Date   

Re: [dead tek 485] Can i fix it ? (newbie here)

Tom Lee
 

Hi Jeff,

Yes, the 2465 targeted the same market segment as did the 485 ("Give me the highest BW in a portable; I've got the budget for it"), and there was some continuity in engineering: Wink Gross, who had partnered with John Addis on the 485, designed the 2465's vertical amp. The 2465 team aimed for higher precision, a modest increase in bandwidth, and far fewer factory adjustments, by taking full advantage of them newfangled 1's and 0's.

The 485's fast cal gen is a very useful feature. It's also a nice little circuit for illustrating to students how one approaches the design of a special-purpose pulse amplifier differently from that of a general-purpose linear amplifier. There are other things, too, such as John implementing a 4.3k resistor as a series connection of two 2.15k resistors to realize a higher-wattage resistance with lower capacitance, the choice of very low output-capacitance transistors, etc.

--Cheers,
Tom

--
Prof. Thomas H. Lee
Allen Ctr., Rm. 205
350 Jane Stanford Way
Stanford University
Stanford, CA 94305-4070
http://www-smirc.stanford.edu

On 1/10/2021 14:11, Jeff Dutky wrote:
Tom,

I was only pointing to the Jim Williams pulser out of sensitivity to Jose's possible financial limits. I agree that the Leo Bodnar pulser is an excellent investment, in both time and money. I might still be trying to get my own Williams pulser working now, two months later, if I hadn't "sent Leo a check" instead.

I did not realize that the 485 had a fast rise calibration signal built in. Again, I'm feeling like I might need to acquire a 485.

Should we consider the 2465 (which has a bandwidth similar to the 485, as well as a multi-frequency calibration signal, and 50 ohm inputs with input protection) as the follow-on to the 485? I had been specifically interested in the 485 after you mentioned that it had input protection, and had tried researching the circuitry for that, but had been disappointed that the service manual only showed it being implemented in a custom IC (155-0076-00). Happily, the internal construction of that IC is detailed on a TekWiki page (https://w140.com/tekwiki/wiki/155-0076-00), and is fascinating reading.

-- Jeff Dutky




Re: Tek 576 step generator problem

Chuck Harris <cfharris@...>
 

The filter capacitor for the 5V logic section is
known to be bad, almost always, with these curve
tracers. It is about 10K uf, and fairly low voltage,
and it becomes open circuit.

When the 5V supply capacitor is bad, it causes very
erratic operation.

-Chuck Harris

hifi-workshop@gmx.at wrote:

Hello Guys,

i need help with my Tek 576.
Theres something pretty strange with the step generator... On some settings of the number of steps the lines start to flicker/jump around like something is not synced right.
I uploaded a little video where you can observe the video (first its set to 5 Steps, then i change the knob to 6 steps and the lines go crazy)

https://youtu.be/DeYiZHzLRo4

Even when its set to 5 every 2nd line of the measurement seems faulty aswell.

Did everyone of you ever experience this problem? Could it be a problem of the output stage of the step gen?

Hope someone can help me
BR,
David






Re: Type 503 — power supply question

brian.kane@...
 

Thanks to all for your advice and wisdom. My previous experience with this kind of thing has been working on restoring radios, test equipment, wire recorders, etc. This is my first oscilloscope.

I have a good sense now of what's involved in putting in a filment transformer for the CRT. I appreciate all the advice, pro and con, about whether it is worth doing.

Regarding Dan's question: I'm a professor at Yale. I teach in the Department of Music and in the Program in Film and Media Studies. I have a strong background in electronic music, but nowadays I'm mostly an historian and theorist. I'm involved in a research project that involves radio and the roots of do-it-yourself electronics. I'm very interested in the informal ways that a whole generation or two of musicians (both popular and avant-garde) learned about electrical engineering and audio engineering through tinkering with radio, hanging out at radio repair shops, reading magazines and Rider manuals, etc. etc.

This is an amazing forum. Thank you for all of your help.

--Brian


Re: I built a TM500 mainframe tester, and updated the design. Someone might find this useful?

Larry McDavid
 

This is another update on the group order for circuit board sets for Jared Cabot's TM500 power supply tester.

I just yesterday remembered there is a Tek500 groups.io Group specifically for TM500 equipment. I asked there for interest in the board sets and got several requests to be included; that has expanded the list to 63 committed and still 2 more with incomplete information or lack of commitment. Please check the two lists of name initials below to ensure you are shown correctly.

These 63 have committed to order board sets:
ML
ML
ML
PA
SM
DC
HB
GL
MH
JR
WR
JT
EW
EW
EW
EW
RD
AW
SG
DS
JF
MU
PE
SN
SC
AA
BH
EPa
WS
KT
RDD
PM
GM
CC
BH
BK
LR
MP
MP
MP
SA
BG
JK
AA
PTC
PD
MU
BO
JG
EP
BL
CT
MD
RG
CB
CH
DM
AD
JR
JR
CR
PM
LM

These 2 are not committed at this time:
rrrr
MK

It is interesting that, while most of these are in USA, there are also members in UK, Canada, Australia, Spain and Germany.

In light of the size of the circuit board order remember that I must ask for prepayment by PayPal Friends and Family before I can place the board order. If you have any problem using PayPal Friends and Family as a means of payment, please contact me off-list.

Final board set cost is still being worked but the estimate is $38. Packaging and shipment to CUSA will be $10; international shipping will be higher and I have discussed that with those several.

Note that Jared has released another update to the Gerber files.

Larry McDavid

On 12/31/2020 11:27 AM, Jared Cabot via groups.io wrote:
Hi all,
I have been accumulating TM500 modules and mainframes for a little while, so I recently finished building an 067-1201-99 TM500 mainframe tester from the Tek construction notes floating around.
I updated the design somewhat and designed it for simplified construction (Not a single wire to strip and solder!) and I thought someone here might be interested in taking a look or even building one themselves.
Here's a link to the youtube video:
https://www.youtube.com/watch?v=Afwqtc6Fxd4
And here's a link to the google drive with all the gerbers and schematics and stuff. I went so far as to rewrite the manual with the alterations and corrections included.
https://drive.google.com/drive/folders/1yj52Z_BtGZ7Q9BB2jS9Lf2Uz39turS4-...
--
Best wishes,

Larry McDavid W6FUB
Anaheim, California (SE of Los Angeles, near Disneyland)


Re: [dead tek 485] Can i fix it ? (newbie here)

 

On Sun, Jan 10, 2021 at 11:11 PM, Jeff Dutky wrote:


I was only pointing to the Jim Williams pulser out of sensitivity to Jose's
possible financial limits.
I guess a lot of people went the Jim Williams pulser-way and drew conclusions that weren't warranted, especially before Leo's pulser became available (not too many years ago). It's a waste of effort as a replacement tool for adjusting a 'scope's BW by step response, especially without a charge line which, as Tom indicates, needs to be adjusted to be useful - and for that you need a fast 'scope!
OTOH, looking at Jim Williams' pulse is about as much fun to see on a not so well-adjusted 'scope as on one that is, unless you know what the pulse really is like, in amplitude and shape.

Raymond


Re: [dead tek 485] Can i fix it ? (newbie here)

 

On Sun, Jan 10, 2021 at 11:11 PM, Jeff Dutky wrote:


Should we consider the 2465 (which has a bandwidth similar to the 485, as well
as a multi-frequency calibration signal, and 50 ohm inputs with input
protection) as the follow-on to the 485?
The 2465's calibrator is nothing like the 485's. It doesn't generate a fast edge and modifies its frequency, depending on the time base speed selected. This in most cases (and in all mine) is just a nuisance.

Raymond


Re: Tek 576 step generator problem

 

On Sun, Jan 10, 2021 at 08:36 PM, Dave S. wrote:


Theres something pretty strange with the step generator... On some settings of
the number of steps the lines start to flicker/jump around like something is
not synced right
You could try this: http://www.hakanh.com/dl/docs/hardtofind/576_mod41067.pdf
/Håkan


Re: [dead tek 485] Can i fix it ? (newbie here)

 

Tom,

I was only pointing to the Jim Williams pulser out of sensitivity to Jose's possible financial limits. I agree that the Leo Bodnar pulser is an excellent investment, in both time and money. I might still be trying to get my own Williams pulser working now, two months later, if I hadn't "sent Leo a check" instead.

I did not realize that the 485 had a fast rise calibration signal built in. Again, I'm feeling like I might need to acquire a 485.

Should we consider the 2465 (which has a bandwidth similar to the 485, as well as a multi-frequency calibration signal, and 50 ohm inputs with input protection) as the follow-on to the 485? I had been specifically interested in the 485 after you mentioned that it had input protection, and had tried researching the circuitry for that, but had been disappointed that the service manual only showed it being implemented in a custom IC (155-0076-00). Happily, the internal construction of that IC is detailed on a TekWiki page (https://w140.com/tekwiki/wiki/155-0076-00), and is fascinating reading.

-- Jeff Dutky


File /Tek 5S14N Sampler Instruction &amp; Service Manual sm_OCR.pdf uploaded #file-notice

TekScopes@groups.io Notification <noreply@...>
 

The following files have been uploaded to the Files area of the TekScopes@groups.io group.

By: Rick

Description:
Tek 5S14N Instruction & Service Manual


Re: [dead tek 485] Can i fix it ? (newbie here)

Tom Lee
 

Yes, a very fast edge (= high bandwidth source) will reveal many blemishes! A slower edge gives you a picture that is equivalent to low-pass filtering the result of a fast-edge excitation, so it will look cleaner.

Tom

--
Prof. Thomas H. Lee
Allen Ctr., Rm. 205
350 Jane Stanford Way
Stanford University
Stanford, CA 94305-4070
http://www-smirc.stanford.edu

On 1/10/2021 13:37, Raymond Domp Frank wrote:
On Sun, Jan 10, 2021 at 10:00 PM, Tom Lee wrote:

Just send Leo a check and buy one of his pulsers (which aren't avalanchers).
You'll be glad you did.
I did and I am. Several years ago, I built a tunneldiode-driven pulser. It uses a 1 MHz quartz oscillator as a clock and generates a pretty clean step of 180-200 ps. I used an S-4 (25 ps nominal rise time) to check and optimize. Since I wanted something faster, I bought Leo's pulser as well; the slightly slower BNC-version. I already had his GPS-synchronized reference oscillator. Another nice toy and no, no affiliation, just a happy customer.
With some 'scopes, edge adjustment is a bit more difficult with such a fast edge and the first ns ends up being a little less clean, i.e. more bumpy than with slightly slower step generators, I guess because of the phase characteristics of their vertical amp.
Several samplers (7854/7T11A/7S11 and S-4, S-6) that I observed Leo's sampler with showed a pretty nice edge, with a slightly slow top 2 - 5 % amplitude part of the shoulder.

Raymond




Re: [dead tek 485] Can i fix it ? (newbie here)

 

On Sun, Jan 10, 2021 at 10:00 PM, Tom Lee wrote:


Just send Leo a check and buy one of his pulsers (which aren't avalanchers).
You'll be glad you did.
I did and I am. Several years ago, I built a tunneldiode-driven pulser. It uses a 1 MHz quartz oscillator as a clock and generates a pretty clean step of 180-200 ps. I used an S-4 (25 ps nominal rise time) to check and optimize. Since I wanted something faster, I bought Leo's pulser as well; the slightly slower BNC-version. I already had his GPS-synchronized reference oscillator. Another nice toy and no, no affiliation, just a happy customer.
With some 'scopes, edge adjustment is a bit more difficult with such a fast edge and the first ns ends up being a little less clean, i.e. more bumpy than with slightly slower step generators, I guess because of the phase characteristics of their vertical amp.
Several samplers (7854/7T11A/7S11 and S-4, S-6) that I observed Leo's sampler with showed a pretty nice edge, with a slightly slow top 2 - 5 % amplitude part of the shoulder.

Raymond


Re: Tek 576 step generator problem

Dave S.
 

Yes i tried to clean it already.. nothing changed at all.

I just checked it again... the problem of the "jumping/unsynced" lines occurs on step-setting 2, 4, 6, 8 and 10


Re: Tek 576 step generator problem

Mlynch001
 

Does the unit flicker on the odd numbered settings? These things typically flicker visibly on the even numbered settings. This flicker is almost nonexistent on the odd numbered settings.
--
Michael Lynch
Dardanelle, AR


Re: [dead tek 485] Can i fix it ? (newbie here)

Tom Lee
 

As Raymond observes, the avalanche pulser that everyone likes to copy from Jim (who copied it from Tek) isn't quite suitable by itself. The rise times you measure with the stock pulser can't be legitimately used to infer bandwidth because the waveshape approximates an impulse more than a step. Most of the measurements I've seen online are grossly in error because the operators don't understand that the rise time-bandwidth relationship only holds for a step. Yes, you can add a charge line to widen the pulse, but if you don't own a fast enough (and clean enough) scope, you'll still be uncertain about what you're generating. I don't see how going down the avalanche pulser path really buys you anything here. Trying to tame the infamous "falling-edge wigglies" of an avalancher will consume a lot of your time, too. Just send Leo a check and buy one of his pulsers (which aren't avalanchers). You'll be glad you did.

All that said, the 1MHz generator on the 485 itself is a better choice than an avalanche pulser for most of the tasks you want to perform to check out a 485. It's certainly more convenient! John Addis designed a very clean generator. Pay attention to results obtained with the rising edge only -- it is cleaner than the falling edge. The official rise time spec, < 1ns, is quite conservative. It's been a long time since I made a measurement, but I recall being impressed how much faster it was than it needed to be to verify the 485's bandwidth.

-- Tom

--
Prof. Thomas H. Lee
Allen Ctr., Rm. 205
350 Jane Stanford Way
Stanford University
Stanford, CA 94305-4070
http://www-smirc.stanford.edu

On 1/10/2021 12:08, Raymond Domp Frank wrote:
On Sun, Jan 10, 2021 at 07:37 PM, Jeff Dutky wrote:

A fast risetime pulse generator would be a big help. I mentioned that you can
buy one (for about $75 U.S. from Leo Bodnar), but you might be able to build a
Jim Williams pulser yourself (
http://dangerousprototypes.com/blog/2013/07/20/avalanche-pulse-generator-and-some-scope-porn/
), which will give you a reasonable stand-in for the time mark generator.
Jim Williams' pulse generator produces a steep rising slope, followed by a less steep falling one. The problem with that is that you can really only see a slew rate, not a rise time, because
1. with a 'scope that is slow by comparison, the pulse will again be on its way down while your 'scope is still before showing the pulse's top (and you normally do not know because you don't know the pulse's full amplitude
2. with a 'scope that is fast enough, you can only conclude that it is fast enough (if you know the pulse's real amplitude)
This is the reason why correct application of Jim Williams' pulse generator is only possible if the pulse is lengthened by a charge line, which can simply be a short piece of coax.

A nice write-up with pictures is to be found here:

https://www.pa4tim.nl/?p=3076

Raymond




Re: Tek 576 step generator problem

-
 

Did you try cleaning the step selector switch? Does it do that in every
position?

On Sun, Jan 10, 2021 at 2:37 PM <hifi-workshop@gmx.at> wrote:

Hello Guys,

i need help with my Tek 576.
Theres something pretty strange with the step generator... On some
settings of the number of steps the lines start to flicker/jump around like
something is not synced right.
I uploaded a little video where you can observe the video (first its set
to 5 Steps, then i change the knob to 6 steps and the lines go crazy)

https://youtu.be/DeYiZHzLRo4

Even when its set to 5 every 2nd line of the measurement seems faulty
aswell.

Did everyone of you ever experience this problem? Could it be a problem of
the output stage of the step gen?

Hope someone can help me
BR,
David






Re: [dead tek 485] Can i fix it ? (newbie here)

 

On Sun, Jan 10, 2021 at 07:37 PM, Jeff Dutky wrote:


which will give you a reasonable stand-in for the time mark generator.
To be usable as a time mark generator, a settable, known and stable period is necessary. While it is possible to sync Jim Williams' pulse generator externally, the simple version is not stable at all.
Also, to be usable with longer intervals, the time marks (needles) will have to be wide enough to be seen at those slower (time base) speeds. A pulse-extender using a charge-line, like the one I mentioned in my earlier post, will only do for not-so-long periods, so not suitable for 'scope time base adjustment in general.

Raymond


Re: [dead tek 485] Can i fix it ? (newbie here)

 

On Sun, Jan 10, 2021 at 07:37 PM, Jeff Dutky wrote:


A fast risetime pulse generator would be a big help. I mentioned that you can
buy one (for about $75 U.S. from Leo Bodnar), but you might be able to build a
Jim Williams pulser yourself (
http://dangerousprototypes.com/blog/2013/07/20/avalanche-pulse-generator-and-some-scope-porn/
), which will give you a reasonable stand-in for the time mark generator.
Jim Williams' pulse generator produces a steep rising slope, followed by a less steep falling one. The problem with that is that you can really only see a slew rate, not a rise time, because
1. with a 'scope that is slow by comparison, the pulse will again be on its way down while your 'scope is still before showing the pulse's top (and you normally do not know because you don't know the pulse's full amplitude
2. with a 'scope that is fast enough, you can only conclude that it is fast enough (if you know the pulse's real amplitude)
This is the reason why correct application of Jim Williams' pulse generator is only possible if the pulse is lengthened by a charge line, which can simply be a short piece of coax.

A nice write-up with pictures is to be found here:

https://www.pa4tim.nl/?p=3076

Raymond


Re: [dead tek 485] Can i fix it ? (newbie here)

Alain Mionnet
 

Hello the gang,

I am here since a long time but due to the yahoogroup transfer and no udapted e-mail I may only receive a digest ( now upgraded)
I have been in contact direct by e-mail with MARCOSJL31 + on phone yesterday afternoon (Paris time)
My first Tek scope* is a 453 around 1980 ? (first contact with in 1972 as a student stage in DANA** Laboratories France))
As student it had seen a demo of a 7000 scope in 1971? in the class of the school (at this time it was scopes from Telequipment , Philips..)

For this topic:
* C2018 has been replaced by an electrolytic cap 100uF 100V
* c2021 has been replaced by an electrolytic cap 47uF 5V

C2021 47µF is not 5v but well 50v but both capacitors had been replaced by chemical, unknown specs.
This must be verified and may be corrected. Most are Tantalum sealed case (military class and very expensive) but some epoxy dipped model (color coded)....
Some chemical are Sprague, certainly good reliability.(as seen on a picture received)
Since the 9v is good it is no more power on trouble (PSU security circuit as described by Reed Dickinson here: https://groups.io/g/TekScopes/topic/20933356)
Recently I re-powered a tek 465 that I had, where I find only bad contacts (as on beamfinder...), no caps trouble....

It is a pleasure to help a Tek Scope owner....

Alain Mionnet
FRANCE
https://groups.io/g/advantestinstrument
owner

* my first contact with a scope should have been in 1969 , it was a primitiv 0-200 KHz Philips with tubes and mono trace like PM 5605.This was at the laborshop of my uncle....(sales and repairs for radio + Tv) , later on a new one with transistors, PM 3200 for theTV team.(still Philips, normal was Philips consumer dealer..)

**DANA LABS I think that must say something here! (Bought by Racal Instruments in 1977)


Tek 576 step generator problem

Dave S.
 

Hello Guys,

i need help with my Tek 576.
Theres something pretty strange with the step generator... On some settings of the number of steps the lines start to flicker/jump around like something is not synced right.
I uploaded a little video where you can observe the video (first its set to 5 Steps, then i change the knob to 6 steps and the lines go crazy)

https://youtu.be/DeYiZHzLRo4

Even when its set to 5 every 2nd line of the measurement seems faulty aswell.

Did everyone of you ever experience this problem? Could it be a problem of the output stage of the step gen?

Hope someone can help me
BR,
David


A different way to make Tek TM500/5000 plug-ins

 

With the discussion of TM500 test plug-ins, I thought it was worthwhile to mention a different approach that is easily implemented to make the mechanical housings, since usable scrap TM500 plug-ins are hard to find these days.. Some years ago I looked at this problem, and realized that a common technique used in the avionics control world would work perfectly for this task, boxes made from tapped square stand-offs.

RAF Hardware makes square cross-section female stand-offs of any length.
you can see them here, and down load the catalog (ignore the picture of the hex parts):

https://www.rafhdwe.com/catalog/female-standoffs

The square ones are not obvious on the site, but they exist,and are incredibly handy for making enclosures. and I got many long samples to experiment with, and made a mock-up of the TM500 case easily. I will see if I can track ti down to provide pics. I( can't remember now if I used 3/16" or 1/4" parts.

To use them to make a housing, it requires 6 pieces, three at the top, with the left hand ones close together to form the narrow top slo fro the frame guide tab, and three at the bottom with the left ones farther apart to bracket the black plug in plastic guide at the bottom of the frame. The faceplate was solid aluminum, with six counter-sunk holes to attach the 6 stand-offs, and the rear is also a similar plate, with a cutout for the card to pass through. It (the rear panel) could be easily made from a PCB, making the slots and holes much easier to fabricate, and quite inexpensive. In this way, little fabrication skill is required, as the front is just a simple sheared plate, as are the sides (in thin material), and the rear is a fabbbed PCB. Interior boards are easily attached to the square stand-offs with tapped holes (with small internal stand-offs to get the riight exact spacing for the mating pcb connector in the frame). Needless to say, a double wide plug-in is just wider, but still needs at least the extra bottom slot stand-offs for alignment. The open top and bottom dramatically improves air circulation and heat removal over the stock box.

The side covers (if needed) can just be flat sheets with 4 countersunk holes, to be screwed to the outside square stand-offs into tapped holes. I did not find the lock essential at all, and I added a small pull handle to the front instead to more easily remove the plug-in, as this is especially relevant for the frame tester unit which is constantly removed and inserted. The lock is just a nuisance in that application. This structure is much more solid that the usual TM500 case due to the rear panel and rigidly attached sides.

Anyway, just passing along my previous escapades in this area, and I should point out that this method really lends itself to local assembly and low cost. If I can track down my original prototype, I will post some pics and details. It's all stuffed away in a box somewhere.

all the best,
walter (walter2 -at- sphere.bnc.ca)
sphere research corp.
https://www.sphere.bc.ca/test/index.html

8501 - 8520 of 184569