 and 11 to 10 on the PCB board B, EHT power. It spread to all the Y Amplifier and the Trigger inputs.
wme_Bill.
WME_bill is offline  
Old 8th Jun 2020, 1:36 pm   #50
MelJon66
Pentode
 
Join Date: Apr 2020
Location: Somerset, UK.
Posts: 173
Default Re: Cossor CDU150 rescue

Hi All,

Thanks for the input Bill.

In the photo I posted it's a distorted square wave at 2000Hz viewed on Channel 2. That made me suspicious as that’s the same frequency as the sine wave my signal generator was inputting into Channel 1 so I have done some further checking. It definitely is not hash put out by the signal generator as I have two and it happens with both of them. It is not specific to one channel either as it happens either way around. I will do my best to explain what I am seeing.

When viewing a DC signal or ground on one channel and there is an alternating signal on the other channel, interference appears on the DC channel being viewed at the same frequency as the input on the other channel. I have changed the frequency of the signal generator and the frequency of the interference on the other channel changes to that frequency. The interference is always a square wave, whatever shape waveform is input to the other channel.

With high vertical sensitivities selected on both channels and with the viewed channel set on AC or DC input the interference is very obvious, as can be seen in the photo I posted earlier, as the ‘scope can trigger on it easily. When the sensitivity of the non-viewed channel is decreased the interference on the viewed channel reduces in amplitude to the point where the ‘scope can’t trigger on it reliably and the flickering starts. If the viewed channel is set to ground the interference is of lower amplitude but is still there so the same problem occurs. For some reason the 5 mv/cm, 50 mv/cm, .5 v/cm and 5 v/cm sensitivities seem to suffer less interference with a grounded input.

I think this shows that there is some form of cross talk between the channels, possibly somewhere in the vertical pre-amp section. Both channels can be viewed at the same time in Alt and Chop mode but there must be something in the circuits that maintains channel separation when only one channel is viewed. I have looked at the vertical pre-amp schematics (Board H) and I wonder if diodes D8, D11, D25 and D28 are involved in this? I will be taking out the vertical pre-amp module fairly soon to replace a position potentiometer so that will gave me an opportunity to probe around and see if I can work out where the interference is being picked up.

In the meantime if anyone has any ideas that can put me in the right direction that will be very much appreciated.

Cheers
Mel
MelJon66 is offline  
Old 8th Jun 2020, 2:50 pm   #51
Chris55000
Nonode
 
Join Date: Aug 2007
Location: Walsall Wood, Aldridge, Walsall, UK.
Posts: 2,853
Default Re: Cossor CDU150 rescue

Hi!

Crosstalk between one vertical channel and another in an Oscilloscope is usually due to deficiencies of the decoupling or earthing of the input amplifiers, or simply direct pickup between Ch 1 and. Ch 2 due to incorrectly dressed or routed wiring between PCBs H1 and J1 and PCB H2 and J2, the simple way to see if there's grot in the amplifiers is to shunt R4 on PCB H1 with a 100 ohm resistor, and R73 on PCB H2 with another 100 ohm resistor – this should produce a flat smooth line on both channels.

If you still get disturbances on the traces with R4 and R73 shunted by 100 ohm resistors,. the disturbances are occuring later on in the amplifiers and you'll need another Oscilloscope to try and locate them.

If you do get smooth flat lines with R4 and R73 shunted, the amplifier circuits should be ok and you'll need to look at PCB Assemblies J1 and J2 carefully to make sure all the earthing and switching is correct and in good condition, and there's no burnt–out or damaged resistors in the attenuator circuits J1 and J2 that can leave an FET gate floating!

Try shunting R4 and R73 first on PCBs H1 & H2 with 100 ohm resistors to try and identify where your breakthrough is occuring and if that's inconclusive, I'll tell you what to try next!

Chris t
__________________
It's an enigma, that's what it is! This thing's not fixed because it doesn't want to be fixed!
Chris55000 is offline  
Old 8th Jun 2020, 3:47 pm   #52
MelJon66
Pentode
 
Join Date: Apr 2020
Location: Somerset, UK.
Posts: 173
Default Re: Cossor CDU150 rescue

Hi Chris,

Thanks for the input. I will try what you suggest.

The only grounding on the attenuator / vertical amp module appears to be through the fixing screws. When I had the unit out and connected via flying leads it didn't work well until I connected a jumper lead between the module and the case to ground it. Also, I remember that when I touched the case it steadied the trace. I'll clean up the threads on all the mounting screws to make sure there is a sound connection.

Cheers
Mel
MelJon66 is offline  
Old 15th Jun 2020, 3:15 pm   #53
MelJon66
Pentode
 
Join Date: Apr 2020
Location: Somerset, UK.
Posts: 173
Default Re: Cossor CDU150 rescue

Hi Chris,

It was my wife's birthday last week so I have had to avoid any tasks that could be construed as "time wasting".

Even now I have had to cheat a bit. It turns out that R4 on Ch1 is accessible from the side of the case so I have shunted that with a 100R resistor on clip leads. It's difficult to get to R73 on Ch2 without disassembly. I hope this provides some useful information.

So, with R4 shunted and a 20v P2P sine wave input in to Ch1 what I see on the CRT for Ch1 varies with sensitivity setting. At 5 mv/cm the sine wave is large and off screen. At 10, 20 & 50 mv/cm I see a reduced sine wave (less than 1 cm high), reducing further in amplitude as the sensitivity is reduced. At 0.1 mv/cm and above I get a fuzzy flat line, except at 0.5 v/cm and 5 v/cm where I see a clear flattened sine wave around 0.5 cm high. As I have seen before the 5 mv/cm, 0.5 v/cm and 5 v/cm settings seem to behave differently to the rest.

With R4 shunted and Ch1 set to 5 mv/cm and Ch2 set to ground the square wave interference on Ch2 is still there at all Ch2 sensitivity settings. At all other Ch1 sensitivity settings the DC trace on Ch2 is flat. There is some flickering of the Ch2 trace but much less than without the shunt. One thing to note is that the amplitude of the interference square wave doesn't change much no matter what the Ch2 sensitivity is set to. It's about 4mm P2P for most settings but it is slightly less at around 3mm P2P at the 5 mv/cm, 0.5 v/cm and 5 v/cm settings.

Does this tell you anything?

Cheers
Mel
MelJon66 is offline  
Old 16th Jun 2020, 7:38 am   #54
Chris55000
Nonode
 
Join Date: Aug 2007
Location: Walsall Wood, Aldridge, Walsall, UK.
Posts: 2,853
Default Re: Cossor CDU150 rescue

Morning Mel!

I wrote a reply out ready for you last nite and my 'fone deleted it!!

You symptoms possibly suggest breakthrough from the chop multivibrator is getting into the ch2 amplifiers and the way to fault–find this is as follows:–

Disconnect the two leads connected to pins 1 and 2 of PCB H2 to isolate the chop multivibrator from the amplifier, making a note of where they were connected originally.

If the spurious squarewave is still visible, then it could be breaking through from the Timebases or horizontal amplifiers, and in this case, you'll need to check the earthing, supply–decoupling and wiring to boards C, D, E and F.

If the interference disappears when you disconnect leads from pin 1 and 2 of PCB H2, either the chop multivibrator is defective or the gating diodes/switching transistors in ch2 need looking at!

Temporarily reverse the connections to pins 1 and 2 of PCB H2 and see if the interference transfers over to Ch 1 – if it does, there is something amiss with the chop multivibrator TR30 and TR31 on PCB H2 and I recommend using another 'scope to trace this to avoid misleading cross–coupling effects.

If the interference remains on ch2 whichever way round the leads from 1 and 2 on PCB H2 are connected, then investigate the isolating/steering diodes D24 to D28 and the switching transistors TR28 and TR29.

Also make sure, of course, that you've not got any yakky switch–contacts on the wafer 3 (both sides) of the CH1/CH2/alt/chop, etc., selector switch SW3 – it's a fairly complex interlinked arrangement of switch contacts, but reference to a blown–up printout of that section of the circuit and a decent d.v.m. with a good reliable continuity range should help you sort out any switching snags!

Good Luck!

Chris Williams
__________________
It's an enigma, that's what it is! This thing's not fixed because it doesn't want to be fixed!

Last edited by Chris55000; 16th Jun 2020 at 7:54 am.
Chris55000 is offline  
Old 16th Jun 2020, 3:41 pm   #55
MelJon66
Pentode
 
Join Date: Apr 2020
Location: Somerset, UK.
Posts: 173
Default Re: Cossor CDU150 rescue

Hi Chris,

With the leads to pins 1 & 2 on board B disconnected and a sine wave input to either channel the signal shows up on both channels at the same amplitude, as if they are operating in parallel. The sensitivity control on the channel without input has no effect and you see the full amplitude sine wave on all settings. Altering the sensitivity on the channel with input reduces the amplitude on the other channel. Even with grounded input on Ch2 the full sine wave is still carried over from the Ch1 input so I can't tell if any square wave interference is present. With no input to Ch1 and grounded input on Ch2 I see a flickering flat line on Ch2.

I haven't swapped pin 1 & 2 inputs around as the interference works both ways anyway. A sine wave input on Ch2 produces square wave interference on Ch1 the same as it does the other way around.

The only visible sign of anything awry on board H2 is that capacitor C40 is leaking some sort of crystalline powder. It has green shrink wrap type cover and it is leaking from one end. It's connected between ground and the -50 v rail going to switch 3. The parts list says it's a PE type (polyester?) 250v and 0.1 ? with no units. It's probably not related to the problem but it's the only visible sign of trouble.

Does that help at all?

Cheers
Mel
MelJon66 is offline  
Old 17th Jun 2020, 3:10 pm   #56
Chris55000
Nonode
 
Join Date: Aug 2007
Location: Walsall Wood, Aldridge, Walsall, UK.
Posts: 2,853
Default Re: Cossor CDU150 rescue

Afternoon Mel!

C40 decouples the –50V line directly going to the chop multivibrator via Wafer 1 on SW3 if it's not doing it's job properly the chop waveform'll be liable to break out everywhere!

Before you go any further with fault location on this Oscilloscope, replace C31, C38, C40, C42, C43 and C44 on PCB H2 first to make sure the supply–lines on H2 are as quiet as they possibly can be – other decoupling capacitors in your Oscilloscope may well be past their best, and when you find lots of them dotted all over the circuit diagram like this example, you can be certain they'll all be needed to do their job for quiet and stable performance of the circuits!

I also recommend you replace C2, C3, C8, C12, C18, C19, C20 and C29 on PCB H1 as well!

All the capacitors on Cossor circuit diagrams are quoted in "microfarads" unless "n" or "p" has been specifically appended – my practice is to normally write 0.1 microfarads as "0u1" when I do a new diagram for anything, but BRC/Thorn and many other British makers, like this one, simply wrote "0.1".

Chris Williams
__________________
It's an enigma, that's what it is! This thing's not fixed because it doesn't want to be fixed!

Last edited by Chris55000; 17th Jun 2020 at 3:29 pm.
Chris55000 is offline  
Old 17th Jun 2020, 8:47 pm   #57
MelJon66
Pentode
 
Join Date: Apr 2020
Location: Somerset, UK.
Posts: 173
Default Re: Cossor CDU150 rescue

Thanks Chris,

A closer look reveals that R144 on board H2, on the -50v rail close to the leaking C40, is burnt on the underside so there has obviously been something going on with the -50v rail. It still measures ok but must be replaced.

I have ordered the parts to get the work done on H2. I have plenty of 1/4W resistors but the ones that fail always seem to be 1/2W, which I don't have. I suppose they are the ones that take the current and heat so it's not surprising.

There are a lot of ceramic caps required so I will total it all up and see if I can buy a batch.

Cheers
Mel
MelJon66 is offline  
Old 22nd Jun 2020, 7:57 pm   #58
MelJon66
Pentode
 
Join Date: Apr 2020
Location: Somerset, UK.
Posts: 173
Default Re: Cossor CDU150 rescue

Hi Chris and all,

Quick update,

I have replaced burnt resistor R144, leaky capacitor C40 and suspect caps C42 & C44. Regrettably, no change, the crossover interference remains. The tantalum for C43 and the various ceramics have yet to arrive.

Just out of interest I disconnected the feed from the attentuators to board H1 at pins 18 and 19, expecting to get no signal. Despite this I still get a signal at high vertical sensitivities. Photo attached of the screen with 5 mv/cm sensitivity selected on Ch1 with pin 18 disconnected and a sine wave input into the BNC connector. A square wave input produces a flat square wave with masses of overshoot. I triangle wave produces a lovely square wave (why?). I assume this is because some signal is still getting across the gap between the pin and the input wire and gets severely distorted as a result.

With one channel input pin connected the crossover interference is still visible on the other channel. I assume this confirms that the crossover is taking place after the attentuators, in the amplifier or chop multi-vibrator as Chris has suggested.

No further progress at present but all ideas and contributions welcome.

Cheers
Mel
Attached Thumbnails
Click image for larger version

Name:	Pin18_Disconnected.jpg
Views:	76
Size:	75.2 KB
ID:	209253  
MelJon66 is offline  
Old 25th Jun 2020, 1:17 pm   #59
MelJon66
Pentode
 
Join Date: Apr 2020
Location: Somerset, UK.
Posts: 173
Default Re: Cossor CDU150 rescue

Hi All,

The good news is that I have fitted the tantalum capacitor C43 and it has stopped the channel crosstalk from Ch1 to Ch2 when Ch2 is grounded. As far as I can tell it has also stopped crosstalk from Ch2 to Ch1 when Ch1 input is shorted (It has no ground selection) but it's a fuzzy line probably due to the length of the shorting leads. The crosstalk is still present on AC or DC input settings. Does anyone know why this would affect grounded inputs only?

While I was waiting for parts I did some fault finding with my Iwatsu 'scope and I am now certain that the crosstalk square wave I am seeing is from the other channel and not related to the chop multi-vibrator signal breaking through. The chop multi-vibrator seems to operate at a fixed frequency and doesn't change with input frequency but the crosstalk does. With a large input signal and high input sensitivity such that the trace extends off the screen the signal through the pre-amp is heavily clipped, resembling a square wave that varies with input frequency. This is what is being picked up on the other channel. Sine, triangle and square waves all look like square waves when they are so badly clipped. If this was an audio amp the distortion would be horrendous so I guess this distorted signal is what is breaking through to the other channel.

When the channel input signal is not clipped the crosstalk on the other channel is at a very low level, just a tiny ripple on a DC trace, but is just enough to fool the trigger circuit to try and trigger on it, hence the flickering. So, if the sensitivity of an AC signal is adjusted so that the trace is within the height of the screen, which it would be in normal usage, the crosstalk on the other channel is minimal. This is fine if both inputs are AC, but a DC trace does tend to flicker unless DC trigger coupling is selected, which stops it. It turns out my Iwatsu scope does the same thing, but the crosstalk is much less and the trace on the channel being viewed does not flicker so it's not noticeable. Maybe my expectations for the Cossor are not realistic.

I still have a batch of ceramic capacitors to replace and hopefully that will improve things still further.

Cheers
Mel
MelJon66 is offline  
Old 25th Jun 2020, 3:52 pm   #60
MelJon66
Pentode
 
Join Date: Apr 2020
Location: Somerset, UK.
Posts: 173
Default Re: Cossor CDU150 rescue

Hi Chris55000 and all,

C43 is a 6.8uF 25V tantalum decoupling capacitor on the +12.6 V supply rail on board H2 (feeding board H1 as well). It measured 7.1uF and ESR was .73 ohms. I believe tantalums are low ESR devices so it seems to measure well but was obviously faulty. It was operating well below its rated voltage and showed no signs of heat.

What failure mode would cause this sort of problem?

Cheers
Mel
MelJon66 is offline  
Closed Thread

Thread Tools



All times are GMT +1. The time now is 9:21 am.


All information and advice on this forum is subject to the WARNING AND DISCLAIMER located at https://www.vintage-radio.net/rules.html.
Failure to heed this warning may result in death or serious injury to yourself and/or others.


Powered by vBulletin®
Copyright ©2000 - 2024, vBulletin Solutions, Inc.
Copyright ©2002 - 2023, Paul Stenning.