Carbon resistors - replace or shunt?

[6SN7WGTB]

Am restoring a Taylor 172A VTVM and have established that all of the resistors around the triode bridge are around 15-20% high. Consequently, calibration is out.

The original resistors are the hollow tube type with dots for value codes. They're all gold spot, so within 5% in theory at some point in their lives...

I was planning to replace the lot. All the panclimatic ladder/divider resistors remain as good as spot on.

It has occurred to me that I could simply shunt each to bring the overall value into kilter. I've done this with three already to prove that my expectations about bringing back to spec will indeed cure my issue.

I'd appreciate views on pros and cons. I can so far think of:

PROS:
Maintains a more original above-board picture
Far less likely to lift a track off of the paxolin board
Not an RF device so no potential issue with shunting

CONS:
Might look a bit shoddy under-board
Doesn't appeal to my OCD of having it just right
Why worry when I have replaced four of the 60-year old caps anyway...?

What I don't have the experience to say is have the old resistors finished their value-shifting? (The whole thing is well acclimatised to a warm and dry atmosphere, so I'm not at risk of correcting it at garden shed conditions then finding it's wandered off in the house).

Overall, I'm more interested in it working really well than looking really old.

However, seems like a legitimate question to pose!

Replace, those old resistors will just keep on moving.

[Sideband]

Yep agree, they'll just keep drifting. Just change them for modern types. It should work reliably rather than just look right!

[frankmcvey]

What the others said, with the possible exception of pink band (hi-stab) Erie resistors - I've checked many of these in the course of MHSVT restorations and I can't recollect having found a bad one.

I have a lot of NOS standard tolerance Erie resistors and I've tried the dodge of spreading some of them out on a baking tray in a warm oven for a few hours to try to bring them back into tolerance - didn't work for me.

[stevehertz]

I just don't get this, I'll call it hang up (at the risk of possibly offending people), of the need to replace resistors with period/style correct ones. In its lifetime almost any piece of electronic equipment that has been repaired will have been done so with a variety of parts that are:

better performance
worse performance
don't look the same
from a later era

That's 'historical' in itself, part of the set's history. To me, performance is the most important thing. For example I replaced the carbon resistors in my Quad 22 with metal film types. It looked really neat and the performance (lower noise) upgrade was worth it. FWIW I would however leave a top chassis mounted electrolytic in place and fit a new one underneath, no harm done, easy to perform, original looks from outside maintained.

Also, if a resistor has changed its value over time, there's no guarantee that it's not going to change further, so fitting a parallel resistor could be risky performance wise.

I would always perform the replacement of any component to be as neat as possible, and that is usually easily obtained by more modern components that are generally smaller. Their 'new', shininess bothers me none.

[6SN7WGTB]

As a stickler for things working to spec, I too would rather have the function over the form.

Obviously nice to do things sympathetically if possible.

Was just curious...!

At the same time, I will have a proper external 1M DC probe as moving it 'indoors' defeats one of the prime reasons for having it 'outdoors'.

Their 'new', shininess bothers me none.

SOme of the larger metal film types have a nice matt finish and look scale in olde equipment. And the colour code is easy to read!

[Leon Crampin]

I'll stick my neck out and offer an alternative view. Being rather fond of pre-war Murphy receivers, I have encountered many large carbon rod type resitors out of tolerance - nearly always high in value.

Unless the original resistor is unacceptably high, the value of shunt resistor is such that its dissipation is very small, so a miniature component can be fitted which sits neatly under or beside the original component. There is no need to unsolder the original, and the shunt can be attached with a non mechanical joint before soldering. This means that again, the disturbance to the original chassis in virtually invisible.

I realise that test equipment is more value sensistive than a radio in many instances, but a perusal of the circuit should reveal the most critical component positions - probably already fitted with carbon film resistors from new.

Together with re-stuffed capacitors, my Murphy sets look completely original and perform as they should. There are no melted or disfigured tag strips and I have not encountered further value shift in the shunted original components. Maybe if I re-check in 70 years' time...

Removing original components is a one way trip - but you can do it later if you really have to. I would shunt them first.

Leon.

[6SN7WGTB]

Originally Posted by merlinmaxwell

SOme of the larger metal film types have a nice matt finish and look scale in olde equipment. And the colour code is easy to read!

Will probably use Vishay CMF60 types, some of which I already have. They look suitably 'industrial'!

Er, but shiny...!

[stevehertz]

Originally Posted by Leon Crampin

Removing original components is a one way trip - but you can do it later if you really have to. I would shunt them first.

Leon.

It is, to a better performing, more stable, longer lasting solution.

[The Philpott]

Extra factor-you only really want to be dismantling an appliance when something needs fettling- small but ever-present risk of damage, stripped threads, lost screws, dust ingress. If something looks original from the outside, and tidy on the inside, then i'm happy.
Totally agree with Franks comment about Erie Hi-Stab resistors, i have some that are older than me and they are right on the bubble, so to speak.

I am sufficiently fussy to ensure that the colour codes and legends of newly fitted resistors &c are all facing the same way, but this is a common folly!

I have just disposed of the last of my carbon comps to someone who will hopefully be able to use them- 'cos i never will.

Dave

[barrymagrec]

If you shunt them you`ll just have to do it all over again in another sixty years time.....

[kalee20]

Replace! Using something which 'looks the part' as much as possible.

I have never had any problem with metal film, or carbon film either - but carbon composition are not stable at all. And that's a worry in test equipment even more than end-user stuff.

[GrimJosef]

I've measured a few carbon comp resistors over time. If I extrapolate from the first reading I took back to when they were new (I know how old they were to better than 10% accuracy) then I can work out their drift rate up to that point. When I measured them over the next decade or so I found that drift rate didn't change a lot. Their values just kept rising steadily.

This has reminded me that I should measure them again !

Cheers,

GJ

[6SN7WGTB]

Originally Posted by The Philpott

I am sufficiently fussy to ensure that the colour codes and legends of newly fitted resistors &c are all facing the same way, but this is a common folly!


Dave

Hope you don't do that with caps...

[The Philpott]

-Speaking of which i believe there was an 'ouch' incident in the relatively recent past of an electrolytic capacitor where all the indications were that the sleeve had been applied in reverse- leaving the minus symbol in the wrong place!
Dave

[Techman]

Even though I like originality, I would definitely advise replacement and not shunting. A drifted resistor will keep on drifting and it'll give trouble again in a lot less than 60 years!

If you want real conservation you could unsolder one end of the original resistor so that it's out of circuit, but still in place, then fit the new resistor - I've actually done this a couple of times with capacitors, but it can look awfully untidy if you're not careful, so I'd generally recommend a genuine 'historical' repair and replace just the parts that actually need replacing and replace them with current new types...you can always keep the old parts in a little bag with the unit.

[Electronpusher0]

I have different personal standards for vintage test equipment and vintage consumer equipment.
I try to keep the vintage consumer equipment as original as possible including re-stuffing caps.
For vintage test equipment it's function over form almost every time. That means modern components.

I said almost, I have restored an Atkins valve tester that is very rare, I can find evidence of only one other example sold on e-bay and that was a different model. Because of its rarity I did try to keep that as original as possible.
I must write up that one for the forum.....

Peter

[John_BS]

Upvote to barrymagrec in post 12!

[G6Tanuki]

I would always replace failed/drifty resistors with modern high-precision components, and then know that when you next need to use the thing-in-question there's a good chance it will still be working to spec.

[I hoenstly don't care what it looks like; if it doesn't work properly I don't want it cluttering up my space!]

Shunting already-failed resistors won't stop the old parts continuing to drift: old carbon-stick resistors are not like a fine wine - they do *not* improve with age ! And I feel free to remind you of "Tanuki's Vintage Test-Gear paradox", where you still need to have modern gear to fix the faults in the vintage your test-gear before you can use it to fix your vintage radio/transmitter/amplifier/scope/tape-recorder...