Yet another 'replacing carbon resistors with metal film' question....


Jared Cabot
 

Hi all,

I have a PG502 TM500 module here I'm poking away at, and although the capacitors all test better than the datasheets in all respects (Those Vishay 30D caps are pretty good!), some of the old brown square ended carbon composite resistors have predictably drifted.
I also found also that the 15V trimpot is high resistance unless you press on the dial leading to no voltage rails, so that will be replaced.

I want to replace all of the fixed carbon composite with new resistors, but I'm not sure which way to go...
I have a set of modern 5% carbon film resistors, and also a set of 50ppm 1% metal film resistors too. I'm not too worried about using film resistors on the power supply etc, but the rest of the circuit is what I'm not sure about. I assume in this situation, the metal film is better than the carbon film seeing as they are both spiral cut?

What's the rule of thumb here? Should I just shotgun the metal film resistors and hope for the best, or use carbon composition resistors instead?

Here's the data sheets of the resistors I have:

Carbon film: http://takman-e.co.jp/pdf/rd_rds.pdf
Metal film: http://takman-e.co.jp/pdf/rlc.pdf


Jean-Paul
 

Mr Cabot: We have PG501, 502, 506, running perfectly with all original parts, needing some switch and pot cleaning on occasion.

Carbon comp are lower self capacitance, and are still used in critical high freq and high voltage applications.

If it needs service, I suggest to replace ONLY failed/out of spec parts causing the issues, and leave all other parts alone.

Most of the changes in resistor type will make no difference.

"if it ain't broke dont fix it"

Regardless I suggest recalibration afterwards.

Bon chance,

Jon


Jared Cabot
 

Yeah, I'll be recalibrating for sure, as I intend to use it to calibrate other gear.

I'll test things thoroughly and might do as you say, just replace what is needed. I already tested the capacitors as perfect so they are back in place to continue being used.
I was just thinking to replace all the carbon composite resistors now as they are a known failure point and failures have already started. I just wanted to avoid having to go back in later and end up having to recalibrate etc multiple times.

It looks like Digikey stock carbon composition resistors too, so maybe I should swap like for like for those that are out of spec?


 

Personally I would *only* replace CC resistors that were causing out of specification behaviour. If e.g. all resistors in a potential divider have drifted high by the same %age (as is probable) then the divider will still have the correct voltages.

Or put another way if you can calibrate it successfully, then don’t replace anything at all!

David

-----Original Message-----
From: TekScopes@groups.io <TekScopes@groups.io> On Behalf Of Jared Cabot via groups.io
Sent: 28 December 2020 09:47
To: TekScopes@groups.io
Subject: Re: [TekScopes] Yet another 'replacing carbon resistors with metal film' question....

Yeah, I'll be recalibrating for sure, as I intend to use it to calibrate other gear.

I'll test things thoroughly and might do as you say, just replace what is needed. I already tested the capacitors as perfect so they are back in place to continue being used.
I was just thinking to replace all the carbon composite resistors now as they are a known failure point and failures have already started. I just wanted to avoid having to go back in later and end up having to recalibrate etc multiple times.

It looks like Digikey stock carbon composition resistors too, so maybe I should swap like for like for those that are out of spec?


Jean-Paul
 

Jared, is this the PG502 pulse gen or PG506 calibration gen?

if a CAL gen with high speed edges, I suggest to NOT change any resistance or cap in the fast pulse output section

In our work, since 1970s, carbon comp and carbon film resistance fail with

a/ over dissipation
b/ high-voltage
c/ mechanical stress eg on leads

They dont fail during normal use, very reliable.

Just the ramblings of an old retired EE

Bon Chance

Jon


Jean-Paul
 

For sure DK and Mouser carbon composition resistance are from China and poor quality

We use only Allen Bradley made in USA and have a huge old stock inventory 1/4---1/2-1-2W

Essential for high-voltage.

The original Allen's Bradley were probably used in the Tektronix, so your reliability may be WORSE with the current Chinese garbage resistors.

Jon


Jared Cabot
 

I have the PG502 250MHz Pulse Generator in front of me now (I also have a PG506 to look over but that is another project for another time).

I guess I'll see how the unit performs with the minimum of modifications to get it functional and make a decision once I see what it says as to further repairs.
I've found a lot of these types of resistors have drifted high in a number of pieces of test equipment I've worked on in less critical applications, hence my thoughts on replacing them in this unit, but I guess if it works, then no point fixing it until it breaks.... :)


Here are the schematics with the Allen Bradley style carbon composite resistors in my unit highlighted:
https://i.imgur.com/8WiK5b9.jpg
https://i.imgur.com/g3KXKlN.jpg
https://i.imgur.com/geXFOIK.jpg


Jean-Paul
 

AHA! Yes PG502 is a fine unit, the back term 50 Ohm is handy. I have several, neither has ever has any problem.

For scope cal, the really fast rise pulses are from PG506 and Leo Bodnar's 40 pS pulser

As a general use PG with fast capability PG502 is best.

The resistors you highlight are not critical to the performance and even a 5% value should not affect it.
I dont recommend Removing, testing and replacing the parts if its performing satisfactorily.


Enjoy,

Jon


Jared Cabot
 

Well, after replacing a couple faulty trimpots (it looks like they were already replaced with some cheap rubbish in the past) it seems to work as well as I can measure on my TDS210 and HP 34461A.
Tonight I'll replace the temporary trimpots I used for testing with some nice Bourns trimpots I picked up today in Akihabara, and after a gentle clean of the switches and stuff I'll call it done.

Next I'll need to get some sampling plugins for my 7704 to really get this thing (and my other TM500 modules) dialed in properly...


Jean-Paul
 

Jared good work! We found the old Bourns trimpots (cermet) are the best, and avail in 3, 5 turns besides one turn.

TEK sampling plugins are interesting but difficult to get working.

Used units can have damaged inputs or hard to find TDs in the sampler.

We have a 7104 mainframe with 7A29 plugins to get 1 GHz response and excellent pulse performance.

Suggest to get some veteran advise on the best sampling PUs for 7704 before plunging in!

Best,

Jon


fiftythreebuick
 

Probably not a factor in this case, but in some cases we have found that metal film resistors (of the same value/wattage as the original carbon comp resistors) would blow open when exposed to high inrush current transients while carbon comp resistors would live in the same environment just fine. We first discovered this when working on the power supply of a 555. Multiple metal film resistors blew while a used carbon comp from a junker instrument stayed in there just fine.

Just for what it's worth....

Tom


Richard Knoppow
 

This resistance to pulse current seems to be about the only virtue carbon comp resistors have. I don't know why. Most carbon film resistors will take tremendous overloads but as steady state. I worked in resistor manufacture many years ago. We made mil spec carbon film and established reliability metal film parts. Carbon film resistors would run at red heat without damage.

On 12/29/2020 9:35 AM, fiftythreebuick wrote:
Probably not a factor in this case, but in some cases we have found that metal film resistors (of the same value/wattage as the original carbon comp resistors) would blow open when exposed to high inrush current transients while carbon comp resistors would live in the same environment just fine. We first discovered this when working on the power supply of a 555. Multiple metal film resistors blew while a used carbon comp from a junker instrument stayed in there just fine.

Just for what it's worth....

Tom



--
Richard Knoppow
dickburk@ix.netcom.com
WB6KBL


Glenn Little
 

The carbon comp can absorb much more transient energy than a film resistor.
The carbon comp is a mass with leads on it, the film is a film deposited on an insulator, much less mass to absorb transients.

Each has its place.
Carbon comps have been replaced with ceramic comps.

Glenn

On 12/29/2020 12:35 PM, fiftythreebuick wrote:
Probably not a factor in this case, but in some cases we have found that metal film resistors (of the same value/wattage as the original carbon comp resistors) would blow open when exposed to high inrush current transients while carbon comp resistors would live in the same environment just fine. We first discovered this when working on the power supply of a 555. Multiple metal film resistors blew while a used carbon comp from a junker instrument stayed in there just fine.

Just for what it's worth....

Tom



--
-----------------------------------------------------------------------
Glenn Little ARRL Technical Specialist QCWA LM 28417
Amateur Callsign: WB4UIV wb4uiv@arrl.net AMSAT LM 2178
QTH: Goose Creek, SC USA (EM92xx) USSVI LM NRA LM SBE ARRL TAPR
"It is not the class of license that the Amateur holds but the class
of the Amateur that holds the license"