Ageing of Carbon Resistors. - UK Vintage Radio Repair and Restoration Discussion Forum

Station X · 16th Oct 2004, 10:02 am

I have just started restoring a 1960s Valve Amateur Transmitter which is full of 1/2 watt 20% tolerance Carbon Resistors. On checking with a meter I find that most of them are high in value and very close to the 20% mark. I generally allow 25% before changing the resistor. A few of the resistors are over 33% high in value.

The tolerance of these resistors doesn't compare very well with those I have checked in much older radios.

My question is will these resistors continue to increase in value, or can I assume they have stabilised after 40 years? If the former, then while I have the radio in bits, I will replace the resistors which are in critical positions in the circuit.

Graham.

Radio_Dave · 16th Oct 2004, 10:16 am

Hi Graham,

I've always assumed that they carry on deteriorating. What I find disturbing is how much some of the these resistors increase in value when heat is applied with a soldering iron

. A resistor which is at the 20% tolurance boarder line can increase dramatically if it's in a position where it can get hot

David

jim_beacon · 16th Oct 2004, 10:21 am

Graham,

heat and ratinghas a lot to do with this, if the components are conservativlry rated and well cooled, they tend to remain stable, as for example, in an old Tek scope (you only change them when you've done something stupid.......), or an early valve radio, where there is a lot of space. In later equipment, where the closest rating was used to save space and cost, and the components are much more closely spaced, you tend to get more problems.

Jim.

Norm_Leal · 16th Oct 2004, 5:05 pm

Hi Graham

In most circuits you won't find any improvement changing out of tolerance resistors. Some resistors open over time. These will need changing.

Even NOS resistors, made 30-40 years ago, may be high in value. Resistance increasing seems to slow down in time?

Norm

ukcol · 16th Oct 2004, 7:19 pm

Hi Graham

When I was a student back in the 1960s, I was told that one of the mechanisms by which carbon composition resistors go high in value was by molecule migration. Apparently in circuits where there are high voltages the molecules responsible for conduction migrate to the ends of the resistor leaving a high resistance part in the middle. This is a continuous process while the circuit is working until the component is more or less open circuit.

Regards

Sideband · 17th Oct 2004, 8:06 am

Hi Graham.

Everything said so far is true. It also depends on the make of resistors used. The old 1/2 watt types were quite stable but most resistors tend to drift high after a time. There are others that will go low but these are generaly used in higher current parts of the circuit like HT feed or cathode bias and usualy if they are (or have been) overrun.

I think you have to make an educated guess as to which are the best resistors to change. From experience, I find that higher value type tend to drift higher, for example anode load resistors on AF amps (like 220k). They may drop perhaps 60v across them at low current and probably results in molecule migration at a higher rate than say a low value dropping only a few volts.

Rich.

XTC · 17th Oct 2004, 3:49 pm

Hi Graham,

I'd check that the wattage of the resistors was right in the first place. If they've been run at or beyond their limits, they'd be expected to drift further than in a domestic set where they are generally conservatively rated.

Also, resistors have a voltage spec and this is often ignored. If a carbon resistor is subjected to more than its specified voltage , but within its wattage, it will show permanent changes in value in a very short time. Maybe the voltages they've been subjected to are higher than in a domestic set.

Sometimes I've found that old carbon composition resistors have values within spec when they are cold, but show really strange behaviour as they get warm.

Whether to replace and what with (NOS or modern) depends on the restoration objectives you have set yourself for this equipment. I'd be inclined to replace those resistors which were hard to get at later with modern 1 or 2 watt modern types, and leave those which were within, or just outside their tolerancees if they were easy to get at .

Generally, valve circuitry is pretty tolerant of resistor values.

Pete.

Station X · 18th Oct 2004, 8:46 am

Thanks for all your replies. You all seem to agree that Carbon Resistors do increase their value with age and that this is due to overheating caused by over running and/or lack of ventilation.

Interestingly the three resistors which are furtherest out of spec are all screen grid droppers. I shall calculate whether they are being over run. I've yet to check the screen grid decoupling capacitors, but they are probably OK, as they are disc ceramics. There isn't a single paper capacitor in this radio.

This radio was built towards the end of the valve era and I wonder whether the resistors used were actually intended for use in low voltage transistor circuits. The radio uses 300 Volts HT.

I shall adopt a sensible approach as to which resistors to change for 2W 500V types in the interests of long term reliability. I won't change any for the sake of it.

Graham.

Biggles · 18th Oct 2004, 1:11 pm

Hi all, it may be worth noting that a lot of older circuits appear not to be too critical of component values. I think that the designers were well aware of the limits of component manufacture at the time, and produced tolerant circuits. I think nowadays that we are used to picking a component up and testing it, and being surprised if it is more than about 1% out, because of modern production techniques. I have wasted a lot of time in the past " being the perfectionist " and replacing components that were just out of tolerance, only to find that it has made no difference at all to the performance of the set. Of course, if a component is proven to be shifting in value, or under-rated by design, then change it , but don't be too critical.
Biggles.

Radio_Dave · 18th Oct 2004, 3:08 pm

It might be worth me mentioning that I won a large quantity of the old nos Eire (I think thats how to spell them :

) resistors.

Although these have never been used alot of them are way out of tolurance. Maybe they were never too good from new!

David

Norm_Leal · 18th Oct 2004, 4:25 pm

Hi David

This is true even with carbon composition resistors made 20 years ago. We had some 1 meg resistors that were on the shelf for a couple years. Many had incresed in value and were out of tolerance. This seems to be more of a problem with higher values which sometimes open.

They were good when received. Every incomming shipment was inspected and sample tested.

Norm

XTC · 18th Oct 2004, 4:50 pm

Here's an interesting website talking about the wonders of carbon composition resistors.

http://www.tone-lizard.com/resistors.htm

I'd guess that the stability of NOS composition resistors depends quite a lot on the conditions they've been stored in, and that the ingress of moisture has a big part to play. 40 years in the average UK garage would be very different to 40 years in a garage in Arizona, say.

In the article, there's a picture of a bag of resistors with a silica dessicant. The author reckons that this type of packing makes a difference.

Pete.

16th Oct 2004, 10:02 am	#1
Station X Moderator Join Date: Jan 2003 Location: Ipswich, Suffolk, IP4, UK. Posts: 21,286	Ageing of Carbon Resistors. I have just started restoring a 1960s Valve Amateur Transmitter which is full of 1/2 watt 20% tolerance Carbon Resistors. On checking with a meter I find that most of them are high in value and very close to the 20% mark. I generally allow 25% before changing the resistor. A few of the resistors are over 33% high in value. The tolerance of these resistors doesn't compare very well with those I have checked in much older radios. My question is will these resistors continue to increase in value, or can I assume they have stabilised after 40 years? If the former, then while I have the radio in bits, I will replace the resistors which are in critical positions in the circuit. Graham. __________________ Graham. Forum Moderator Reach for your meter before you reach for your soldering iron.

16th Oct 2004, 10:16 am	#2
Radio_Dave Nonode Join Date: Jun 2004 Location: Norwich, Norfolk, UK. Posts: 2,543	Re: Ageing of Carbon Resistors. Hi Graham, I've always assumed that they carry on deteriorating. What I find disturbing is how much some of the these resistors increase in value when heat is applied with a soldering iron . A resistor which is at the 20% tolurance boarder line can increase dramatically if it's in a position where it can get hot David __________________ http://www.youtube.com/ My Nixie Clocks

16th Oct 2004, 10:21 am	#3
jim_beacon Retired Dormant Member Join Date: Mar 2003 Location: Bishop's Waltham, Hants, UK. Posts: 939	Re: Ageing of Carbon Resistors. Graham, heat and ratinghas a lot to do with this, if the components are conservativlry rated and well cooled, they tend to remain stable, as for example, in an old Tek scope (you only change them when you've done something stupid.......), or an early valve radio, where there is a lot of space. In later equipment, where the closest rating was used to save space and cost, and the components are much more closely spaced, you tend to get more problems. Jim.

16th Oct 2004, 5:05 pm	#4
Norm_Leal Retired Dormant Member Join Date: Jan 2003 Location: Livermore, California, USA. Posts: 139	Re: Ageing of Carbon Resistors. Hi Graham In most circuits you won't find any improvement changing out of tolerance resistors. Some resistors open over time. These will need changing. Even NOS resistors, made 30-40 years ago, may be high in value. Resistance increasing seems to slow down in time? Norm

16th Oct 2004, 7:19 pm	#5
ukcol Rest in Peace Join Date: May 2003 Location: Harlaxton, Lincolnshire, UK. Posts: 3,944	Re: Ageing of Carbon Resistors. Hi Graham When I was a student back in the 1960s, I was told that one of the mechanisms by which carbon composition resistors go high in value was by molecule migration. Apparently in circuits where there are high voltages the molecules responsible for conduction migrate to the ends of the resistor leaving a high resistance part in the middle. This is a continuous process while the circuit is working until the component is more or less open circuit. Regards

17th Oct 2004, 8:06 am	#6
Sideband Dekatron Join Date: Oct 2003 Location: Croydon, Surrey, UK. Posts: 7,566	Re: Ageing of Carbon Resistors. Hi Graham. Everything said so far is true. It also depends on the make of resistors used. The old 1/2 watt types were quite stable but most resistors tend to drift high after a time. There are others that will go low but these are generaly used in higher current parts of the circuit like HT feed or cathode bias and usualy if they are (or have been) overrun. I think you have to make an educated guess as to which are the best resistors to change. From experience, I find that higher value type tend to drift higher, for example anode load resistors on AF amps (like 220k). They may drop perhaps 60v across them at low current and probably results in molecule migration at a higher rate than say a low value dropping only a few volts. Rich. __________________ There are lots of brilliant keyboard players and then there is Rick Wakeman.....

17th Oct 2004, 3:49 pm	#7
XTC Retired Dormant Member Join Date: Oct 2004 Location: Bristol, UK. Posts: 837	Re: Ageing of Carbon Resistors. Hi Graham, I'd check that the wattage of the resistors was right in the first place. If they've been run at or beyond their limits, they'd be expected to drift further than in a domestic set where they are generally conservatively rated. Also, resistors have a voltage spec and this is often ignored. If a carbon resistor is subjected to more than its specified voltage , but within its wattage, it will show permanent changes in value in a very short time. Maybe the voltages they've been subjected to are higher than in a domestic set. Sometimes I've found that old carbon composition resistors have values within spec when they are cold, but show really strange behaviour as they get warm. Whether to replace and what with (NOS or modern) depends on the restoration objectives you have set yourself for this equipment. I'd be inclined to replace those resistors which were hard to get at later with modern 1 or 2 watt modern types, and leave those which were within, or just outside their tolerancees if they were easy to get at . Generally, valve circuitry is pretty tolerant of resistor values. Pete.

18th Oct 2004, 8:46 am	#8
Station X Moderator Join Date: Jan 2003 Location: Ipswich, Suffolk, IP4, UK. Posts: 21,286	Re: Ageing of Carbon Resistors. Thanks for all your replies. You all seem to agree that Carbon Resistors do increase their value with age and that this is due to overheating caused by over running and/or lack of ventilation. Interestingly the three resistors which are furtherest out of spec are all screen grid droppers. I shall calculate whether they are being over run. I've yet to check the screen grid decoupling capacitors, but they are probably OK, as they are disc ceramics. There isn't a single paper capacitor in this radio. This radio was built towards the end of the valve era and I wonder whether the resistors used were actually intended for use in low voltage transistor circuits. The radio uses 300 Volts HT. I shall adopt a sensible approach as to which resistors to change for 2W 500V types in the interests of long term reliability. I won't change any for the sake of it. Graham. __________________ Graham. Forum Moderator Reach for your meter before you reach for your soldering iron.

18th Oct 2004, 1:11 pm	#9
Biggles Rest in Peace Join Date: Feb 2004 Location: Hexham, Northumberland, UK. Posts: 2,234	Re: Ageing of Carbon Resistors. Hi all, it may be worth noting that a lot of older circuits appear not to be too critical of component values. I think that the designers were well aware of the limits of component manufacture at the time, and produced tolerant circuits. I think nowadays that we are used to picking a component up and testing it, and being surprised if it is more than about 1% out, because of modern production techniques. I have wasted a lot of time in the past " being the perfectionist " and replacing components that were just out of tolerance, only to find that it has made no difference at all to the performance of the set. Of course, if a component is proven to be shifting in value, or under-rated by design, then change it , but don't be too critical. Biggles.

18th Oct 2004, 3:08 pm	#10
Radio_Dave Nonode Join Date: Jun 2004 Location: Norwich, Norfolk, UK. Posts: 2,543	Re: Ageing of Carbon Resistors. It might be worth me mentioning that I won a large quantity of the old nos Eire (I think thats how to spell them : ) resistors. Although these have never been used alot of them are way out of tolurance. Maybe they were never too good from new! David __________________ http://www.youtube.com/ My Nixie Clocks

18th Oct 2004, 4:25 pm	#11
Norm_Leal Retired Dormant Member Join Date: Jan 2003 Location: Livermore, California, USA. Posts: 139	Re: Ageing of Carbon Resistors. Hi David This is true even with carbon composition resistors made 20 years ago. We had some 1 meg resistors that were on the shelf for a couple years. Many had incresed in value and were out of tolerance. This seems to be more of a problem with higher values which sometimes open. They were good when received. Every incomming shipment was inspected and sample tested. Norm

18th Oct 2004, 4:50 pm	#12
XTC Retired Dormant Member Join Date: Oct 2004 Location: Bristol, UK. Posts: 837	Re: Ageing of Carbon Resistors. Here's an interesting website talking about the wonders of carbon composition resistors. http://www.tone-lizard.com/resistors.htm I'd guess that the stability of NOS composition resistors depends quite a lot on the conditions they've been stored in, and that the ingress of moisture has a big part to play. 40 years in the average UK garage would be very different to 40 years in a garage in Arizona, say. In the article, there's a picture of a bag of resistors with a silica dessicant. The author reckons that this type of packing makes a difference. Pete.