More 1S1 fun...a closer look at tunnel diodes and power supply ripple


Sean Turner
 

Back at studying my prototype 1S1. I've been systematically studying the circuit descriptions and schematics in attempt to figure out how it is supposed to work. Another potential trouble area I have identified is in the comparator circuit in the fast ramp generator. Specifically, there are a pair of 4.7 MA tunnel diodes that are part of this circuit. Specifically, D345 is supposed to control Q344 depending on what state it is in. However, the curve for D345 does not look much like a tunnel diode to me. I've uploaded a picture of both curves to (https://groups.io/g/TekScopes/album?id=261841) leveraging my 577D1 to overlay them. It seems to me that D345 might be stuck 'turned on'.

Another problem I have noted during live troubleshooting is that there is still severe 120 Hz ripple appearing all over the circuits of the plug-in even after replacing the extremely crusty 100 uF filter caps on the + and -19V rails. It's about 200 mVpp, and shows up on the vertical signal out, the dc offset out, and to a lesser extent on the horizontal out. It also shows up at many of the suggested test points in the troubleshooting steps. I'm about at wit's end trying to figure out where it is coming from. It is worth noting that I am running the 1S1 using a Type 81 plug in adapter in my 585A. I believe it should be compatible though...and I have other plug ins that work fine on the same adapter!

Any ideas are much appreciated.


Ed Breya
 

Sean, are you checking the TDs in-circuit? If so, the curves will be different from the TDs themselves. You could have a bad TD, or another failure that makes it look bad in-circuit. Also, remember the possibility of "back diodes" being used in some spots (https://groups.io/g/TekScopes/message/179519).

Ed


cmjones01
 

On Sat, 13 Mar 2021, 21:10 Sean Turner, <sdturne@q.com> wrote:

Another problem I have noted during live troubleshooting is that there is
still severe 120 Hz ripple appearing all over the circuits of the plug-in
even after replacing the extremely crusty 100 uF filter caps on the + and
-19V rails. It's about 200 mVpp, and shows up on the vertical signal out,
the dc offset out, and to a lesser extent on the horizontal out. It also
shows up at many of the suggested test points in the troubleshooting steps.
I'm about at wit's end trying to figure out where it is coming from. It is
worth noting that I am running the 1S1 using a Type 81 plug in adapter in
my 585A. I believe it should be compatible though...and I have other plug
ins that work fine on the same adapter!
That sounds to me like you're missing a ground somewhere. Can you confirm
continuity between the 0V points on the 1S1's PCBs, the plugin's chassis
and the mainframe's chassis and 0V points?

Chris


Jonathan Pyle
 

If you are getting ripple, maybe you need to replace the 1000uF capacitor, C632.


Sean Turner
 

Ed,

I called out the test conditions in the description of the album. :o) Indeed, it is in-circuit as I am not wanting to desolder potentially ok TDs unless I have to. How much of a change are we talking here? I've done in-circuit tests of known good TDs and you get the curve you are supposed to. Perhaps the measured values would be slightly different if removed and tested out of circuit, but it's at least recognizable as a tunnel diode curve with Ip matching 4.7 or 10MA depending on the particular TD.

I also doubt that any of these are back diodes...according to my handy dandy Tek measurement concepts book about semiconductor devices, back diodes seem to have an Ip that is _much_ smaller than a tunnel diode.

I'm happy be to be proven wrong, of course, but I haven't yet seen any reason why you can't use an in-circuit measurement to at least verify that "yes, this looks like a healthy TD".

Sean

On Sat, Mar 13, 2021 at 12:45 PM, Ed Breya wrote:


Sean, are you checking the TDs in-circuit? If so, the curves will be different
from the TDs themselves. You could have a bad TD, or another failure that
makes it look bad in-circuit. Also, remember the possibility of "back diodes"
being used in some spots ( https://groups.io/g/TekScopes/message/179519 ).

Ed


Sean Turner
 

Chris,

Good idea, but it looks like all the grounds are intact. It may well be the 1000 uF cap as another commenter pointed out.

Sean

On Sat, Mar 13, 2021 at 01:04 PM, cmjones01 wrote:


On Sat, 13 Mar 2021, 21:10 Sean Turner, <sdturne@q.com> wrote:


Another problem I have noted during live troubleshooting is that there is
still severe 120 Hz ripple appearing all over the circuits of the plug-in
even after replacing the extremely crusty 100 uF filter caps on the + and
-19V rails. It's about 200 mVpp, and shows up on the vertical signal out,
the dc offset out, and to a lesser extent on the horizontal out. It also
shows up at many of the suggested test points in the troubleshooting
steps.
I'm about at wit's end trying to figure out where it is coming from. It is

worth noting that I am running the 1S1 using a Type 81 plug in adapter in
my 585A. I believe it should be compatible though...and I have other plug
ins that work fine on the same adapter!
That sounds to me like you're missing a ground somewhere. Can you confirm
continuity between the 0V points on the 1S1's PCBs, the plugin's chassis
and the mainframe's chassis and 0V points?

Chris


 

On Sat, Mar 13, 2021 at 10:29 PM, Sean Turner wrote:


I'm happy be to be proven wrong, of course, but I haven't yet seen any reason
why you can't use an in-circuit measurement to at least verify that "yes, this
looks like a healthy TD".
Not trying to be a smart ass here but we're trying to verify the other way around here, as in "can something not looking like a healthy TD still be one?" One curve looks very much like a healthy TD, the other doesn't but could be a poor back diode. "Poor" if bridged by a relatively small parallel R.

Raymond


Sean Turner
 

No worries, no offense taken.

I think that could be the missing piece of the puzzle. There is, in fact, a small parallel R (R345, 20 ohms).

Sean

On Sat, Mar 13, 2021 at 01:47 PM, Raymond Domp Frank wrote:


On Sat, Mar 13, 2021 at 10:29 PM, Sean Turner wrote:



I'm happy be to be proven wrong, of course, but I haven't yet seen any
reason
why you can't use an in-circuit measurement to at least verify that "yes,
this
looks like a healthy TD".
Not trying to be a smart ass here but we're trying to verify the other way
around here, as in "can something not looking like a healthy TD still be one?"
One curve looks very much like a healthy TD, the other doesn't but could be a
poor back diode. "Poor" if bridged by a relatively small parallel R.

Raymond


Sean Turner
 

Raymond was correct. I lifted one end of R345, which makes D345 have a perfect curve for a 4.7 MA tunnel diode. Mystery solved on that front, and I learned something new.

I suppose my next move is going to be to get out the big can capacitor and replace it. Maybe I'll stuff it to preserve the look. Hopefully that will kill the ripple once and for all.

Sean

On Sat, Mar 13, 2021 at 02:02 PM, Sean Turner wrote:


No worries, no offense taken.

I think that could be the missing piece of the puzzle. There is, in fact, a
small parallel R (R345, 20 ohms).

Sean


Jonathan Pyle
 

Electrolytics are a lot smaller these days (I guess?), so I was able to just clip the leads to C632 and add a 1000uF cap on the underside of the plug-in, hanging there not touching anything.


Sean Turner
 

Do you have a part number for the cap you used?

Sean

On Sat, Mar 13, 2021 at 02:35 PM, Jonathan Pyle wrote:


Electrolytics are a lot smaller these days (I guess?), so I was able to just
clip the leads to C632 and add a 1000uF cap on the underside of the plug-in,
hanging there not touching anything.


Jonathan Pyle