Repairing a Metrix OX 8050 Digital Oscilloscope

[T.Drakes Eng]

It works in a fasion!! It has a little curve on the peaks of the digital probe test signal that my other scope doesn't pick up so I'm guessing on another cap gone down in the input section? It does it on both channels though which points to something more generic! See what y think!

 

[T.Drakes Eng]

I should also add that this doesn't change when you change the channel to DC seen as my pic had it still set to AC!!

 

[rumpfy]

Post 38 describes a 'mono accelerator' type of CRT.
These types have a relatively insensitive CRT system and the vertical amplifier drive required usually limits the bandwidth to 20 MHz.
I note the bandwidth for the metrix 8030 is up t0 60 MHz, and for this bandwidth, a PDA type crt is required, in my experience.
So, if the CRO has a 60 MHZ bandwidth AND a 'mono accelerator' type of CRT, does this mean that the CRO operates as a sampling CRO on analog, or is the CRT a PDA type.
Sorry for the intrusion; but my mind works funny sometimes!!.

 

[itsthatidiotagain]

It works! Excellent news. I suspect the rounded peaks are simply because the scope probe needs to be calibrated to the scope. i.e. push the end of the scope probe in to the hole marked "Probe Adjust" on the front of the scope, and adjust the trimmer on the probe (either on the BNC body or on the tip body, depending on the probe type) for best results. Probes from one scope don't always work well on another without adjustment.

For comparison, here is a picture of my OX 803B screen just after I fixed it, and pre-calibration.

**broken link removed**

If that is the issue, then just get some new probes, calibrate them and leave them attached to the scope (check them every so often depending on how much use the scope gets, to ensure they haven't drifted off).

Oh.. and give the case and the buttons a treat...

(Other cleaning products are available. )

 

[itsthatidiotagain]

Post 38 describes a 'mono accelerator' type of CRT.
These types have a relatively insensitive CRT system and the vertical amplifier drive required usually limits the bandwidth to 20 MHz.
I note the bandwidth for the metrix 8030 is up t0 60 MHz, and for this bandwidth, a PDA type crt is required, in my experience.
So, if the CRO has a 60 MHZ bandwidth AND a 'mono accelerator' type of CRT, does this mean that the CRO operates as a sampling CRO on analog, or is the CRT a PDA type.
Sorry for the intrusion; but my mind works funny sometimes!!.

I am making a bold assumption that the tube in the OX 8050 is the same as the tube in the OX 803B (the tube I previously identified). This may in fact not be the case. The OX 863 for example has a PDA type CRT and a stated bandwidth of 150 MHz, so based on that observation, the tube in the 8050 may be better than the 803.

60 Mhz seems to be about as far as you could realistically push a mono accelerator type tube (although one or two of the tubes I looked at recently appeared to claim higher bandwidth, up to 80MHz or more). The OX 803 only claims a bandwidth of 40 MHz.

Without the part number from the tube though I am only speculating, and since we have just managed to get it working, I would not recommend stripping out the tube just to see what it is.

 

[itsthatidiotagain]

I see you very wisely swapped a bunch of capacitors, just for the record, which ones, and what are the values.

If you get a chance and think it worth the risk, I would be keen to see what waveforms you get on J18, J19 and J301. If you don't want to risk prodding about in there, I fully understand. It is a little tricky to get at some of those pins.

Caps replaced and no dig at your soldering intended!!

I'm out of practice...

 

[T.Drakes Eng]

I just want to say I'm really grateful of all the help you guys have given me and those voltages and what they do on the plates in the tube have proved invaluable! I just want to take the opportunity to tell you a little about myself! As you might have seen I'am only 20 years old and just won't to say I am completely self taught in electronics as I'am a plant electrician by trade! My father is a Mech engineer and he to is self taught in electronics! Together we run a little service repairing all sorts of electronics from amplifiers to power supplies to digital lighting control circuits! We do it out of interest only! I'd love to take it up full time but I don't know enough yet and it's hard to get a break into but I think we've made a start! This is the first time I've ever opened a scope up and I say it has been a learning curve!

Again thanks for your help and yes I will be happy to return the favour by having a play with the new scope on those plug connectors!
Many Thanks Tom
PS look out for a new post soon about repairing a Roland digital sound desk with a randomly popping channel!

 

[T.Drakes Eng]

By the way just had a go with the probe adjust but it isn't the problem entirely it was a little out since I last did it on my other scope however the lines still aren't perfectly horizontal they are continuously concaved front down, back up! They are perfectly straight now but still on the pss! Tom

 

[itsthatidiotagain]

Hi Tom,

I would still go for the probes as the most likely source of the non linear square waves, since they are from a different scope and they may not match particularly well. If I am correct in my identification, your other scope is a Gould OS 250, which is only a 10 MHz scope, so the probes for that may not be very well matched to the OX 8050.

If you don't mind the wait, those **broken link removed** seem reasonably good although you might have a bit of a delay while they ship from China.

Certainly for the price they make a good spare, for when the inevitable happens and you break a probe. They claim to be good for 100 MHz so they should work fine on your 8050. But reading the small print, it is likely that 100 MHz is only available on x 10 mode (better response on x 10 is usually the case, unless you spend $100+ on a probe), the x 1 response is probably lower, perhaps as low as 6MHz. Spec. of the probes I linked is below.

Module: P6100
Attenuation ratio: 1:10
Input resistance:1M¸/10M¸
Input Resistance: 1M:10M
Bandwith: DC-6MHz, DC-100MHz
Input Capacitance: 1X:85pF-120pF, 10X:18.5pF-22.5pF
Compensation Range: 10pF-30pF
Working Voltage:1X:<300VDC+AC Vp-p, 10X:<600VDC+AC Vp-p

Definitely good enough for most things, and certainly good enough for audio work and low end microcontrollers. The response doesn't drop off a cliff at 6MHz on x 1, but becomes progressively poorer and using the x 10 attenuation mode with them on your OX 8050 should still allows you to fill the screen with a +5v signal at 60MHz. In the unlikely event that one of those probes also shows a problem, then we need to look at the set up of the front end.

That would be a learning experience for both of us, since we don't have a service manual. Don't go fiddling with the trimmers on the front end until we are certain we have a problem though. It can be a bit of a nightmare getting things working accurately again after someone has been for a random twiddle.

There is a good video of Dave taking us through adjusting a Tektronix 222v 50 MHz scope. If you haven't seen the EEV blog, bookmark it, subscribe and join the forum. Lots of good stuff there.

If we do end up having to go down that route, I suggest we would need to take a few pictures of the board to reverse engineer the layout and function of the necessary trim pots and caps.

Scope probes are a little misunderstood, their response is always a bit of a compromise due to the way they work. If you are up for a bit of a 'scope probe electronics lesson you will find this fascinating. http://www.dfad.com.au/links/THE SECRET WORLD OF PROBES OCt09.pdf

Of particular interest is Figure 3 - the diagram shows the multiple square waves obtained by trimming a scope probe, overlaid one on top of the other. You can see the effect of over and under compensation (the rounded shoulder you are seeing versus the spike and dip of over compensation). Your dual Analog/Digital scope will give you an insight in to the value and limitations of both types of scope. If you ever see another one of those OX 8050's going cheap let me know, it looks like a very useful tool.

Self taught is probably how most of us learned the majority of what we know. I've been learning electronics all my life, and am still discovering things. The first rule of electronic and electrical engineering is safety first, so if you do go prodding around on those headers I mentioned (J18,J19 etc) be aware that there are some high voltages near there (particularly next to J18/19) so proceed with caution. Also worth noting is that I am based in Scotland, so expect some delays in my replies.

EDIT: Just checked the probes on my Tektronix TDS 1012B (100 MHz - 1GS/s) in the office, they are also only good for full bandwidth (200 MHz) on x 10 and 6 MHz on x 1

 

[itsthatidiotagain]

Any joy with squaring up your square waves?

 

[T.Drakes Eng]

Hi sorry I've been a little quiet lately but I'm working away from home at minute! As soon as I get back I will follow up our previous works! What I can say is that before I went away I tried adjusting the probe to the test signal and it got rid of the kick in the signal but the digital wave is still concaved! More pics to come soon when I get home!! Cheers Tom

 

[itsthatidiotagain]

Hi sorry I've been a little quiet lately but I'm working away from home at minute! As soon as I get back I will follow up our previous works! What I can say is that before I went away I tried adjusting the probe to the test signal and it got rid of the kick in the signal but the digital wave is still concaved! More pics to come soon when I get home!! Cheers Tom

Do you have access to any other (higher bandwidth) probes, I'm still convinced this is caused by the probe, and not the scope.

 

[itsthatidiotagain]

Hi sorry I've been a little quiet lately but I'm working away from home at minute! As soon as I get back I will follow up our previous works!

I hope you are not working too hard...

 

[KeepItSimpleStupid]

I agree, the first order problem is to match the probes t the scope. 1 M || 20 pf for an input Z is pretty typical. Probe bandwidth has to be = or greater than the scope. There are some probes that are non-standard. Don't get 50 ohm probes, for instance.

 

[itsthatidiotagain]

Any progress?

 

[itsthatidiotagain]

I'm back for more punishment... Recently I picked up a Metrix OX 2000 oscilloscope (because it sounded like a good idea)... which is missing its analog board, can anybody suggest where I might get one? I suspect "from the manufacturer" is unlikely to be the correct answer since they have thus far completely ignored all of my emails.

If you want pictures of the insides of the scope, these will be available soon.