Re: 5642 rectifier tubes in a Tek 535: Replace or swap for silicone?
The idea of replacing tube rectifiers in TVs and other CRT devices is an old one. One of the most important considerations is the capacitance of the reverse biased diode. The higher the current capacity of a silicon rectifier, the higher the capacitance. Many TVs would have serious horizontal linearity issues as well as reduced high voltage when a solid state rectifier was substituted.? A string of 1N4007 diodes would likely have too much capacitance, especially at the 40 KC or higher frequency a scope high voltage supply usually operates at. A high voltage rectifier has several carefully designed and/or chosen junctions in series to share the reverse voltage, 1N4007 diodes may not balance their reverse voltage sharing properly. They are almost always used with balancing resistors in high voltage power supplies.
??? Bruce Gentry, KA2IVY
toggle quoted message
Show quoted text
On 7/28/21 11:45, saipan59 (Pete) wrote: Just wondering: Is it feasible to replace a 5642 with a string of (for example) cheap 1N4007's?
I don't have a need to do this at the moment, but I already have a bunch of them.
Pete
|
Re: Tek Scopes found in old picture book
On 2021-07-28, at 17:02, Dave Peterson via groups.io <davidpinsf@...> wrote:
And are those 400 series scopes I see hints of (blue screens) on the far right of the racks?
My brain tells me that I saw Hamegs in the rack when I visited seven years ago. My picture at is not clear enough to read what¡¯s on the scope in the rack; it doesn¡¯t shout ¡°Hameg¡± at me either. It doesn¡¯t look to me like a 475 or some such. Gr¨¹?e, Carsten
|
Re: 5642 rectifier tubes in a Tek 535: Replace or swap for silicone?
Yep. 1N400x devices are very slow to turn off; they are charge sponges that rectify as a side effect. UF4007s would be a much better choice.
As to equalizing networks (R and/or C), if the individual rectifiers all came from the same batch, I haven't found them to be necessary. Adjacent devices from an SMD reel or strip will be from the same batch, to a high probability.
The ST app note I linked to earlier has much more detailed information.
Tom
--
Prof. Thomas H. Lee
Allen Ctr., Rm. 205
350 Jane Stanford Way
Stanford University
Stanford, CA 94305-4070
toggle quoted message
Show quoted text
On 7/28/2021 09:04, SCMenasian wrote: It appears that the high voltage power supply runs at ~60kHz. I wouldn't trust a garden variety silicon diode like 1N400x to hack it. Fast recovery diodes might be OK but you might need to take precautions to ensure that the high frequency voltage is distributed evenly along the string.
|
|
|
Re: 5642 rectifier tubes in a Tek 535: Replace or swap for silicone?
It appears that the high voltage power supply runs at ~60kHz. I wouldn't trust a garden variety silicon diode like 1N400x to hack it. Fast recovery diodes might be OK but you might need to take precautions to ensure that the high frequency voltage is distributed evenly along the string.
|
Re: 5642 rectifier tubes in a Tek 535: Replace or swap for silicone?
Absolutely! Silicon HV rectifiers beyond a kV or so are actually made of series strings of lower-voltage units. A clue is their much higher forward voltages (which makes them difficult to test with conventional DMMs).
One consideration is ensuring even partitioning of the reverse voltage across the individual diodes. If the diodes all came from the same wafer, you generally won't have to worry about adding shunt resistors, especially if you're not trying to push the limits of breakdown voltage. There's a lot more detail in this app note from ST:
-- Cheers
Tom
--
Prof. Thomas H. Lee
Allen Ctr., Rm. 205
350 Jane Stanford Way
Stanford University
Stanford, CA 94305-4070
toggle quoted message
Show quoted text
On 7/28/2021 08:45, saipan59 (Pete) wrote: Just wondering: Is it feasible to replace a 5642 with a string of (for example) cheap 1N4007's?
I don't have a need to do this at the moment, but I already have a bunch of them.
Pete
|
Re: 5642 rectifier tubes in a Tek 535: Replace or swap for silicone?
Regular 1n4007s are too slow. UF4007s are fast enough, but it's easiest to just get proper HV rectifiers (which are actually made from strings of series diodes inside).
Ed
|
Re: Tek Scopes found in old picture book
On 28. Jul 2021, at 09:07, Carsten Bormann <cabocabo@...> wrote:
...Yes. See the picture at I like the large black resistors to the right of their heads. Look like oversized fuses in their holders... In the original pic I didn't notice all the tubes on top of the shelves. What are they for? Are these thyratrons? My first guess leant towards analogue computers, thinking there were operational amplifiers all over. Stupid me, I even read about that Manchester Baby some months ago. cheers Martin
|
Re: 5642 rectifier tubes in a Tek 535: Replace or swap for silicone?
Just wondering: Is it feasible to replace a 5642 with a string of (for example) cheap 1N4007's?
I don't have a need to do this at the moment, but I already have a bunch of them.
Pete
|
Re: Tek Scopes found in old picture book
That makes perfect sense, thanks Carsten!
Looking forward to the goods :-)
toggle quoted message
Show quoted text
On Wed, Jul 28, 2021 at 9:07 AM Carsten Bormann <cabocabo@...> wrote:
Williams tube.
Gr¨¹?e, Carsten You might be talking about the loose tube in a box at the bottom of
the picture. I mean a rack mounted unit between the two PCs.
Yes. See the picture at
(because of the inexplicable restriction on sending attachments on this list, I just put it up in a random location. Copy that into the groups.io archive if you can stomach navigating that entirely unneeded bureaucracy) ¡ª zoom in to see the faint green image of the bits.
There are
two scopes and a third thing, in the rack, that also looks like a
scope. A Williams tube device has no front display because the front
of the tube is covered by the read-out plate. The problem is ¡ª how do you debug a Williams tube?
I seem to remember that the makers of the Manchester Baby (SSEM [1]) just put another tube on the same signals so they could see what was written to the storage tubes (ah, [2] has the details). During restoration, they couldn¡¯t get the original size display tube, so they have a half-size in the rack now.
BTW I tried getting in touch off list but haven't heard back :) Would
you mind checking :) thanks Everything is on its way, thank you.
Gr¨¹?e, Carsten
[1]:
[2]:
|
Re: Tek Scopes found in old picture book
And are those 400 series scopes I see hints of (blue screens) on the far right of the racks?
I'm back from our travels, but the book is now packed away in the shipment bound for home. I'll look up the picture credits when it arrives.
I could be wrong, but in my scan of the text when I came across the picture didn't say anything about the Manchester Baby, restoration, or anything of note w.r.t. the picture. I'm curious to read it now to see what it does say. I'm glad I brought it up with you all. What a fascinating back story. I'd love to give credit to the photographer, and wonder what his involvement was. Perhaps there's more to be had there? Unfortunately it has been a few years since.
My wife and I both worked at Intel back in the 90s. Actually never knew each other while there, but that's another story. Talk about getting off topic! This book must have been a 30th anniversary thing, and I'd left by then. I suspect she gifted this to her parents - we'd just met, so I don't ever recall seeing it before. Well, maybe. I recall Intel being a pretty fun place to work, and recall being rewarded nicely at times, and chastised by Andy at other times. I appreciated the gifts. Still have one sitting right behind the monitor I'm looking at. As well as my office art which is a layout plot of my first design there. Intel was my first job out of school, so I guess left a lasting impression. Probably why we decided to keep the book. Now my kids can deal with it when it's time to clean out my house! Along with a few scopes.
Thanks all for the background.
Dave
toggle quoted message
Show quoted text
On Wednesday, July 28, 2021, 2:02:00 AM PDT, mikecaa <4sumik@...> wrote:
Think it's a replica Manchester baby at the Manchester computer museum. See below;
Regards - Mike
|
Re: 7000 test/cal module(s) and backplane breakout board.
Schematic fixed, thanks for pointing out the error.
toggle quoted message
Show quoted text
On Tue, Jul 27, 2021 at 11:46 PM Andy Warner <andyw@...> wrote: You are right about Q1-4, I was in a rush, will fix it tomorrow.
I can test things with a simple current mirror, but I usually resort to a
Wilson if I am not using matched parts. Just habit, perhaps a bad one.
On Tue, Jul 27, 2021 at 23:33 Ed Breya via groups.io <edbreya=
[email protected]> wrote:
Looks about right, except the PNPs appear upside down. If you're going
with emitter ballast resistors in the mirror, then you probably don't need
transistor matching or the Wilson configuration, and two NPNs per mirror
would suffice. The compliance is pretty much constant since the row and
column nodes hold very close to ground - there's hardly any Early effect.
Ed
-- Andy
--
Andy
|
Re: Tek Scopes found in old picture book
Think it's a replica Manchester baby at the Manchester computer museum. See below;
Regards - Mike
|
Re: Tek Scopes found in old picture book
Williams tube.
Gr¨¹?e, Carsten You might be talking about the loose tube in a box at the bottom of
the picture. I mean a rack mounted unit between the two PCs.
Yes. See the picture at (because of the inexplicable restriction on sending attachments on this list, I just put it up in a random location. Copy that into the groups.io archive if you can stomach navigating that entirely unneeded bureaucracy) ¡ª zoom in to see the faint green image of the bits. There are
two scopes and a third thing, in the rack, that also looks like a
scope. A Williams tube device has no front display because the front
of the tube is covered by the read-out plate. The problem is ¡ª how do you debug a Williams tube? I seem to remember that the makers of the Manchester Baby (SSEM [1]) just put another tube on the same signals so they could see what was written to the storage tubes (ah, [2] has the details). During restoration, they couldn¡¯t get the original size display tube, so they have a half-size in the rack now. BTW I tried getting in touch off list but haven't heard back :) Would
you mind checking :) thanks Everything is on its way, thank you. Gr¨¹?e, Carsten [1]: [2]:
|
Re: 7000 test/cal module(s) and backplane breakout board.
Andy wrote: "The 7M13 uses a trick
where it diode ORs TS2-10 together, thus requiring only 2 pins of input,
TS1 to flag start of readout, and all the subsequent timeslots simply
increment a counter..."
Haha, it turns out I did pretty much the same thing. I couldn't recall the details, but even better, I found some of my notes on this project, including the schematic of the section. I'm having camera problems right now, but will try to take and send a good picture, or a scan of it soon, maybe by tomorrow.
Ed
|
Re: 7000 test/cal module(s) and backplane breakout board.
You are right about Q1-4, I was in a rush, will fix it tomorrow. I can test things with a simple current mirror, but I usually resort to a Wilson if I am not using matched parts. Just habit, perhaps a bad one. On Tue, Jul 27, 2021 at 23:33 Ed Breya via groups.io <edbreya= [email protected]> wrote: Looks about right, except the PNPs appear upside down. If you're going
with emitter ballast resistors in the mirror, then you probably don't need
transistor matching or the Wilson configuration, and two NPNs per mirror
would suffice. The compliance is pretty much constant since the row and
column nodes hold very close to ground - there's hardly any Early effect.
Ed
-- Andy
|
Re: 7000 test/cal module(s) and backplane breakout board.
Looks about right, except the PNPs appear upside down. If you're going with emitter ballast resistors in the mirror, then you probably don't need transistor matching or the Wilson configuration, and two NPNs per mirror would suffice. The compliance is pretty much constant since the row and column nodes hold very close to ground - there's hardly any Early effect.
Ed
|
Re: 7000 test/cal module(s) and backplane breakout board.
Also, here is the dimensional drawing of the outline for my breakout board:
it seems to fit well with the plug-in frames that I have.
I make no apology for the mixed units - the connector and board width are
exact inch dimensions, and are shown that way.
All other dimensions were measured and I work in metric.
I hope this helps and can be of use to others making 7000 series plug-ins.
toggle quoted message
Show quoted text
On Tue, Jul 27, 2021 at 10:15 PM Andy Warner <andyw@...> wrote: Here is the schematic for the digitally controlled current sink:
That needs to be duplicated for each row/column.
Using 1% E96 resistors and unmatched transistors, it is accurate to
within +/- 10uA in my testing, so I do not think I need to resort to
matched parts.
Values of R5-R8 would need to be tweaked if your logic was 3.3V, but I
plan to design to 5V parts.
On Tue, Jul 27, 2021 at 6:26 PM Andy Warner <andyw@...> wrote:
Simple mistake on the prototype wiring was the cause of the poor TS pulse
detection previously reported.
Here is what it should look like:
This is basically the TS to TTL conversion circuit from the 7M13 readout
module (correctly wired this time.)
Now we have nice sharp edges, and about 35us from the rising edge of the
blue trace (the logic level output) to set up the current sinks
appropriately, which I think is plenty of time.
I am planning to use SPI port expanders, to reduce the number of I/O pins
required on the microcontroller, and also make the circuit more universal,
so I am planning to use an input pin per timeslot. The 7M13 uses a trick
where it diode ORs TS2-10 together, thus requiring only 2 pins of input,
TS1 to flag start of readout, and all the subsequent timeslots simply
increment a counter, no need for interpolation. However, I am planning to
give each timeslot its own pin, that way the circuit can diagnose missing
timeslots and other issues.
On Tue, Jul 27, 2021 at 4:39 PM Ed Breya via groups.io <edbreya=
[email protected]> wrote:
I think you'll need at least two TS detectors (like maybe the first and
last), so you can figure out the actual running frequency and interpolate
the rest of the slots. In my readout exerciser, it is all inside the
plug-in, and uses all of the TSs for direct coding, so it was
straightforward - except for needing 14 interface signal lines.
Ed
--
Andy
--
Andy
--
Andy
|
Re: 7000 test/cal module(s) and backplane breakout board.
Here is the schematic for the digitally controlled current sink:
That needs to be duplicated for each row/column.
Using 1% E96 resistors and unmatched transistors, it is accurate to
within +/- 10uA in my testing, so I do not think I need to resort to
matched parts.
Values of R5-R8 would need to be tweaked if your logic was 3.3V, but I plan
to design to 5V parts.
toggle quoted message
Show quoted text
On Tue, Jul 27, 2021 at 6:26 PM Andy Warner <andyw@...> wrote: Simple mistake on the prototype wiring was the cause of the poor TS pulse
detection previously reported.
Here is what it should look like:
This is basically the TS to TTL conversion circuit from the 7M13 readout
module (correctly wired this time.)
Now we have nice sharp edges, and about 35us from the rising edge of the
blue trace (the logic level output) to set up the current sinks
appropriately, which I think is plenty of time.
I am planning to use SPI port expanders, to reduce the number of I/O pins
required on the microcontroller, and also make the circuit more universal,
so I am planning to use an input pin per timeslot. The 7M13 uses a trick
where it diode ORs TS2-10 together, thus requiring only 2 pins of input,
TS1 to flag start of readout, and all the subsequent timeslots simply
increment a counter, no need for interpolation. However, I am planning to
give each timeslot its own pin, that way the circuit can diagnose missing
timeslots and other issues.
On Tue, Jul 27, 2021 at 4:39 PM Ed Breya via groups.io <edbreya=
[email protected]> wrote:
I think you'll need at least two TS detectors (like maybe the first and
last), so you can figure out the actual running frequency and interpolate
the rest of the slots. In my readout exerciser, it is all inside the
plug-in, and uses all of the TSs for direct coding, so it was
straightforward - except for needing 14 interface signal lines.
Ed
--
Andy
--
Andy
|
Re: 5642 rectifier tubes in a Tek 535: Replace or swap for silicone?
I agree Harvey and Silicon Valley is close to Palo Alto/ Sunnyvale/ Mountain View area.
oops back to the original thread.
(I have a 575 just obtained that is going to need work the only one upon quick look is a 5651 that has its gas missing...zener will not be stable enough under temps.)
..It is currently stored at a friends so i have been looking at the circuits seems to be very similar and I can see silicon diodes should do just fine if the tubes become unobtainable. I would think 100ma more than enough for that application. the 5642 tube has 250ma max.
¸é±ð²Ô¨¦±ð
toggle quoted message
Show quoted text
On 7/27/21 7:04 PM, Harvey White wrote: Silicone valley is obviously located in Hollywood.
Harvey
On 7/27/2021 9:39 PM, Dominic O'Kelly wrote:
Ha! Oh dear... I wrote that on my phone so perhaps I can blame autocorrect.... either way it's a good case for proof reading one's work.
Yes I'd imagine silicone valley is in a different part of California... more focused on the front end work ;)
|
Martin
saipan59 (Pete)
Dave Peterson
mikecaa