Resistor codes? Which way round?

chillidoggy

Update:- 

I've worked out what DMM means, and resistors going low. MF is metal film, PSU Is a power supply unit, but as for a feedback path, you've got me there. The remainder will remain a mystery, but if I get to building an amp, the definitions may become clearer as time goes on!

Cirrus

Bias resistors and grid stopper resistors are there to help determine how the power tubes run. When a power tube overheats/overloads/shorts spectacularly (valve flashover is one way to describe that) the resistors attached to it can end up overheating too as too much power goes through them. Carbon comps are a type of resistor used in older amps. They can drift in value over time or when they heat up, become noisy etc, (up. Down, sideways...)more often than more modern resistor designs. But they do have a tonal signature some enjoy.

jpfamps

Cirrus said:

Bias resistors and grid stopper resistors are there to help determine how the power tubes run. When a power tube overheats/overloads/shorts spectacularly (valve flashover is one way to describe that) the resistors attached to it can end up overheating too as too much power goes through them. Carbon comps are a type of resistor used in older amps. They can drift in value over time or when they heat up, become noisy etc, (up. Down, sideways...)more often than more modern resistor designs. But they do have a tonal signature some enjoy.

Carbon comps do have one advantage over other resistor types in that they are better able to survive large overloads.

ICBM

jpfamps said:

Carbon comps do have one advantage over other resistor types in that they are better able to survive large overloads.

Not as much as wirewounds though… which is why I prefer those for screen resistors.

Cirrus

jpfamps said:

Cirrus said:

Bias resistors and grid stopper resistors are there to help determine how the power tubes run. When a power tube overheats/overloads/shorts spectacularly (valve flashover is one way to describe that) the resistors attached to it can end up overheating too as too much power goes through them. Carbon comps are a type of resistor used in older amps. They can drift in value over time or when they heat up, become noisy etc, (up. Down, sideways...)more often than more modern resistor designs. But they do have a tonal signature some enjoy.

Carbon comps do have one advantage over other resistor types in that they are better able to survive large overloads.

Is that large short term or large overloads as a matter of regular operation? Reason I ask is that in the vox tone circuit there's a 56k resistor that has about 180v across it. I've thought this'd be a good place to try a carbon comp but never risked it because the highest rated 56k carbon comp I can find is 0.5w.

Sorry for the hijack!

ESchap

Cirrus said:

Is that large short term or large overloads as a matter of regular operation? Reason I ask is that in the vox tone circuit there's a 56k resistor that has about 180v across it. I've thought this'd be a good place to try a carbon comp but never risked it because the highest rated 56k carbon comp I can find is 0.5w.

Sorry for the hijack!

HiFi collective have 1W Allen and Bradleys in 5% and 10% http://www.hificollective.co.uk/catalog/gb5631-allen-bradley-rc32gf563k-rcr32g563k-p-5717.html

jpfamps

ICBM said:

jpfamps said:

Carbon comps do have one advantage over other resistor types in that they are better able to survive large overloads.

Not as much as wirewounds though… which is why I prefer those for screen resistors.

Wirewounds are good at surviving over dissipation, but CC resistors can cope with voltage overload better than wirewounds.

Wirewounds often have quite low voltage (for a valve amp), with 350V being typical, so 5W Cement box resistors often fail open circuit when used as screen grid resistors.

jpfamps

Cirrus said:

jpfamps said:

Cirrus said:

Bias resistors and grid stopper resistors are there to help determine how the power tubes run. When a power tube overheats/overloads/shorts spectacularly (valve flashover is one way to describe that) the resistors attached to it can end up overheating too as too much power goes through them. Carbon comps are a type of resistor used in older amps. They can drift in value over time or when they heat up, become noisy etc, (up. Down, sideways...)more often than more modern resistor designs. But they do have a tonal signature some enjoy.

Carbon comps do have one advantage over other resistor types in that they are better able to survive large overloads.

Is that large short term or large overloads as a matter of regular operation? Reason I ask is that in the vox tone circuit there's a 56k resistor that has about 180v across it. I've thought this'd be a good place to try a carbon comp but never risked it because the highest rated 56k carbon comp I can find is 0.5w.

Sorry for the hijack!

You can always employ two resistors in parallel to increase the power dissipation, so two 110 k, 1/2 Ws in parallel will be equivalent of a 1W 55k part.

In fact I'm using this very technique this afternoon by replacing a 10W 110 ohm resistor with 2 x 220 7W resistors. I didn't have a 10W 110 ohm resistor anyway, but using 2 x 7W will up the power rating to 14W to reduce the chance of future failure.

ICBM

jpfamps said:

Wirewounds are good at surviving over dissipation, but CC resistors can cope with voltage overload better than wirewounds.

Wirewounds often have quite low voltage (for a valve amp), with 350V being typical, so 5W Cement box resistors often fail open circuit when used as screen grid resistors.

I don't really like those - the ones I prefer are the Welwyn W20 series vitrified ceramic ones, which are rated for 500V if I remember right - they seem close to indestructible unless some numpty has foiled the fuse.

chillidoggy

So what's the deal with resistors being rated for power? I can understand the colour code idea, but how would i know if I needed to buy a new one, or replace one, what to ask for?

Also, I see in my amps that some resistors are bigger than others , like in the photo I posted above. What's that about?

ICBM

The size is related to the power rating - bigger = more surface area for heat dissipation. When you get experienced at this you can judge the rating just by looking at the size, give or take a bit. Modern types can have slightly higher ratings than the same size old ones, due to higher temperature-rated materials.

In your pic the blue one looks like 1/2W and the grey ones 3W, possibly 5W - although I tend to be a bit suspicious of some of the claimed modern ratings, particularly in hot environments. (You have to de-rate the power handling at higher operating temperatures.)

You can always use a higher rating if you can physically fit it in, but never a lower one unless you know for certain that the manufacturer has used an unnecessarily big one - very rare in commercial mass-produced amps, because bigger ones cost more! And as jpfamps mentioned earlier, resistors also have a voltage rating, which is very often overlooked - and is rarely important outside valve circuits - and which is often related to the power rating, so occasionally you might find one which is "too big" for that reason rather than power rating itself.

As a general rule, if you need to replace a resistor because it's burned out, use a bigger one! It tends to indicate that the dead one wasn't up to the job. At least unless you know that the initial fault would have caused an unsurvivable failure for just about any resistor. (eg Fender "virtual centre tap" resistors burning out if a power valve shorts badly.)

Some people think it's good to use resistors as "fuses" so that a low-rated resistor will burn out before other damage occurs - while this can sometimes be true, it's bad practice really - resistors are not designed to fail safely usually, and can sometimes even catch fire. If you need a fuse, use a fuse… that's what they're for.

As a general rule you should keep a safety margin as well - if you calculate the power dissipation of a particular resistor, it's best to double it (at least, if you can) to get the correct rating.

ecc83

chillidoggy said:

So what's the deal with resistors being rated for power? I can understand the colour code idea, but how would i know if I needed to buy a new one, or replace one, what to ask for?

Also, I see in my amps that some resistors are bigger than others , like in the photo I posted above. What's that about?

Any resistor has to be of a size and type that is capable of dissipating the power caused by the current through it. This power (in watts) is the product of the volt drop across the resistor and the current thru' it.( It is often  more convenient by simple algebra to arrive at P =V.V /R) . To take a simple example.

ECC83 with a 100k anode load and a 300V supply. Say Va =150V then P=150x150/100,000=0.225W. That should mean we can get away with a very dinky (and cheap!) 1/4W component? Not really. What would happen if the the valve shorted anode to cathode (not likely I know but bear with!) ?

Now P= 300x300/R = 0.9W! So, should we fit a 1 W component for safety? No, a 0.6W R will suffice since it is almost unheard of for a pre amp triode to short in that way.However! Peeps DO fekk about with the wrong valves and fitting say an ECC82 in there might well burn out a 1/4W resistor....Then, we can't expect the mains supply to be stable, we can (legally!) get 253V here and the mains can get even more out of specc' abroard. 

So, the power rating must be sufficient for the basic design but a good manufacturer will build in some "headroom" . For instance the cathode bias R for 2x EL84 need only be just over a 1W component, 2W is well over the static dissipation  but under drive the cathode volts will rise to some 20V meaning over 3W will be found. Now ok, that only last for a few seconds at a time and a good quality 2W R would cope. But not maybe with a valve short? Fit a 5W or better 10W jobbie. In any case the bigger thermal capacity of a 10W means it runs much cooler than the same value 2W.

Of course "bombproofing " costs, but yer pays yer money........?

Dave.

ICBM

ecc83 said:

For instance the cathode bias R for 2x EL84 need only be just over a 1W component, 2W is well over the static dissipation  but under drive the cathode volts will rise to some 20V meaning over 3W will be found. Now ok, that only last for a few seconds at a time and a good quality 2W R would cope.

Not with some musicians I know .

In my opinion components should always be rated for at least the maximum continuous power the amp can produce when driven beyond the point of common sense or musicality… because sooner or later it will be!

I would regard 5W as the minimum safe rating for that application.

ecc83

ICBM said:

ecc83 said:

For instance the cathode bias R for 2x EL84 need only be just over a 1W component, 2W is well over the static dissipation  but under drive the cathode volts will rise to some 20V meaning over 3W will be found. Now ok, that only last for a few seconds at a time and a good quality 2W R would cope.

Not with some musicians I know .

In my opinion components should always be rated for at least the maximum continuous power the amp can produce when driven beyond the point of common sense or musicality… because sooner or later it will be!

I would regard 5W as the minimum safe rating for that application.

Yes IC...YOU would and so would I! But I was pointing out the process that goes from "just" good enough thru what most cheapo firms would fit (and get away with 95% of the time) and what good, conservative design would be.

In any event, IMO a good quality 2W MF mounted slightly off the PCB would cope "forever" because the amp is never going to be driven to that extreme point for minutes on end. It is heat over time, duty cycle, that matters and music does have gaps!

Dave.

ICBM

ecc83 said:

In any event, IMO a good quality 2W MF mounted slightly off the PCB would cope "forever" because the amp is never going to be driven to that extreme point for minutes on end. It is heat over time, duty cycle, that matters and music does have gaps!

Not if some nutcase decides to do a theatrical feedback gig ending and leave the guitar leaning up against the amp at full volume .

Not that I've ever done that of course… er… cough .

I would always go with enough rating to withstand the amp being cranked to full distorted power output and left cooking like that indefinitely… that way you know it's musician-proof. Or at least the best you can do!

Cirrus

ESchap said:

Cirrus said:

Is that large short term or large overloads as a matter of regular operation? Reason I ask is that in the vox tone circuit there's a 56k resistor that has about 180v across it. I've thought this'd be a good place to try a carbon comp but never risked it because the highest rated 56k carbon comp I can find is 0.5w.

Sorry for the hijack!

HiFi collective have 1W Allen and Bradleys in 5% and 10% http://www.hificollective.co.uk/catalog/gb5631-allen-bradley-rc32gf563k-rcr32g563k-p-5717.html

Thanks @ESchap, just ordered a 2w 56k cc from there! (2w to make my amp musician proof, in the unlikely event my amp is actually played by one) :P

ecc83

ICBM said:

ecc83 said:

In any event, IMO a good quality 2W MF mounted slightly off the PCB would cope "forever" because the amp is never going to be driven to that extreme point for minutes on end. It is heat over time, duty cycle, that matters and music does have gaps!

Not if some nutcase decides to do a theatrical feedback gig ending and leave the guitar leaning up against the amp at full volume .

Not that I've ever done that of course… er… cough .

I would always go with enough rating to withstand the amp being cranked to full distorted power output and left cooking like that indefinitely… that way you know it's musician-proof. Or at least the best you can do!

In agree on an ad hoc repair basis, but for production stuff making every component choice on a "worse case" basis would simply price you out of the market, so you do what you can WHERE you can.

Peeps don't know the brutal testing that SHOULD be done and how it adds to costs, they just see the differential twixt a £300 20watter and a sub £100 one!

Dave.

jpfamps

ecc83 said:

ICBM said:

ecc83 said:

For instance the cathode bias R for 2x EL84 need only be just over a 1W component, 2W is well over the static dissipation  but under drive the cathode volts will rise to some 20V meaning over 3W will be found. Now ok, that only last for a few seconds at a time and a good quality 2W R would cope.

Not with some musicians I know .

In my opinion components should always be rated for at least the maximum continuous power the amp can produce when driven beyond the point of common sense or musicality… because sooner or later it will be!

I would regard 5W as the minimum safe rating for that application.

Yes IC...YOU would and so would I! But I was pointing out the process that goes from "just" good enough thru what most cheapo firms would fit (and get away with 95% of the time) and what good, conservative design would be.

In any event, IMO a good quality 2W MF mounted slightly off the PCB would cope "forever" because the amp is never going to be driven to that extreme point for minutes on end. It is heat over time, duty cycle, that matters and music does have gaps!

Dave.

A 2W MF cathode resistor will not survive a valve short.

I've just replaced a 10W wirewound cathode resistor that experienced such a failure.

jpfamps

ecc83 said:

chillidoggy said:

So what's the deal with resistors being rated for power? I can understand the colour code idea, but how would i know if I needed to buy a new one, or replace one, what to ask for?

Also, I see in my amps that some resistors are bigger than others , like in the photo I posted above. What's that about?

Any resistor has to be of a size and type that is capable of dissipating the power caused by the current through it. This power (in watts) is the product of the volt drop across the resistor and the current thru' it.( It is often  more convenient by simple algebra to arrive at P =V.V /R) . To take a simple example.

ECC83 with a 100k anode load and a 300V supply. Say Va =150V then P=150x150/100,000=0.225W. That should mean we can get away with a very dinky (and cheap!) 1/4W component? Not really. What would happen if the the valve shorted anode to cathode (not likely I know but bear with!) ?

Now P= 300x300/R = 0.9W! So, should we fit a 1 W component for safety? No, a 0.6W R will suffice since it is almost unheard of for a pre amp triode to short in that way.However! Peeps DO fekk about with the wrong valves and fitting say an ECC82 in there might well burn out a 1/4W resistor....Then, we can't expect the mains supply to be stable, we can (legally!) get 253V here and the mains can get even more out of specc' abroard. 

So, the power rating must be sufficient for the basic design but a good manufacturer will build in some "headroom" . For instance the cathode bias R for 2x EL84 need only be just over a 1W component, 2W is well over the static dissipation  but under drive the cathode volts will rise to some 20V meaning over 3W will be found. Now ok, that only last for a few seconds at a time and a good quality 2W R would cope. But not maybe with a valve short? Fit a 5W or better 10W jobbie. In any case the bigger thermal capacity of a 10W means it runs much cooler than the same value 2W.

Of course "bombproofing " costs, but yer pays yer money........?

Dave.

Most 0.6W MF resistors are only rated for 350V. I've seen these fail in amps (Blues Deville), in positions where the dissipation is not an issue.

ICBM

jpfamps said:

A 2W MF cathode resistor will not survive a valve short.

I've just replaced a 10W wirewound cathode resistor that experienced such a failure.

Wot no HT fuse?

jpfamps said:

Most 0.6W MF resistors are only rated for 350V. I've seen these fail in amps (Blues Deville), in positions where the dissipation is not an issue.

Same here, and even in the Hotrod Deluxe where the voltage is lower - typically the PI plate resistors.

ecc83 said:

In agree on an ad hoc repair basis, but for production stuff making every component choice on a "worse case" basis would simply price you out of the market, so you do what you can WHERE you can.

Peeps don't know the brutal testing that SHOULD be done and how it adds to costs, they just see the differential twixt a £300 20watter and a sub £100 one!

Not that much. You don't have to do it to every component, there are only a few where it's really expensive to uprate and the cumulative cost of those doesn't add anything like that.

A little HT fuse goes a long way too, and doesn't cost much.

But having seen the insides of many modern mass-produced amps, I do know the extreme penny-pinching that is applied, and feel glad I don't work in that industry since I would be having constant fights with the pinchers.

In any case, there are companies - eg Peavey - who don't seem to have any problem making amps that are both competitively priced and pretty reliable.