Topics

Help replacing a resistor/inductor in SG503


Jared Cabot
 

Hi all,
I recently got myself an SG503 Leveled Sine Wave generator for a good price and I am currently repairing it back to operational condition.

I found a few components in the oscillator section were damaged, from what I guess, is from some ham-fisted poking from someone's previous repair attempt.
Most parts were easy to replace, but I'm a bit stuck on one. LR140 is a 20nH inductor consisting of a couple turns of fine wire around a 1/8w 47ohm carbon composite resistor.
It looks like it has been poked at and one leg was broken off.
Below is linked a page from the schematics with the part highlighted in red.
https://groups.io/g/TekScopes/photo/260912/0/SG503%20Oscillators%20LR140%20Detail.jpg

My question is, How's the best way to recreate this sort of part? I tried winding a few turns on a 1/8 metal film resistor and didn't have much luck, I have a DE-5000 LCR meter and a HP 4276A etc but it's a little difficult to measure such small values reliably.
Should I get a carbon composite resistor from Digikey/Mouser to recreate the part more true to the original design instead of the metal film I used? Or is a plain 20nH axial inductor without the resistor ok to use? Any other ideas or suggestions?

Thanks!


Stephen
 

In my spare time I’m a guitar pickup winder. Been doing that for over 35 years. I have very fine wire available, and tons of Allen Bradley carbon comp resistors.
I could probably build you one.


Ed Breya
 

If you have the original, broken part, then you have a good reference point. The damping resistance may be important, or it may be just a handy coil form, but it's hard to tell. It's best to just replicate it reasonably closely.

Ed


Jared Cabot
 

That would be awesome and greatly appreciated! :) The wire looks to be around 32 or 33 AWG.
If you are able to whip something up, are you able to send it to Japan? I can paypal you any costs involved.

Thanks!
Jared.

On Fri, Feb 19, 2021 at 01:39 AM, Stephen wrote:


In my spare time I’m a guitar pickup winder. Been doing that for over 35
years. I have very fine wire available, and tons of Allen Bradley carbon comp
resistors.
I could probably build you one.


Stephen
 

On Thu, Feb 18, 2021 at 05:51 AM, Jared Cabot wrote:


That would be awesome and greatly appreciated! :) The wire looks to be around
32 or 33 AWG.
If you are able to whip something up, are you able to send it to Japan? I can
paypal you any costs involved.

Thanks!
Jared.


On Fri, Feb 19, 2021 at 01:39 AM, Stephen wrote:


In my spare time I’m a guitar pickup winder. Been doing that for over 35
years. I have very fine wire available, and tons of Allen Bradley carbon
comp
resistors.
I could probably build you one.
I have an SG503. I’ll take a look at that inductor and see what I can come up with. Don’t worry about the cost for now.
AWG 32 or 33 are pretty thick compared to what I usually work with (AWG 42 and 43). These are hair thin.
Give me time to investigate.


-
 

Winding inductors on the body of a carbon comp resistor is an OLD
practice that you very seldom see any more. I don't know how well it would
work on a film resistor but I would think that the film might affect it's
inductance. Also I've never seen one wound on that low of a resistance.
Are you SURE that the coil is open and that you're not reading the DC
resistance of the coil and not the resistor? Usually the resistance of the
resistors used in that practice are 100,000 Ohms or more. Several orders of
magnatude more than the coil.

Yes, a 2nH coil should be fine. Everytime that I've seen a coil wound on
a resistor body, the resistor resistance was FAR higher than the inductance
of the coil and the resistor had no effect on the circuit..

Also the film resistors have a very different inductance than the carbon
comp resistors so that could be interferring with your circuit as well.

On Thu, Feb 18, 2021 at 11:30 AM Jared Cabot via groups.io <jaredcabot=
protonmail.com@groups.io> wrote:

Hi all,
I recently got myself an SG503 Leveled Sine Wave generator for a good
price and I am currently repairing it back to operational condition.

I found a few components in the oscillator section were damaged, from what
I guess, is from some ham-fisted poking from someone's previous repair
attempt.
Most parts were easy to replace, but I'm a bit stuck on one. LR140 is a
20nH inductor consisting of a couple turns of fine wire around a 1/8w 47ohm
carbon composite resistor.
It looks like it has been poked at and one leg was broken off.
Below is linked a page from the schematics with the part highlighted in
red.

https://groups.io/g/TekScopes/photo/260912/0/SG503%20Oscillators%20LR140%20Detail.jpg

My question is, How's the best way to recreate this sort of part? I tried
winding a few turns on a 1/8 metal film resistor and didn't have much luck,
I have a DE-5000 LCR meter and a HP 4276A etc but it's a little difficult
to measure such small values reliably.
Should I get a carbon composite resistor from Digikey/Mouser to recreate
the part more true to the original design instead of the metal film I used?
Or is a plain 20nH axial inductor without the resistor ok to use? Any other
ideas or suggestions?

Thanks!






Roy Thistle
 

On Thu, Feb 18, 2021 at 08:30 AM, Jared Cabot wrote:


is a plain 20nH axial inductor without the resistor ok to use?
It's labeled LR140, and its manufacturer is manufacturer 8009... that's Tektronix.
To me, the LR designation indicates the R is important for the circuit... but, that's my analysis.
Just the same... the "golden rule"... well espoused here on TekScopes, by cranky seniors... is... if you don't know what the part does in the circuit... then don't alter the component.
Forming a low value inductance (nH), around a resistor body, and soldering the ends to the resistor leads... used to be a common way to wind small value inductors... using the resistor body as a coil form... you'd see this in Ham radio RF construction... all the time... where you'd see specd.. so many turns, of such and such diameter enameled cu wire, around a CC resistor, of so and so a wattage [just because CC resistors were common, and their body dimensions...dependent on their heat dissipation/wattage were well specd... so you could get/construct the inductor value specd ... of reasonably good tolerance. This method of low value (nH) coil winding/forming was also frequently used by manufacturers too, before automation.
I'd measure the wire diameter (with a mic)... also, the number of turns of the wire .. and the resistor body diameter with a vernier ... the winding spacing too... so as too calculate... at least... the value of the "single turn wound, air core inductor." If you know that calculated value in nH... you can compare tp what Tektronix has specd, for LR140.
Posting a pic of the part may help.


 

I feel like we had this same conversation when I was looking at replacing some CC resistors in the Z-axis amp of my 475A a couple months ago, but shouldn't a metal film resistor have some inherent inductance of it's own (because of the spiral cut in the film)?

Also, it looks like the part might be sensitive to very small (i.e. single ohm) changes in the resistor value, so it could well be that the CC resistor has drifted out of spec.

-- Jeff Dutky


Stephen
 

On Thu, Feb 18, 2021 at 06:29 AM, - wrote:

Also I've never seen one wound on that low of a resistance.
Are you SURE that the coil is open and that you're not reading the DC
resistance of the coil and not the resistor? Usually the resistance of the
resistors used in that practice are 100,000 Ohms or more. Several orders of
magnatude more than the coil.
The schematic clearly shows a 47R... 🤷‍♂️


Jean-Paul
 

Dear Jared: The parisitci chokes in radio and amateur transdmitters use this same construction, to create a damped LR circuit.

You will need a 47 Ohm 1/2 W carbon compr R,

The 47 Ohm is to damp the L and the LR combination is required.

The L will be duplicated if you layer wind the same number of turns of same wire gauge, (for same turns per cm) and the resistor body is the same length and diameter.

But it should be a 1/2W carbon and not MF, or other sizes/types.

I have some stock of old Allen Bradley hot molded parts far superior to usual Chines copies nowadays. I could post one or a few to you!



Bon Chance,



Jon


Roy Thistle
 

On Thu, Feb 18, 2021 at 10:13 AM, Jeff Dutky wrote:


I feel like we had this same conversation when I was looking at replacing some
CC resistors in the Z-axis amp of my 475A a couple months ago, but shouldn't a
metal film resistor have some inherent inductance of it's own (because of the
spiral cut in the film)?
Everything that conducts, has inductance... which, I sure you know.
That are "kinds" of metal film resistor technology; some with lower inductances than others.
Then too, they laser trim the deposited metal film, into the appropriate helix: at least, increasing the number of turns, to increase the resistance (which increase the inductance.)
Beside the question of inductance, I guess it bears repeating that cc resistors have, compared to other common resistor technologies) the best peak pulse power ratings because of their body bulk, thermal conduction, and power/(surface area).
How did the Z-axis thing work out?


Tom Lee
 

Hi Jared,

You've gotten a lot of advice, not all of it self-consistent...

Although resistors are certainly sometimes used as coil forms, that's not the sole purpose of the resistor here; the fact that it is 47 ohms instead of 47k, for instance, tells us that. As Ed said, tt's part of a damping network to quench a troublesome parasitic oscillation at other than the desired mode. The precise value of the inductance wound around that resistor isn't critical, and neither is the resistance value. Something that is eyeball-close to the original would be fine. Do use a carbon comp here, simply to ensure that you don't get some odd effects from a core that isn't magnetically transparent.

If you can't obtain exactly the same gauge wire wound to the exact same pitch, don't worry. You can get a surprisingly good estimate of the inductance from Wheeler's nearly century-old formula:

L = n^2*r^2/(9r + 10l), where L comes out in microhenries if the radius r and the length l are in inches.

For inductors whose lengths are at least as great as the diameter, the formula typically has an error of 5% or less, provided that the total length of the wire used to wind the inductor is a tiny fraction of the shortest wavelength of interest.

That should be enough for you to roll your own successfully.

-- 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 2/18/2021 08:30, Jared Cabot via groups.io wrote:
Hi all,
I recently got myself an SG503 Leveled Sine Wave generator for a good price and I am currently repairing it back to operational condition.

I found a few components in the oscillator section were damaged, from what I guess, is from some ham-fisted poking from someone's previous repair attempt.
Most parts were easy to replace, but I'm a bit stuck on one. LR140 is a 20nH inductor consisting of a couple turns of fine wire around a 1/8w 47ohm carbon composite resistor.
It looks like it has been poked at and one leg was broken off.
Below is linked a page from the schematics with the part highlighted in red.
https://groups.io/g/TekScopes/photo/260912/0/SG503%20Oscillators%20LR140%20Detail.jpg

My question is, How's the best way to recreate this sort of part? I tried winding a few turns on a 1/8 metal film resistor and didn't have much luck, I have a DE-5000 LCR meter and a HP 4276A etc but it's a little difficult to measure such small values reliably.
Should I get a carbon composite resistor from Digikey/Mouser to recreate the part more true to the original design instead of the metal film I used? Or is a plain 20nH axial inductor without the resistor ok to use? Any other ideas or suggestions?

Thanks!




 

Roy,

The Z-axis amp fix worked out quite well. I did NOT end up replacing the two CC resistors standing on end, not least because it was too difficult to get a soldering iron in between the bottom ends of the resistors and the pads they are soldered to (also because I was assured, and if performance is any indicator the assurances were correct, that the specific value of those resistors was not vital the circuit).

-- Jeff Dutky


Jared Cabot
 

Hi all,
The resistor in question is definitely 47ohm, I was able to read the colour code on the resistor, so I assume it is required for damping or the like instead of just a former.
I think I'll try to replicate the original part as close as possible to avoid chasing my tail.

I wonder if lower value resistors in this application are critical enough that the usual carbon comp resistor value drift is a concern? Should I look at the other inductors of this construction type or avoid poking the bear? There are 5 or so of varying values around the place.


John Kolb
 

Looking at the schematic, there are a number of LR's and they are wound on different value resistors. Noting that LR142 and LR150 are both 0.25uH/51 ohms and LR135 and LR19? are both 2.5uH/2560 ohms, it may be that Tek was only using the resistor value to identify the component. Nevertheless you should duplicate the broken part as exactly as possible. With the bad part intact except for the broken lead, with even a cheap vernier caliper, you should be able to match the part very well.

On 2/18/2021 8:30 AM, Jared Cabot via groups.io wrote:
Hi all,
I recently got myself an SG503 Leveled Sine Wave generator for a good price and I am currently repairing it back to operational condition.
I found a few components in the oscillator section were damaged, from what I guess, is from some ham-fisted poking from someone's previous repair attempt.
Most parts were easy to replace, but I'm a bit stuck on one. LR140 is a 20nH inductor consisting of a couple turns of fine wire around a 1/8w 47ohm carbon composite resistor.
It looks like it has been poked at and one leg was broken off.
Below is linked a page from the schematics with the part highlighted in red.
https://groups.io/g/TekScopes/photo/260912/0/SG503%20Oscillators%20LR140%20Detail.jpg
My question is, How's the best way to recreate this sort of part? I tried winding a few turns on a 1/8 metal film resistor and didn't have much luck, I have a DE-5000 LCR meter and a HP 4276A etc but it's a little difficult to measure such small values reliably.
Should I get a carbon composite resistor from Digikey/Mouser to recreate the part more true to the original design instead of the metal film I used? Or is a plain 20nH axial inductor without the resistor ok to use? Any other ideas or suggestions?
Thanks!


Tom Lee
 

Hi John,

Your speculation might  be reasonable for LR142 vs LR150, although I think that it is more likely that Tek used LR combos that were already in inventory, rather than endure the paperwork in specifying a new part. The resistance value isn't too critical for either LR142 or LR150. As for LR190, though, its resistance actually matters: LR190 participates in a series stack of parallel LR sections to produce a broadband collector load resistance that is shunted at DC to keep the collector swing about VCC. A single parallel LR wouldn't get the job done, owing to the low self-resonant frequency of a 2.5mH choke. As the frequency increases past the self-resonance of one section, another section cuts in to keep the resistance high.

-- 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 2/18/2021 22:17, John Kolb wrote:

Looking at the schematic, there are a number of LR's and they are wound on different value resistors. Noting that LR142 and LR150 are both 0.25uH/51 ohms and LR135 and LR19? are both 2.5uH/2560 ohms, it may be that Tek was only using the resistor value to identify the component. Nevertheless you should duplicate the broken part as exactly as possible. With the bad part intact except for the broken lead, with even a cheap vernier caliper, you should be able to match the part very well.

On 2/18/2021 8:30 AM, Jared Cabot via groups.io wrote:
Hi all,
I recently got myself an SG503 Leveled Sine Wave generator for a good price and I am currently repairing it back to operational condition.

I found a few components in the oscillator section were damaged, from what I guess, is from some ham-fisted poking from someone's previous repair attempt.
Most parts were easy to replace, but I'm a bit stuck on one. LR140 is a 20nH inductor consisting of a couple turns of fine wire around a 1/8w 47ohm carbon composite resistor.
It looks like it has been poked at and one leg was broken off.
Below is linked a page from the schematics with the part highlighted in red.
https://groups.io/g/TekScopes/photo/260912/0/SG503%20Oscillators%20LR140%20Detail.jpg

My question is, How's the best way to recreate this sort of part? I tried winding a few turns on a 1/8 metal film resistor and didn't have much luck, I have a DE-5000 LCR meter and a HP 4276A etc but it's a little difficult to measure such small values reliably.
Should I get a carbon composite resistor from Digikey/Mouser to recreate the part more true to the original design instead of the metal film I used? Or is a plain 20nH axial inductor without the resistor ok to use? Any other ideas or suggestions?

Thanks!








Dave Wise
 

?We were strongly encouraged to use parts already in the Common Design Parts Catalog. ISTR a new part number cost something like $10,000 in administrative overhead.


Dave Wise

Information Display, 1980-1994?

________________________________
From: TekScopes@groups.io <TekScopes@groups.io> on behalf of Tom Lee via groups.io <tomlee=ee.stanford.edu@groups.io>
Sent: Thursday, February 18, 2021 11:02 PM
To: TekScopes@groups.io
Subject: Re: [TekScopes] Help replacing a resistor/inductor in SG503

Hi John,

Your speculation might be reasonable for LR142 vs LR150, although I
think that it is more likely that Tek used LR combos that were already
in inventory, rather than endure the paperwork in specifying a new part.
The resistance value isn't too critical for either LR142 or LR150. As
for LR190, though, its resistance actually matters: LR190 participates
in a series stack of parallel LR sections to produce a broadband
collector load resistance that is shunted at DC to keep the collector
swing about VCC. A single parallel LR wouldn't get the job done, owing
to the low self-resonant frequency of a 2.5mH choke. As the frequency
increases past the self-resonance of one section, another section cuts
in to keep the resistance high.

-- 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<http://www-smirc.stanford.edu>

On 2/18/2021 22:17, John Kolb wrote:

Looking at the schematic, there are a number of LR's and they are
wound on different value resistors. Noting that LR142 and LR150 are
both 0.25uH/51 ohms and LR135 and LR19? are both 2.5uH/2560 ohms, it
may be that Tek was only using the resistor value to identify the
component. Nevertheless you should duplicate the broken part as
exactly as possible. With the bad part intact except for the broken
lead, with even a cheap vernier caliper, you should be able to match
the part very well.

On 2/18/2021 8:30 AM, Jared Cabot via groups.io wrote:
Hi all,
I recently got myself an SG503 Leveled Sine Wave generator for a good
price and I am currently repairing it back to operational condition.

I found a few components in the oscillator section were damaged, from
what I guess, is from some ham-fisted poking from someone's previous
repair attempt.
Most parts were easy to replace, but I'm a bit stuck on one. LR140 is
a 20nH inductor consisting of a couple turns of fine wire around a
1/8w 47ohm carbon composite resistor.
It looks like it has been poked at and one leg was broken off.
Below is linked a page from the schematics with the part highlighted
in red.
https://groups.io/g/TekScopes/photo/260912/0/SG503 Oscillators LR140 Detail.jpg<https://groups.io/g/TekScopes/photo/260912/0/SG503 Oscillators LR140 Detail.jpg>


My question is, How's the best way to recreate this sort of part? I
tried winding a few turns on a 1/8 metal film resistor and didn't
have much luck, I have a DE-5000 LCR meter and a HP 4276A etc but
it's a little difficult to measure such small values reliably.
Should I get a carbon composite resistor from Digikey/Mouser to
recreate the part more true to the original design instead of the
metal film I used? Or is a plain 20nH axial inductor without the
resistor ok to use? Any other ideas or suggestions?

Thanks!









Jared Cabot
 

My next step is finding a 1/8W 47ohm carbon comp resistor or two... Tiny things, about 4mm long..

I don't think I'd even be able to find one in Akihabara here in Tokyo... seems 1/4W and up are far more common around the place.


Jared Cabot
 

And now resistors are ordered from Hi-Fi Collective.
I found they have a large stock of Allen Bradley resistors for a decent price and good shipping rates, so once they arrive, I'll be able to get some fine wire and wind some inductors while holding my tongue at the right angle to get them hopefully close to 20nH.....


Dave Seiter
 

$10K just to pull a new part number???  I guess it depends on what it is and where/how it's used and how inefficient the documentation system is.  I've created complete documentation packages, including all the compliance related items in less than10 minutes.  OTOH, I recall parts at a few companies that easily cost that much (mostly due to silo culture and subsidiaries that didn't want to play ball).  At one place I worked, they routinely created part numbers for the babies of employees (as a joke, of course).
-Dave

On Friday, February 19, 2021, 07:43:33 AM PST, Dave Wise <david_wise@phoenix.com> wrote:

?We were strongly encouraged to use parts already in the Common Design Parts Catalog.  ISTR a new part number cost something like $10,000 in administrative overhead.y


Dave Wise

Information Display, 1980-1994?

________________________________
From: TekScopes@groups.io <TekScopes@groups.io> on behalf of Tom Lee via groups.io <tomlee=ee.stanford.edu@groups.io>
Sent: Thursday, February 18, 2021 11:02 PM
To: TekScopes@groups.io
Subject: Re: [TekScopes] Help replacing a resistor/inductor in SG503

Hi John,

Your speculation might  be reasonable for LR142 vs LR150, although I
think that it is more likely that Tek used LR combos that were already
in inventory, rather than endure the paperwork in specifying a new part.
The resistance value isn't too critical for either LR142 or LR150. As
for LR190, though, its resistance actually matters: LR190 participates
in a series stack of parallel LR sections to produce a broadband
collector load resistance that is shunted at DC to keep the collector
swing about VCC. A single parallel LR wouldn't get the job done, owing
to the low self-resonant frequency of a 2.5mH choke. As the frequency
increases past the self-resonance of one section, another section cuts
in to keep the resistance high.

-- 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<http://www-smirc.stanford.edu>

On 2/18/2021 22:17, John Kolb wrote:

Looking at the schematic, there are a number of LR's and they are
wound on different value resistors. Noting that LR142 and LR150 are
both 0.25uH/51 ohms and LR135 and LR19? are both 2.5uH/2560 ohms, it
may be that Tek was only using the resistor value to identify the
component. Nevertheless you should duplicate the broken part as
exactly as possible. With the bad part intact except for the broken
lead, with even a cheap vernier caliper, you should be able to match
the part very well.

On 2/18/2021 8:30 AM, Jared Cabot via groups.io wrote:
Hi all,
I recently got myself an SG503 Leveled Sine Wave generator for a good
price and I am currently repairing it back to operational condition.

I found a few components in the oscillator section were damaged, from
what I guess, is from some ham-fisted poking from someone's previous
repair attempt.
Most parts were easy to replace, but I'm a bit stuck on one. LR140 is
a 20nH inductor consisting of a couple turns of fine wire around a
1/8w 47ohm carbon composite resistor.
It looks like it has been poked at and one leg was broken off.
Below is linked a page from the schematics with the part highlighted
in red.
https://groups.io/g/TekScopes/photo/260912/0/SG503 Oscillators LR140 Detail.jpg<https://groups.io/g/TekScopes/photo/260912/0/SG503 Oscillators LR140 Detail.jpg>


My question is, How's the best way to recreate this sort of part? I
tried winding a few turns on a 1/8 metal film resistor and didn't
have much luck, I have a DE-5000 LCR meter and a HP 4276A etc but
it's a little difficult to measure such small values reliably.
Should I get a carbon composite resistor from Digikey/Mouser to
recreate the part more true to the original design instead of the
metal film I used? Or is a plain 20nH axial inductor without the
resistor ok to use? Any other ideas or suggestions?

Thanks!