Maker Pro
Maker Pro

Problem with Heathkit IO-102 oscilloscope - vertical deflection doesn't change

seanspotatobusiness

Sep 11, 2012
193
Joined
Sep 11, 2012
Messages
193
I have a Heathkit IO-102 oscilloscope - a modern marvel of electronic engineering! However my particular scope has a problem. It may have been caused by me applying a voltage that was too high for the settings (don't know whether they're susceptible to that) whilst I was figuring out how to use it (I had trouble finding the trace initially and didn't know about the 1V-PP output) or it may be something else entirely.

OlVvI.jpg


The problem is that regardless of the voltage applied, the deflection is 1.5 of the (smallest) subdivisions, in the wrong direction (red to positive gives a downward deflection). I get the same deflection with 3.8 volts or 9 volts with the vertical set to 1x DC and maximum gain.

I opened it up (had been unplugged for over a month) and I see no burned or burst components which is about the extent of my electronics troubleshooting skills. I paid particular attention to the vertical gain board.

attachment.php


attachment.php


I took the extended cover off the back of the CRT and inspected the wires there. Perhaps it was always like this and is of no importance, but I noticed that one of the vertical amplifier wires connected to the tube was a bit bare (purple demarcation). I used my multimeter to determine that the resistance between the two wires was 11K ohm (right green demarcation). The resistance between the two wires from the horizontal amplifier is >1M ohm (left green demarcation). Maybe this is of no relevance whatsoever?

I have the schematic here: http://i.imgur.com/RCzb8.gif - has anyone any idea of what I could try checking?
 

Attachments

  • RZbLo.jpg
    RZbLo.jpg
    101.9 KB · Views: 1,514
  • hQSFE.jpg
    hQSFE.jpg
    61.5 KB · Views: 1,289
Last edited by a moderator:

KrisBlueNZ

Sadly passed away in 2015
Nov 28, 2011
8,393
Joined
Nov 28, 2011
Messages
8,393
I guess the prime suspects are the JFETs, Q1 and maybe Q3. JFETs can be easily damaged by high voltages.

What is the part number? It looks like EL131. Google EL131 JFET, there are lots of hits including someone who was looking for a replacement for his Heathkit scope.

I've never heard of EL131 and I don't know what manufacturer it might be. The chances of finding a data sheet would be close to zero. Q1 and Q3 are both being used as source followers, so the gate-source voltage is the only important factor. Digikey have several N-channel JFETs that might be suitable - J105, J106, J107, J111, J112, J113. Mouser have an enormous range. Even after I narrowed the selection criteria to through-hole and 10 mA source current or more, there were 44 matches!

Anyway, get in there with a multimeter and measure all three terminal voltages on Q1 with no input, and with an input connected to cause deflection. That will tell us whether Q1 needs to be replaced or not.
 

seanspotatobusiness

Sep 11, 2012
193
Joined
Sep 11, 2012
Messages
193
Hi Kris, I was at home for a few weeks and unable to test my oscilloscope. Now I'm back and have it behind me. The leads on Q1 and Q3 appear to be darker in colour than the similar components around them - they're a sort of bronze compared to the silver of the others. I would have measured the voltages already but I don't know what to use as a reference/ground? To where should I connect my black lead for these measurements?

Thanks a lot for your help!
 

KrisBlueNZ

Sadly passed away in 2015
Nov 28, 2011
8,393
Joined
Nov 28, 2011
Messages
8,393
Hi Kris, I was at home for a few weeks and unable to test my oscilloscope. Now I'm back and have it behind me. The leads on Q1 and Q3 appear to be darker in colour than the similar components around them - they're a sort of bronze compared to the silver of the others.
I don't think that's a sign of damage. The little bit of silicon inside the package is far more vulnerable to damage than the wires are. Probably they're just made of a different metal, or are untinned vs. tinned on the other semiconductors, or something like that.
I would have measured the voltages already but I don't know what to use as a reference/ground? To where should I connect my black lead for these measurements?
To the chassis, or the ground of the input connector.
Oscilloscopes, like most electronic circuits, have a global "rail", which is called variously "ground", "common", "0V" ("zero volts"), "earth" and sometimes a few other names. This is the point in the circuit that other voltages are all specified in relation to ("with respect to"), unless circuit voltages are specified as being between two specific other points in the circuit, and it's the rail that you connect the negative probe of your multimeter to.
 

seanspotatobusiness

Sep 11, 2012
193
Joined
Sep 11, 2012
Messages
193
I used the collector leg of D1 as ground, based on the schematic.

With the vertical dial set to GND, Q1 voltages as follows:
G: 0
D: 9.91
S: 0.02

With the vertical dial set to x1 and a 3.58 V potential applied to IN and GND vertical inputs, Q1 voltages as follows:
G: 2.25
D: 9.71
S: 1.81

Should I try it again with a different voltage applied to the inputs?
 

KrisBlueNZ

Sadly passed away in 2015
Nov 28, 2011
8,393
Joined
Nov 28, 2011
Messages
8,393
Those were helpful tests. It looks like Q1 may be OK. Those voltages are kind of compatible with a working device. Can you do some more testing.

Q1 operates as a voltage follower, so its source voltage should follow its gate voltage, which in turn comes from the input. Can you check with both positive and negative input voltages that the source voltage follows the input voltage. Normally a FET has a voltage difference between the gate and source; in this case, according to the voltages marked on the schematic, it has zero volts drop. I would like to see the gate and source voltages for several DC input voltages - two positive and two negative. You've already given me the voltages with 0V at the input.

When you measure these, make sure the input range selector is set so that you would expect a noticeable vertical deflection. For example if your highest input voltage is 9V, select 5V per division, so you'd expect nearly two divisions of vertical deflection (assuming a division is around 1 cm).

The drain voltage comes from a rail that is supposed to be regulated by ZD6 and it shouldn't change much; also it should be lower than 9.9V. Perhaps ZD6 is faulty. This won't be the cause of the "no vertical deflection at all" problem but it will affect accuracy once you get the deflection working.

It could also be useful to know the voltages on Q3 under those same five conditions, with the vertical gain adjustment turned to maximum.

While you're at it, could you measure the voltages on all terminals of all three-terminal components with no input applied. The input stage may be OK, so there may be a failure elsewhere in the circuit. A DC voltage check on all active devices will tell me a lot.
 
Last edited:
Top