Continue to Site

Welcome to our site!

Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

  • Welcome to our site! Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

Repairing a Metrix OX 8050 Digital Oscilloscope

Status
Not open for further replies.

T.Drakes Eng

New Member
The display is constantly fuzzy and displays this noise from somewhere! The odd thing is I can tune in to this frequency using the oscilloscope functions although the writing from the digital processor is also displayed fuzzy! I'm banking on it maybe been a noisy power supply seen as it's all driven from a switched mode power supply, but we'l see!
The internals are relatively empty for an oscilloscope this size!! Can't seem to find anything on Metrix at all in terms of a service manual or a schematic!
See What you think!
Tom
 

Attachments

  • 2014-10-02 22.09.14.jpg
    2014-10-02 22.09.14.jpg
    1 MB · Views: 1,385
  • 2014-10-02 22.18.57.jpg
    2014-10-02 22.18.57.jpg
    1.1 MB · Views: 1,474
Right been all round the power supply today and there is no noise on the outputs at all! The only output I haven't tested for noise is the 2kv output to the tube seen as it's so high! I'm not sure what the exact voltages outgoing this power supply should be but I have 4 common voltages that I recognise and a couple of ones I don't! I have the usual +5v -5v, +12v and -12v but also have +110v and +170v all going down the same ribbon. Any ideas what the 110v and 170v maybe for and if they are actually supposed to be those voltages?? Anyway seen as there's no schematic I will have to start tracing!
Just out of interest I fed a 1200Hz frequency into the scopes channels to see what it would display. I could tune into the waveform using the oscilloscopes controls but the display of the waveform was very fuzzy? Thinking now this could possibly be a fault on the scopes tube driving circuit! The tube circuit is also new to me as it has no flipper transformer so I'm not entirely sure what type of tube I have! Any advice would be really helpful right now?
Cheers Tom
 
This post is a little long in the tooth, (several months old, with no reply) but if you are still looking for info you might be interested in my similar issue here...

https://www.badcaps.net/forum/showthread.php?t=42768

These scopes all look to be very similar in basic construction as you can see below. As to the buzzing, its most likely a duff capacitor, which one exactly I don't know, but I might be able to offer some assistance by comparing with my now functional OX 803B

What I would be interested in are some better pictures of the inside of your OX 8050, as it might give me some clues about the unused connectors on the main board of my 803.

RIMG0233.JPG


One other suggestion, remove all sources of power, and clean *everything* with a clean dry paint brush to remove all dust, particularly anything associated with power supplies or high voltage (the connector on the tube and so forth), and inspect all of the joints on the back of the tube and the PSU board for dry joints with a magnifier. My issue was with the PSU board, but since the scope was dead, this was probably a bit easier to track down.

One other trick, turn the brightness right down, turn off the lights and in a darkened room, carefully look for arcing on the tube and the PSU.
 
Last edited:
Hi thanks for the reply! I will give those few things a try because yes it is still broken and yes it did just get tossed to the end of the bench for a rainy day!!! I think it is more likely a cap as I have been round the psu joints with solder and re seated all the connectors once already! I viewed all the low output's from the psu with my other scope and found a noisy output there due to a bad cap, so there's bound to be more on the way out! I will dig it out again and take some pics of the boards close up, plus I'm definitely trying that trick in the dark! Thanks Tom
 
One other thing to check, make sure the can with the boards on the back of the tube is on straight and that there is no damage to those boards. When mine arrived from the ebay seller, that can was on squint (as you may be able to see from the above picture) suggesting the thing had been bounced around.

First thing I did was very carefully straighten everything up and check that none of the tracks on that board have been lifted.

The noise you are hearing could also be due to issues with the switch mode PSU glitching, so if you follow through my post on the badcaps site, you will see the parts I replaced on my PSU there. If you have an ESR meter then give all of the PSU caps the once over (pay particular attention to the caps I swapped as they are the most likely ones to fail).

Check the values of the resistors associated with the TDA4605 and the MOSFET.

If the switch mode controller circuit is not working correctly then the MOSFET may be overstressed or firing when not required, and this might also cause noise. (The usual warning about high voltages applies, there are some pretty hazardous voltages in that part of the world, and these persist for a while even after switching off and unpluging the mains lead).

Take care when replacing components, I found my board was very prone to tracks lifting. That may be because my de-soldering technique is a little rusty, but I'm blaming it on the fragility of the board. If reworking any of the TDA4605 stuff, carefully discharge the associated caps with a high value resistor before replacing the TDA4605 or the Mosfet, particularly the big 400V main input electrolytic (C11 - 100uF 400V on my version). Pay attention to the accuracy (1%) and wattage and voltage rating (500V plus, depending on the location) if changing any of the PSU resistors, use good quality low ESR high temperature replacement caps if you want the thing to still be working in 10 years time.

As I said previously, carefully remove as much dust as you can with a clean dry paintbrush. Dust + HT = horrible crackling noises, arcing leading to untimely demise. Many years ago I used to work with faulty CRT monitors and terminals a lot. The grubbier they were, the more likely they were to fail. Dust + airborne grease and grime can have a surprisingly low resistance.

I couldn't find the schematic for the PSU online, but there is a similar design here..

http://www.tvservice.org/files/img/n583_big.gif

Obviously there are differences in component values, layout, and so forth, but this is the closest TDA4605 based design I could find. Similar but different. Close enough for you to get an idea of how it works though.

If you do get a chance, see if you can get some clear sharp pictures of all of the boards in your scope (no need to take them out, but the pictures need to be clear enough to identify traces and chip markings). I have a feeling that it might be possible to do some interesting hacks on these things. My first target is trying to add a serial port. Next is to try to figure out how the digital card in the 8040, 8050, 8100 series does its magic.
 
Last edited:
Interesting. Lots of stuff on the A to D card. I didn't realize it was that complex.
Some of the more interesting parts are listed below.

12bit ADC
FPGA
4Mb CMOS Flash
2x8bit ADCs - 80MS/s
**broken link removed**

If you compare the PSU pictures of the 803B, mine looks like a slightly later revision of the same PSU as this. There is a fairly complete set of pictures of all of the boards in my OX 803B here.

https://plus.google.com/u/0/photos/111082960064282217370/albums/6094881872199591361

My gut feeling is that your problem lies on the PSU board. There are a number of heat marked areas on that, and C5 looks a little the worse for wear. Also check the values of R3 and R4, if they are more than say 10% out, then I would suspect them too. Strangely all of the resistors on my version are 1% and possibly higher wattage.

I think I am going to have to grab some waveforms for you to compare with yours. That will have to wait though as it is getting late here.
 
Last edited:
PSU board
Compare your PSU with...
https://lh3.googleusercontent.com/-...AAAABec/BNGRNo8kByw/w754-h566-no/RIMG0234.JPG
... layout is slightly altered, but component names look to be the same.

While I remember... I would check and if necessary replace C15 - on mine it looked a little cooked, and was one of the things I replaced.
What is the story with the Diode next to W15 and C63, it looks from the photo as if it has a bodge wire soldered across it, and there appears to be another bodge next to R10 - are those your modifications?
 
Last edited:
Yer I understand where your coming from with the PSU however I'm a little sceptical as I checked all the voltages from the regulators and they are correct and show as being smooth on my scope. That's voltages -5, -12, +5, +12! I find it a little odd however that diode D10 has been shorted out with some wire?? The trace is working and if I input a waveform I can see it changing, it just seems to have a really bad shadow! Plus you can see the text being displayed on the scope from the digital CPU, but again it is all blurred! The display on the tube seems to be all over the place, it just won't focus! I'm thinking It's more to do with the tube driving circuit or the display driver! But if that takes power from another part of the PSU then it could still be the PSU! I think the over heating marks on the PSU board are just down to the last person to to do work on it as it has a label saying serviced on the 17/9/01 by Chauvin Arnoux! And no they are not my mods I have just replaced a few caps to get voltages smooth then it got put aside!!
 
The tube drive side of the PSU is a distinct possibility, especially as you describe things as blurred. We do however *have* HT otherwise we wouldn't see anything.

Are the images stable or shaking?

Do you have a high voltage (>2.5KV) probe for your other scope?

Did you clean all of the dust out of the HT side of the PSU, and the tube connector?

Is the image brighter/squashed at the top or the bottom, or brighter/squashed at the left or the right, or anything odd like that?

If so, that would suggest that the ramp voltages to the plates are getting distorted by the problem, and this would also tend to suggest the high voltage and/or HT side of the switch mode PSU is the probable cause.

Can you show me what the "Probe Adjust" test waveform look like?

When you change from Analog to Digital mode, and vice versa do things get any better?

I would check the caps on the HT driver (C31 and so forth) they are only pennies, so perhaps just replace them anyway and see if things settle down. Watch out when working on that part of the board, I imagine you can get a nasty belt from there even after things have been powered off for a while.

That shorted diode is a little bizarre.. since I don't know what that part of the circuit does I can only speculate on the effect, but it seems odd to fit a diode and then short it out. Certainly shorting it would screw up any rectification it might have provided but I would have thought that might result in a... lot of high frequency noise, or magic smoke as the output from that switch mode transformer coil would then be passed un-rectified to wherever.

EDIT: I just took a look at some of the pictures of my PSU, and there looks to be a link labeled E1 that allows shorting of D10, so there must be a reason for this strange arrangement. Different models presumably use different arrangements. Very odd, needs closer study.

I would need to know what is going on, on the back of the board (pictures of both sides of the board) to figure out what that the effect of that bodge is likely to be. I wouldn't remove that bodge wire just yet, until we know what its intended purpose is. It may have a deliberate purpose, rather than being some previous attempt to fix this issue.

One final question unrelated to the fault, does your scope have a serial port fitted?
 
Last edited:
I will have PSU board out again and have a look where d10 feeds! hv sounds like a possible cause then as the picture does dance around quickly insinuating hf noise! the trace is like it is compressed in the center yet as you move it side to side it stretches! Intresting thread on the crt tube manufactures there!!
 
I suspect the link across D10 may be a red herring. I would concentrate on the other voltages. I suspect that whatever voltage is used to power the ramp voltages to control the plates is sagging and having issues, and that probably means a duff capacitor on the PSU side. HT breakdown is another possibility. The dancing picture certainly suggests something is leaking/breaking down. Is the shimmering only in one axis? In other words does the picture shimmer left to right, or up and down, or both?
 
Well I think I have cleared up the mystery of the diodes (i.e. the link across the D10 diode).

There is the option to have one or two diodes in series on one of the power rails, selectable by linking out one of the diodes. This must be model dependent, so not the result of some misguided attempt to bodge up the PSU.

mZgPTgazUAaxC-3gYQW8Kht7PZdaa3NWF5YMCscz8ZE=w587-h241-p-no


If you line up the front and rear images here you will see what I mean. Note that E1 (the link presumably) has been removed on my variant.

Perhaps this is to select two different heater voltages for different tube types, who knows. I didn't get a chance to measure what was coming out on that rail, I simply lined up the two pictures to see what was going on.

So the moral of the story is... Don't cut the black wire!

EDIT: One final word on the subject tonight... since we lack a diagram for this particular scope, I thought you might be interested in this site. It offers a good description of a simple CRT Oscilloscope with a couple of examples to teach/remind us how the thing is supposed to work.

https://www.electronixandmore.com/projects/simplescope/

The voltages in the different Metrix scope versions will depend on the tubes being used, but the basic design is pretty much the same for most CRT scopes (and very similar in many ways to a CRT television or monitor, except for the use of deflection plates rather than scan coils).

Here is the spec sheet for a possibly very similar Philips tube, just for your interest. (It uses fairly high voltages at around 6KV. For comparison some colour TV tubes run at up to 30KV or so and x-ray tubes much higher, up to 150KV). https://frank.pocnet.net/sheets/186/d/D14383-123A.pdf

Since the tube is a Philips tube, this might also be of interest, and since it has a similar bandwidth it may well use the same or a similar tube.

https://www.qsl.net/vk5bar/AHARS-Resources/Philips%20PM3055/PM3055%20Service%20Manual.pdf

If we knew what tube was in your scope, we could then figure out what voltages to expect, since the spec sheet is likely to be available on the internet, perhaps here https://frank.pocnet.net/sheetsD.html

If you do go looking for a label on the tube do so carefully, it is obviously fragile, and there may be hazardous voltages lurking in there even after it has been switched off for a while.

Hence my question about having a suitable HT probe and another scope, we may need to probe the high voltages, and that might prove a little interesting without a suitable probe (we may be able to fabricate something if you don't have access to one, so don't chuck the thing in the scrap box just yet).

Also all of the other gubbins on the main board, the front panel and the A/D converter card may well be working time will tell.
 
Last edited:
Hi yes I'm nearly certain the rest of the scope is working fine and it's just a crappy display problem!! I think next is to test the HV rails and I shall have to rig up a suitable dropper resistor for the oscilloscope to measure it with and also work out the divisions on the scope! I really won't this scope working as the digital trace will prove so helpful when trying to decode dmx signal and other digital signals!
 
If I get a chance today, I'll see if I can get you some signals to compare from this working scope.

I'm angling for this to be the plate voltages rather than the HT, since the image is bright, but squashed at the top and to the right.

I think this part of the supply probably consists of D18 (or is it 19, the picture is unclear, but it is next to c36) the one with the heat marks on the PCB and its associated capacitor(s) so these would be the things I would check and or replace next, assuming of course that they are the rectifier and smoother for one or both plate voltages.

What do the plate voltage outputs look like on the PSU board? Typically it (or they) should be a solid flat DC voltage somewhere if the order of +200V to +280V DC, depending on the tube type, nice and smooth. If they are a lot lower and or have a lot of ripple or noise, we have our fault.

Looking at the pictures of my PSU PCB I think we need to scope out pin 9, 10 and 12 on the green ribbon cable and see what is there.

Having done a little more digging around, I am certain there are different CRT tubes in the different models (higher voltage tubes typically allow you to display faster signals) the OX 863 for example has a 15.5KV tube, whereas the OX 803B (my version) has a 2KV tube, yours perhaps may be 2KV or 6KV.

This makes good sense, but it also means there may possibly be slightly different X and Y amplifier driver voltages (plate voltages), although this is not necessarily the case. In other words the higher voltages on my PSU may not be exactly the same as those on your 8050. What will be the same though is the shape of the ramp/sweep and signal voltages driving the X and Y amplifiers, the X voltage should be clean saw tooth waves with a constant gradient and period dependent on the time base, the Y voltage should be an amplified, undistorted exact copy of your test signal.

What you will probably find on your scope is that the X sweep ramp and the Y signal look fine on the low voltage side, but by the time they hit the plates, on the X1/X2 and or Y1/Y2 pins on the tube (i.e. after the X and Y amplifier stages on the board on the back of the tube) they are flattened off towards the top. The cause of this is yet to be determined, but is probably breakdown of the smoothing cap across the associated PSU output (or possibly the diode or both).

If the DC X and Y plate driver voltages are clean, then one of the X or Y driver circuits on the tube neck/can boards would be the thing to check after that.

Does your model have the serial port?
 
Last edited:
Forensics on your 1st photo indicate the digital vector output is very limited. A working display would be very complex series of dots filtered to show only the lines connected between the dots.

In this case 5 dots are on the screen and 6 loops that would appear to be spurious loops between the dots.

I would expect hundreds or thousands of dots on the screen to form the characters and the signal but filtered to convert the dots into slew rate limited vectors between the dots.

Therefore I would start looking at the Dual DAC for problems either digital going in , or analog going out. It appears the Y axis DAC is not working at all.
 
Last edited:
Status
Not open for further replies.

Latest threads

New Articles From Microcontroller Tips

Back
Top