I'm going over a couple of receivers that I'll be giving to friends - a Harmon Kardon 330B and a Sansui QRX4500.
One thing that I've nticed is that the HK receiver is packed full of 10% tolerance resistors.
This novice has been led to believe that 1% resistors should be used whenever possible. Believe me - I'm working on my understanding of things.
But for now - would it worthwhile to replace them with 1% metal film resistors? I probably have all values on hand.
Thanks, folks.
One thing that I've nticed is that the HK receiver is packed full of 10% tolerance resistors.
This novice has been led to believe that 1% resistors should be used whenever possible. Believe me - I'm working on my understanding of things.
But for now - would it worthwhile to replace them with 1% metal film resistors? I probably have all values on hand.
Thanks, folks.
LowRedMoon,
Nothing wrong with 1%'s at ALL. 1%'s are the only thing I put in my bins. Not because because they're "better"... it just minimzes what I have to keep around... you can always use a 1 in place of a 5 or 10... doesn't work the other way though.
10% resistors have all but gone away... 5%'s are on their way as well. Will it make a difference in a circuit designed for 10%'s? ...not really... might help once in awhile... but it will never hurt. And they don't cost that much more these days.
Nothing wrong with 1%'s at ALL. 1%'s are the only thing I put in my bins. Not because because they're "better"... it just minimzes what I have to keep around... you can always use a 1 in place of a 5 or 10... doesn't work the other way though.
10% resistors have all but gone away... 5%'s are on their way as well. Will it make a difference in a circuit designed for 10%'s? ...not really... might help once in awhile... but it will never hurt. And they don't cost that much more these days.
CaliforniaBob said:
There are two issues here. The tolerance of the resistors should not be critical. If HK uses 10% tolerance that means that they have competently designed the circuits to be stable and predictable even if resistors slightly vary in value. Going to 5%, 2% or 1% wouldn't make a difference.
The other is the type of resistors. The 10% types may be carbon film types. Replacing them with metal film types CAN lower the noise floor, but it depends on a lot of other factors too.
Jan Didden
This page gives an indication that the statement could be correct:
Carbon film resistor: cheap general purpose resistor, works quite well also on high frequencies...
Thin film resistor: good long time stability, good temperature statiblity, good voltage dependity rating, low noise, not good for RF...
Metal film is somewhere in the middle I guess.
P.S. Much of what I learned in the past was offered by people who were in business for decades. The fact that I can reproduce many of their claims in daily life convinces me they mostly got it right. But not always up to the smallest details...
/Hugo
Carbon film resistor: cheap general purpose resistor, works quite well also on high frequencies...
Thin film resistor: good long time stability, good temperature statiblity, good voltage dependity rating, low noise, not good for RF...
Metal film is somewhere in the middle I guess.
P.S. Much of what I learned in the past was offered by people who were in business for decades. The fact that I can reproduce many of their claims in daily life convinces me they mostly got it right. But not always up to the smallest details...
/Hugo
peranders said:
Per-Anders, replacing ( for example ) electrolytic caps in old gear is a must-do if you want the device to keep working for several more years. So leaving gear just like they are can possibly result in failure. I would not think of replacing carbon resistors nor caps in the tuning part. The amplifier part however might benefit from replacing them for metal film resistors.
Just leave the RF section as it is. Especially the power supply caps need replacement when they're older than 10 years.Preventive maintenance is old fashioned but it has its merits.
Carbon composition resistors were non-inductive, carbon film was very inductive because the film was spiral cut to get a long path. Ferrous leads don't help either. Modern resistors are generally much better and smd is best.
RF used to mean VHF to most people, these days 2.4GHz and beyond products are all over the place, so RF performance of consumer components has had to keep up.
One thing to watch on new resistors is the voltage rating, smds especially are getting to the point that they cannot take typical amplifier rail voltages.
RF used to mean VHF to most people, these days 2.4GHz and beyond products are all over the place, so RF performance of consumer components has had to keep up.
One thing to watch on new resistors is the voltage rating, smds especially are getting to the point that they cannot take typical amplifier rail voltages.
There seem to be several issues here.
Would replacing the 10% tolerance resistors with 1% resistors improve the device? The short answer is not likely. Resistor manufacturig processes are precise, but not necessarily accurate. The 10% tolerance parts are within +/-10% of the marked value, but that says nothing about the matching of values which is typically much closer than the +/-10%. Most modern audio circuits (certainly those manufactured in massive quantities) use differential architecture whose performance depends more on the matching of the parts than the absolute accuracy of the values.
You are risking damage to the receiver by just going through and replacing the resistors. You may damage the PCB, and/or you may insert an incorrect value in one or more locations, resulting in a dead receiver. Hence the adage "If it ain't broke, don't fix it."
In very old equipment such as vacuum tube radios made in the 1920s, 30s, and to a lesser extent the 1940s, the bulk carbon resistors used in them typically drift upward in value over time and the radio will often require replacement of the resistors in order to function properly. The resistors set the bias points of the vacuum tubes and if those go off too far, the circuit ceases to function properly. I have restored many such radios to original operating condition by simply replacing failed capacitors and drifted resistors.
Are metal film resistors lower noise than other type of resistors? A little, but not much. Noise from resistors is dominated by the resistance value. The material is a very small contributor to the overall noise.
Why do many hobbyists use 1% tolerance resistors? Because they are not much more expensive than 5% and 10% parts, and they provide some peace of mind that the value of the part is what is marked. A mass market manufacturer prefers not to use 1% parts because if they cost a penny more than a 5% part they may lose tens of thousands of dollars.
Why do many audio hobbyists use metal film resistors? Because they come in 1% tolerance values for little extra cost, and because they mistakenly think that the circuit noise is largely a function of the material, when it is actually mainly a function of the resistance value, regardless of the material.
Metal film resistors can be a problem at RF frequencies because of their physical construction. They are typically made by applying a film to a ceramic tube, then removing a thin line of the film from the tube as it is rotated. The result is a resistive track that looks very much like a coil, which has a relatively large inductance.
In some critical applications there are other considerations such as thermal coefficients that make metal film resistors desirable over other types.
I_F
Would replacing the 10% tolerance resistors with 1% resistors improve the device? The short answer is not likely. Resistor manufacturig processes are precise, but not necessarily accurate. The 10% tolerance parts are within +/-10% of the marked value, but that says nothing about the matching of values which is typically much closer than the +/-10%. Most modern audio circuits (certainly those manufactured in massive quantities) use differential architecture whose performance depends more on the matching of the parts than the absolute accuracy of the values.
You are risking damage to the receiver by just going through and replacing the resistors. You may damage the PCB, and/or you may insert an incorrect value in one or more locations, resulting in a dead receiver. Hence the adage "If it ain't broke, don't fix it."
In very old equipment such as vacuum tube radios made in the 1920s, 30s, and to a lesser extent the 1940s, the bulk carbon resistors used in them typically drift upward in value over time and the radio will often require replacement of the resistors in order to function properly. The resistors set the bias points of the vacuum tubes and if those go off too far, the circuit ceases to function properly. I have restored many such radios to original operating condition by simply replacing failed capacitors and drifted resistors.
Are metal film resistors lower noise than other type of resistors? A little, but not much. Noise from resistors is dominated by the resistance value. The material is a very small contributor to the overall noise.
Why do many hobbyists use 1% tolerance resistors? Because they are not much more expensive than 5% and 10% parts, and they provide some peace of mind that the value of the part is what is marked. A mass market manufacturer prefers not to use 1% parts because if they cost a penny more than a 5% part they may lose tens of thousands of dollars.
Why do many audio hobbyists use metal film resistors? Because they come in 1% tolerance values for little extra cost, and because they mistakenly think that the circuit noise is largely a function of the material, when it is actually mainly a function of the resistance value, regardless of the material.
Metal film resistors can be a problem at RF frequencies because of their physical construction. They are typically made by applying a film to a ceramic tube, then removing a thin line of the film from the tube as it is rotated. The result is a resistive track that looks very much like a coil, which has a relatively large inductance.
In some critical applications there are other considerations such as thermal coefficients that make metal film resistors desirable over other types.
I_F
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.