Topics

let's bring this back to life Re: Tektronix 453 high voltage problem


william_b_noble
 

I decided to dig into an early 453. Interestingly it does not match the schematics for the 453 or the 453A - some boards seem closer to the A, some to the earlier one - for example the Z axis board is clearly dated (C) 1965, but its layout matches the board found in the A manual (unless I'm just loosing my mind). This scope sat around for several years after I got it and never plugged it in - when I did plug it in there was initially no light on the CRT (except the scale illuminator) - as I messed with it for about an hour trying to make sure what was wrong, it fired up and showed a spot and vertical deflection, I could center it, but there was no trace. then it stopped. The HV circuit is one of those that will stop oscillating if there is an overload, so it's hard to troubleshoot, and harder without an HV probe. But it's clear that something is overloading the circuit. If I disconnect the cathode and grid bias circuits, it makes HV and powers the CRT filament just fine. I've tried all sorts of things, one at a time, and disconnecting the .015uf 3Kv capacitors by itself doesn't work, disconnecting the diode that drives the grid string does let it work, and the list goes on. I've come to the conclusion without a good way to verify it that it is most likely that at least one of those 3KV ceramic caps is leaky, and if one is bad, the others are probably not far behind, so I've ordered some. The HV diodes are also suspect, though they both seem to act like diodes at 115V (with a 10 W lightbulb as the load), I ordered some microwave oven diodes just to be on the safe side. I don't think my 1M CRT Grid Bias control is bad like was found upthread, when the power supply works, I can adjust the grid bias. But now, if the supply oscillates the current can rise to about 2.9 amps (it's fused at 2 amps) with all the loads connected. The transformer isn't potted, so I think it's probably not the issue.


anyone care to offer some advise on something I could have missed? I'd like to get this scope working just to say that I did it - I have a couple of newer ones to look at next, I figured this one first, it's all discretes, how hard could that be >>>


 

I believe microwave oven HV diodes will be too slow to use for the HV diodes in a scope. The scope HV inverter runs at 20 kHz or more. The oven diodes are used at 60 Hz and their reverse recovery time may be way too slow.

Regards

----- Original Message -----
From: "william_b_noble" <william_b_noble@...>
To: <TekScopes@groups.io>
Sent: Monday, June 11, 2018 1:50 AM
Subject: [TekScopes] let's bring this back to life Re: Tektronix 453 high voltage problem


I decided to dig into an early 453. Interestingly it does not match the schematics for the 453 or the 453A - some boards seem closer to the A, some to the earlier one - for example the Z axis board is clearly dated (C) 1965, but its layout matches the board found in the A manual (unless I'm just loosing my mind). This scope sat around for several years after I got it and never plugged it in - when I did plug it in there was initially no light on the CRT (except the scale illuminator) - as I messed with it for about an hour trying to make sure what was wrong, it fired up and showed a spot and vertical deflection, I could center it, but there was no trace. then it stopped. The HV circuit is one of those that will stop oscillating if there is an overload, so it's hard to troubleshoot, and harder without an HV probe. But it's clear that something is overloading the circuit. If I disconnect the cathode and grid bias circuits, it makes HV and powers the CRT filament just fine. I've tried all sorts of things, one at a time, and disconnecting the .015uf 3Kv capacitors by itself doesn't work, disconnecting the diode that drives the grid string does let it work, and the list goes on. I've come to the conclusion without a good way to verify it that it is most likely that at least one of those 3KV ceramic caps is leaky, and if one is bad, the others are probably not far behind, so I've ordered some. The HV diodes are also suspect, though they both seem to act like diodes at 115V (with a 10 W lightbulb as the load), I ordered some microwave oven diodes just to be on the safe side. I don't think my 1M CRT Grid Bias control is bad like was found upthread, when the power supply works, I can adjust the grid bias. But now, if the supply oscillates the current can rise to about 2.9 amps (it's fused at 2 amps) with all the loads connected. The transformer isn't potted, so I think it's probably not the issue.


anyone care to offer some advise on something I could have missed? I'd like to get this scope working just to say that I did it - I have a couple of newer ones to look at next, I figured this one first, it's all discretes, how hard could that be >>>


ArtekManuals
 

Bill

1) (C) dates have no real bearing on date of manufacture they can be
years apart or even decades apart.
2) Unless you know or have access to the repair history for your unit,
anything is possible as relates to actual board versions in your unit.
If it was repaired at some point in the past by an "Authorized Service
Center" it can have a "later" board installed as part of the repair, by
the early 80's there was already little incentive for these centers to
repair to the component level, swap a board instead, takes less time and
generates more revenue. If it was repaired by some "entrepreneur" using
boards from  an "organ donor - parts scope" the board in question can be either or later than the board it left the factory with.
3) Coming to life after an hour can be a function of some electrolytic
capacitor forming up after time
4) Replacing caps can usually never hurt unless someone gets sloppy with
a soldering iron, after all this scope is getting close to 50 years old.
5) Does the beam finder work at all?
6) Are all the LV power Supply voltages in spec ?
7) Why do you suspect HV diodes?

-DC
manuals@...

On 6/11/2018 1:50 AM, william_b_noble wrote:
I decided to dig into an early 453. Interestingly it does not match the schematics for the 453 or the 453A - some boards seem closer to the A, some to the earlier one - for example the Z axis board is clearly dated (C) 1965, but its layout matches the board found in the A manual (unless I'm just loosing my mind). This scope sat around for several years after I got it and never plugged it in - when I did plug it in there was initially no light on the CRT (except the scale illuminator) - as I messed with it for about an hour trying to make sure what was wrong, it fired up and showed a spot and vertical deflection, I could center it, but there was no trace. then it stopped. The HV circuit is one of those that will stop oscillating if there is an overload, so it's hard to troubleshoot, and harder without an HV probe. But it's clear that something is overloading the circuit. If I disconnect the cathode and grid bias circuits, it makes HV and powers the CRT filament just fine. I've tried all sorts of things, one at a time, and disconnecting the .015uf 3Kv capacitors by itself doesn't work, disconnecting the diode that drives the grid string does let it work, and the list goes on. I've come to the conclusion without a good way to verify it that it is most likely that at least one of those 3KV ceramic caps is leaky, and if one is bad, the others are probably not far behind, so I've ordered some. The HV diodes are also suspect, though they both seem to act like diodes at 115V (with a 10 W lightbulb as the load), I ordered some microwave oven diodes just to be on the safe side. I don't think my 1M CRT Grid Bias control is bad like was found upthread, when the power supply works, I can adjust the grid bias. But now, if the supply oscillates the current can rise to about 2.9 amps (it's fused at 2 amps) with all the loads connected. The transformer isn't potted, so I think it's probably not the issue.


anyone care to offer some advise on something I could have missed? I'd like to get this scope working just to say that I did it - I have a couple of newer ones to look at next, I figured this one first, it's all discretes, how hard could that be >>>

--
Dave
Manuals@...
www.ArtekManuals.com


william_b_noble
 

Ok, more info. I have no repair history, I got it for scrap price from an e-waste guy who used to sell at a,small flea market. It had a nice manual with it, but for a very different scope.

I have tested the low voltage supplies, they were all close, they are now within a millivolt of their proper value. I tested all the carbon resistors in the hv divider networks, and all but one were within tolerance, the 150k resistor in series with the grid adjust pot had dropped to 30k, so I changed it.

I don't really suspect the diodes, but it pretty much has to be a leaky ceramic cap or two, or a diode, right?

I also suspect some issue on the z axis board, for now, I disconnected it

Thanks


RustySparks
 

Are you sure this is not the typical leaky HV transformer winding problem that seems to affect practically all the 453 series scopes and other similar models simply due to the transformer HV winding becoming leaky because of chemical changes in the epoxy used in transformer assembly over the years?
My 453A has pretty much the same problem with HV, initially it would stop oscillating after being on for 30 minutes or so, got progressively worse after a year or so, eventually wouldn't start at all. It was through this forum that I learned what the problem was. At least one member here has corrected his by rewinding the HV transformer secondary. Pretty amazing project!

-----Original Message-----
From: TekScopes@groups.io <TekScopes@groups.io> On Behalf Of william_b_noble
Sent: Monday, June 11, 2018 1:51 AM
To: TekScopes@groups.io
Subject: [TekScopes] let's bring this back to life Re: Tektronix 453 high voltage problem

I decided to dig into an early 453. Interestingly it does not match the schematics for the 453 or the 453A - some boards seem closer to the A, some to the earlier one - for example the Z axis board is clearly dated (C) 1965, but its layout matches the board found in the A manual (unless I'm just loosing my mind). This scope sat around for several years after I got it and never plugged it in - when I did plug it in there was initially no light on the CRT (except the scale illuminator) - as I messed with it for about an hour trying to make sure what was wrong, it fired up and showed a spot and vertical deflection, I could center it, but there was no trace. then it stopped. The HV circuit is one of those that will stop oscillating if there is an overload, so it's hard to troubleshoot, and harder without an HV probe. But it's clear that something is overloading the circuit. If I disconnect the cathode and grid bias circuits, it makes HV and powers the CRT filament just fine. I've tried all sorts of things, one at a time, and disconnecting the .015uf 3Kv capacitors by itself doesn't work, disconnecting the diode that drives the grid string does let it work, and the list goes on. I've come to the conclusion without a good way to verify it that it is most likely that at least one of those 3KV ceramic caps is leaky, and if one is bad, the others are probably not far behind, so I've ordered some. The HV diodes are also suspect, though they both seem to act like diodes at 115V (with a 10 W lightbulb as the load), I ordered some microwave oven diodes just to be on the safe side. I don't think my 1M CRT Grid Bias control is bad like was found upthread, when the power supply works, I can adjust the grid bias. But now, if the supply oscillates the current can rise to about 2.9 amps (it's fused at 2 amps) with all the loads connected. The transformer isn't potted, so I think it's probably not the issue.


anyone care to offer some advise on something I could have missed? I'd like to get this scope working just to say that I did it - I have a couple of newer ones to look at next, I figured this one first, it's all discretes, how hard could that be >>>


Phillip Potter
 

I have a late serial number 453, that seems to have a combination of boards, seemingly like yours. To your advantage, I submitted my manual for scanning, so you might want to get ahold of Toby, a member, to see where that scan is parked and then download it. It might just be the ticket for your scope...

Phil

Phil, N6OMM

On Jun 10, 2018, at 11:50 PM, william_b_noble <william_b_noble@...> wrote:

I decided to dig into an early 453. Interestingly it does not match the schematics for the 453 or the 453A - some boards seem closer to the A, some to the earlier one - for example the Z axis board is clearly dated (C) 1965, but its layout matches the board found in the A manual (unless I'm just loosing my mind).


toby@...
 

On 2018-06-11 12:21 PM, Phillip Potter wrote:
I have a late serial number 453, that seems to have a combination of boards, seemingly like yours. To your advantage, I submitted my manual for scanning, so you might want to get ahold of Toby, a member, to see where that scan is parked and then download it. It might just be the ticket for your scope...
I'm so glad to hear about your rescue, William! I also obtained a 453
rescued from a potentially terrible fate.

Phil's 453 manual has been digitised here:
http://chiclassiccomp.org/docs/index.php?dir=%2Ftesteq/Tektronix

I hope this helps.

--Toby

Phil

Phil, N6OMM

On Jun 10, 2018, at 11:50 PM, william_b_noble <william_b_noble@...> wrote:

I decided to dig into an early 453. Interestingly it does not match the schematics for the 453 or the 453A - some boards seem closer to the A, some to the earlier one - for example the Z axis board is clearly dated (C) 1965, but its layout matches the board found in the A manual (unless I'm just loosing my mind).




Phillip Potter
 

Awesome attention to detail, Toby! You rock! Thanks for the link, as I was able to download it to my iPad for safe keeping. I hope this helps William, too!

Phil, N6OMM

On Jun 11, 2018, at 11:09 AM, toby@... wrote:

Phil's 453 manual has been digitised here:
http://chiclassiccomp.org/docs/index.php?dir=%2Ftesteq/Tektronix

I hope this helps.

--Toby


william_b_noble
 

The reason I don't think that the transformer epoxy is going bad is because the transformer is not potted at all. I can see the windings. My understanding is that later transformers were potted. I have the early style supply with the vacuum tubes for the 12kV high voltage doubler.


Dave Wise
 

A winding can impregnated with epoxy without being potted in it. I restored my early 453 by rewinding, and it wasn't potted. See topic and photo album titled "453 HV Transformer Rewind".

Dave Wise
________________________________________
From: TekScopes@groups.io <TekScopes@groups.io> on behalf of william_b_noble <william_b_noble@...>
Sent: Monday, June 11, 2018 11:55 AM
To: TekScopes@groups.io
Subject: Re: [TekScopes] let's bring this back to life Re: Tektronix 453 high voltage problem

The reason I don't think that the transformer epoxy is going bad is because the transformer is not potted at all. I can see the windings. My understanding is that later transformers were potted. I have the early style supply with the vacuum tubes for the 12kV high voltage doubler.


Phillip Potter
 

Wow, then I will guess that my manual isn’t appropriate, sorry!

Phil, N6OMM

On Jun 11, 2018, at 12:55 PM, william_b_noble <william_b_noble@...> wrote:

I have the early style supply with the vacuum tubes for the 12kV high voltage doubler.


RustySparks
 

Yes, my 453A also has vacuum tube rectifiers, and a non-potted HV transformer. I really went through the whole HV supply until there simply was nothing else to blame except the transformer, before finally giving up to the very unwelcome realization that the HV transformer was bad in the scope I had used since it was new. Very sad indeed.

-----Original Message-----
From: TekScopes@groups.io <TekScopes@groups.io> On Behalf Of william_b_noble
Sent: Monday, June 11, 2018 2:55 PM
To: TekScopes@groups.io
Subject: Re: [TekScopes] let's bring this back to life Re: Tektronix 453 high voltage problem

The reason I don't think that the transformer epoxy is going bad is because the transformer is not potted at all. I can see the windings. My understanding is that later transformers were potted. I have the early style supply with the vacuum tubes for the 12kV high voltage doubler.


John
 

I've tackled a couple of 453 EHT problems. In one case (actually a 454) it was the final HV capacitor (before the CRT anode) which had broken down. In the other instance, it was a thermally-related runaway problem with the transformer. It would take increasing current as it heated up until the -12V line was affected. A third 'scope had a gassy tube, which resulted in a purple glow in the neck and excessive EHT current. You can elimiate this from your enquiries by (carefully! discharge first) lifting the anode output lead and re-powering. Is your scope fitted with 5642 tubes or silicon rectifiers?


John
 

Sorry, for -12V read +12V above.


ArtekManuals
 

NOTHING TO READ ABOVE ????

On 6/11/2018 5:52 PM, John wrote:
Sorry, for -12V read +12V above.



--
Dave
Manuals@...
www.ArtekManuals.com


william_b_noble
 

my scope has the vacuum tube rectifiers - I am pretty sure the CRT is OK because I have gotten it to run for a bit and it makes a fine display. For now, I'm inclined to blame caps - specifically the ones in the grid circuits with the silicon diodes, not the HV ones. I disconnected the grid circuits by removing the diodes, and the HV supply made HV and the current drawn through the oscillator transistor was well under an amp (sorry, I didn't write down the number). There are some orange colored rectangular axial leaded ceramic caps, .015 at 2.5KV. I've ordered some .03 ceramic caps and I just got a bag of .01 ceramic caps, both rated 3KV (so safely more than the original ones). they are quite a bit smaller though - I'm inclined to say that they have newer ceramic with better dielectric but I worry that there is some other parameter that isn't specified that the original caps meet and these don't - I suppose the "easy" thing is to just swap them and try.

I am also thinking that if it is really the HV transformer, a 3KV supply from a Xerox machine is a tiny little module, and 12KV isn't all that big either, so a modern potted replacement might make sense rather than rewinding. but I'll cross that bridge when I come to it.


John
 

The HV circuit takes less than 500mA when healthy: current is inversely proportional to the DC voltage in (unregulated, but normally around 16 to 17V raw DC).
I wouldn't recommend removing the diode D940 and then powering the rest, in case you end up with +ve grid volts.
Have you eliminated C953/C961 by just lifting C954 and powering?
I'll send you some useful component location info.
John


william_b_noble
 

replacing the ceramic caps seems to have fixe it - the orange ones that your very nice picture
C940, 945, 946, and 952, 966 - I used .01 3KV ceramics - two in parallel, except for C945/946 which are two .015 in parallel so I just paralleled 3 of the blue caps. when I can figure out how to upload an image in association with this post, I'll post a picture


william_b_noble
 

well, that was overly optimistic - I'm back where I started - I replaced the caps, I tested the diodes (again) and I had HV and a trace for a bit, but the intensity and focus didn't work, so I dug around a bit, found a wiring error, fixed it, and I've got nothing. when it was "working", the drive transistor (Q930) drew about .25 to .3 amps from the 12V supply. After fixing it (so to speak), I don't have a trace, the -1950 volt bias is gone, and the current wanders around from 1/2 amp to about 1.8 amps (at which point I turn it off and let the transistor cool down). The transformer doesn't get hot, the caps don't get hot, and occasionally the neon lamps in the regulating circuit flash - so I'm going to put it aside for a while. If anyone wants to pick up from where I left off, drop me a note - if it was working they seem to sell for $200+, as a parts/not working scope it's got to be worth something, and I'd much rather get it to someone who will finish fixing it - I think it is otherwise working - I have gotten a valid trace and the controls seem to work