Tunnel Diode Substitution, according to Stan Griffiths


 

I'm researching how to replace the tunnel diodes in my 475 with Russian parts, and I found that the page on TekWiki references a page by Stan Griffiths on how to modify a circuit to lower current tunnel diodes. Unfortunately the page (http://www.reprise.com/host/tektronix/reference/tunnel_diode.asp) no longer exists on the internet, so the link from TekWiki is dead. Happily, I found a copy archived on the Wayback Machine. For everyone else's edification I will post the text here while I figure out how to add it back to TekWiki in less ephemeral manner:

Sweep Trigger TD Substitution (Bill & Stan's Tektronix Resource Site)

Tunnel Diodes are getting hard to find, although some types are more rare than others. Many Tektronix oscilloscopes use tunnel diodes in their sweep trigger circuits. When these diodes fail, the entire instrument is useless until you find a replacement.

If you cannot find the Tunnel Diode you need, it it sometimes possible to use an alternate type. This page describes modifications that were performed on a Tektronix 547 sweep trigger circuit to use an available 4.7mA Tunnel Diode in place of the impossible-to-find 10mA original.

Our thanks to Morris Odell, VK3DOC, who devised these modifications and has allowed us to share them with you here.

The original 547 A Sweep Trigger circuit diagram is available here.

Step by Step Modification Instructions:

1. Replace the bad 10mA Tunnel Diode (D245) with a new 4.7mA Tunnel Diode (Tektronix part number 152-0125-00) (available from Stan Griffiths).

2. Replace transistors Q234 and Q244 with silicon transistors having Ft around 200MHz and Beta around 100, such as 2N4248.

3. Replace resistor R236 (30.1K) with a 47K 1Watt resistor.

4. Add new 30K 1 watt resistors in series with R247 and R248, which are each 15K 2 watts. (Total resistance is now 45K each).

5. Because the Tunnel Diode must be biased below its trigger point, there is less adjustment margin with a 4.7mA Diode than with the original 10mA Diode. This means that the values of R247 and R248 can be quite picky. Because of the variations between individual tunnel diodes, it may be necessary to use slightly smaller or larger resistances.

Morris points out that the B Sweep Trigger circuit is identical to the A Sweep Trigger. This modification should therefore be equally applicable for that circuit.

Tek and Tektronix are trademarks of Tektronix, Inc.
Not affiliated with, or supported by, Tektronix, Inc. Last modified on 9/11/04 7:18:10 PM


 

It just so happens that there are a bunch of 4.7 mA tunnel diodes available on eBay at the moment (1N3717s) at the lower end of usurious prices. I'm thinking of buying a couple just to have them in case of future need, but I've already done that with a dozen Russian tunnel diodes that are better matches for the 1N3718s in my 475.

The problem with the Russian tunnel diodes is that they use a very different package than the gold leaded top hats currently in the 475, and I'm trying to figure out how to modify them (or the 475) so that they can be mounted in the circuit. Having gold leaded top hats would be a lot easier to deal with, as they would just plug into the existing sockets.

-- Jeff Dutky


adesilva_1999@...
 


 

True, that's not a bad price at all, but the next offering up is over $50. I can get four tunnel diodes for that price, conveniently packaged in a free oscilloscope!

The thing is, I've already got 1N3718s in my parts mule and my back up 475, and I am interested in what would need to be done if a tunnel diode failed and I was not able to find a replacement.

My first preference, after pulling one from a parts mule, is to find a part with similar characteristics, which I have already done with several Russian parts. I will, however, need to figure out how to fit the Russian parts to the existing circuit, which is not entirely straightforward. It was while searching for any other people's experience doing this that I stumbled on Stan's article, describing what I would now call my second preference.

The second preference being modifying the circuit, not merely to mount a tunnel diode in a different style of package, but to use a tunnel diode with different specs. Since the original page no longer exists, and since we were just discussing this in another thread (without anybody mentioning this article, or the possibility that it could be done) I figured I should drop this here.

I've updated the TekWiki to include a page containing the article from Stan Griffiths as a new entry.

-- Jeff Dutky


adesilva_1999@...
 

Got it. I was thinking that you were desperately in need of one.


Morris Odell
 

I'm pleased to see my tunnel diode hack has lived on! I still use the 547 as my primary bench scope and the trigger works nicely with the 4.7 mA TD as described above.

Morris


um-gs@...
 

All,
being a German Tek fan, I just wanted to point out that there is a shop offering Russian Tunnel Diodes at VERY reasonable prices.
I used them already without problems in a 581A as well as a 547 as replacements to restore full triggering capability.

https://www.mos-electronic-shop.de/tunneldioden-c-87_86.html

As well this probably well known link might be helpful:
https://w140.com/tekwiki/images/8/8f/Rca1963TunnelDiodeManual-page147.png

Hope that helps a bit,
Gordian


 

So I've done just enough research to be able to speak intelligibly (if not intelligently) on this subject: my issue with the Russian tunnel diodes is that they are provided in obsolete diode outline packages: DO-18, DO-20, or what Tek called a "pill" package.

The DO-18 and DO-20 packages have flat leads that I can figure a way to work with, but the pill package, which is a tiny "top hat" like package about 3 mm in diameter and 2 mm tall, is more challenging. I think that the proper way to work with the pill package is to have a spring clip that pinches the pill between two contacts. I have tried, furtively, to find "diode clips" for sale on line with no success, and that's as far as I've gotten.

I have done almost no research into crafting my own spring clips, though I think I will need some kind "spring bronze" strip or sheet to make the contact clips. Absurdly, the easiest source for this looks like weatherstripping for exterior doors.

The pill package Russian diodes that I bought for the 475 are 1I308V-80 (which looks like 1N308B in Cyrillic, but the N is backwards), which appear to have the best specs to replace the 1N3718/STD927 tunnel diodes in a 475 (9-11 mA current, 4-10 pF capacitance).

-- Jeff Dutky


Greg Muir
 

I am not aware if anyone has referenced this site but this fellow has done considerable legwork with regards to tunnel diodes.

https://welcolab.wixsite.com/discovery/1-tunnel-diodes-xref

Greg


Ed Breya
 

Jeff wrote: "I have done almost no research into crafting my own spring clips, though I think I will need some kind "spring bronze" strip or sheet to make the contact clips."

You likely already have plenty of springy contact clip materials available in the form of junk parts, like board edge connectors, for instance. Just figure out some good candidates and a configuration to attach them to the board and properly hold the pill package. Salvaged contacts from board edge connectors are nice and springy and gold plated, and ready to go once figured out mechanically.

Pill packages aren't obsolete - there are all sorts, built for stripline, coaxial, and other structures, where lead inductance is minimized.

Ed


 

Greg,

I am not able to connect to the discovery site at this time. I found a snapshot (incomplete?) on the Wayback Machine,

His experiments, assuming I'm seeing the full page in the Wayback snapshot, seem more focused on measurement than on installation in a specific instrument. He appears to be soldering directly to the tunnel diode leads, and he is working with DO-80 or DO-20 style parts, so it's not really applicable to my questions about interfacing "pill" packages to through-hole assemblies.

He does have some more complete discussion of the difficulties in measuring tunnel diode IV curves than I have seen anywhere else. He also brushes up to the topic of forming inductors around the body of a small resistor, which has been discussed in recent threads on this group, but he seemed oddly reluctant to do the obvious build.

There are some very nice images of Tek instruments in action.

-- Jeff Dutky


 

Thanks Ed! I do, in fact, have a small collection of partially mangled connectors with nice, springy contacts in my junk bin. Sadly, none of them appear to be gold plated, but I will have courage.

-- Jeff Dutky


Miguel Work
 

I have uploaded in files section a pdf with my tested Russian diodes:

https://groups.io/g/TekScopes/files/Russian%20Tunnel%20diodes%20for%207S14

Regards!

-----Mensaje original-----
De: TekScopes@groups.io [mailto:TekScopes@groups.io] En nombre de Jeff Dutky
Enviado el: lunes, 12 de abril de 2021 19:26
Para: TekScopes@groups.io
Asunto: Re: [TekScopes] Tunnel Diode Substitution, according to Stan Griffiths

Thanks Ed! I do, in fact, have a small collection of partially mangled connectors with nice, springy contacts in my junk bin. Sadly, none of them appear to be gold plated, but I will have courage.

-- Jeff Dutky