known overheating power resistor issue with guitar amp

[BadSolder242]

Hi all.I have a popular brand name guitar amp,which has a known issue with 2 overheating wirewound cement type resistors (5W 330ohm) The 2 resistors have an input voltage of 42v which is reduced to 16v. I have set up a test rig with 10w 330 ohm resistors(a suggested fix,which are physically oversize)and I have attached heatsinks to both(see pics) with a PC thermo controlled fan driven by a 9v battery.I have successfully reduced the operating temperature from 27.6C to 21.3C. As I have proved the mod works,I need to install a suitable 230v to 12vDC power supply to provide the power.I have 2 fans,one is rated at 0.20A 12vDC and pushing out 53CFM(120mm diameter) the other is 0.16A 12vDC 31CFM 80mm diameter. Anyone with a suggestion for a suitable PSU to do the job and which fan? I have a copy of the scematic to view. Many thanks

[Buriedcode]

...I have successfully reduced the operating temperature from 27.6C to 21.3C. ..

Is that a typo? I don't think 28C is particularly hot for a resistor, I mean, yes using a higher wattage resistor is the best option, but if the resistor is <50C, does it need a fan?

[Benta]

Seriously? Those resistors can run at 150 C without any problem.

[ejeffrey]

You are way overdoing it.  There is no need for a fan. If the resistors were overheating (its probably not) sizing up to 10 watt and/or adding a heatsink are good fixes.  A fan is just adding parts to fail and attract dust.

Those resistors, as explained can run at well over 100 C.  You can run into problems before that if it is cooking nearby parts but 50C is nothing.  If your circuit has a problem with a power resistor at 50C the problem is not the resistor.

[John B]

Lol @ plastic cable ties. Luckily the temps are still quite controlled in this situation anyways.

If I understand you right, each resistor is dropping 26V, meaning they dissipate ~2 watts each. Quite reasonable for a 5W resistor. The only benefit to a physically larger resistor here is more surface area -> more inherent heatsinking ability.

By the way welcome to the forum. Your first lesson is about solutions looking for a problem.

[wraper]

The way you measure temperature is beyond ridiculous. You cannot expect even remotely correct measurement just by placing a huge plastic encapsulated temperature sensor onto something. Not to say it's barely touching. The only thing you measure is average air temperature in the area. The way heatsinks are attached is ridiculous too. Thermal contact is really poor as you put heatsinks on a non flat side and without any thermal paste. Also these resistors can safely operate at higher temperature than what is enough to melt cable ties which will result in heatsinks falling off and shorting something. Yeah, and this most likely will fail if amp needs to be moved somewhere. It's unlikely that resistor terminals will be able to support this contraption in presence of any mechanical impact and vibration. Not to say these resistors dissipate only 2W each and with added fan which blows air right on them it becomes totally ridiculous. Original 5W resistors are OK-ish even with passive cooling. When replaced with 10W resistors, everything else you added is rudimentary.

[BadSolder242]

Measuring the average air temp was simply to establish approximately how much heat the resistor was creating,I know they get hot! The trace side of the PCB has scorch marks on it,a diode was damaged and parts of the copper trace had lifted from the board. As I mentioned,it is a known issue and I have seen examples of same amp PCB toasted beyond repair,luckily my PCB is repairable. Use of the fan was to establish how much cooler the ambient air temp could be, as the PCB enclosure  is tight fitting on all sides,top and bottom within a 2x12" speaker cabinet enclosure.
Thanks   

[Buriedcode]

Apologies if the replies seemed a bit harsh - I think we all need to be reminded this is the beginners section.

I'll second what John B suggested.  For 2W it isn't going to go nuclear, but raising them off the board will reduce the chances of damage should these reistors suddenly (unlikely) decide to dissipate more power.

Larger resistors rated for higher power can simply dissipate more heat, so they run cooler than lower wattage devices.  The heat still has to go somewhere but raising them above the board allows better air flow and less heat conducted to the board.

It doesn't look like they are any other power devices in close proximity, so its doubtful the ambient temp where those resistors are will be high enough to push up the temperature the resistors will run at.

Honestly, I don't think you should worry.  If the voltage across it you provided, and the 330 value is correct, then it'll dissipate 2W, in a rather large space with adequate airflow, away from the power transistors (which will also heat up). I can't see these getting anything more than warm.

Edit: typos

[coppice]

Measuring the average air temp was simply to establish approximately how much heat the resistor was creating,I know they get hot! The trace side of the PCB has scorch marks on it,a diode was damaged and parts of the copper trace had lifted from the board.

How do you suppose that heat gets from the 2 big resistors to the far side of the board, when the non-trace side of the PCB does not seem to have suffered. I see those resistors have been mounted well off the board, presumably to avoid them seriously heating the PCB. You don't usually get a huge amount of heat conducted down a resistor's leads to cook the other side of the board. Many high power resistors raised well above a PCB have little impact on the PCB temperature when run at up to 150C (the typical maximum recommended temp for many power resistors). Maybe its something else which is cooking the board. Maybe its something you aren't seeing, as it only gets hot under particular operating conditions. Most amps run cool until they are thrashed.

[james_s]

The fan should be totally unnecessary, as others have mentioned it's totally normal for resistors of that style to run at 100C or more and not a problem at all. If it is scorching the PCB then try mounting them further up off it so less heat is conducted by the leads. Plastic zip ties are not suitable for anything in contact with a power resistor like that, Normal operating temperatures can melt them and if they don't get hot enough to melt they will still harden and degrade over time.

[MathWizard]

If you still wanted some more cooling, what about just drilling a few air holes ?

A typical 12V PC fan, will run just fine off 5V, from say a 5V wall charger, and probably use 50-60mA. IDK your amps' PSU section, but it probably has 5 or 12V rails. In all likely hood, it could handle running a fan. IDK if that mould make inference in the audio circuits tho.

If it was me and those resistors were cooking the PCB over time, I'd do what others said and raise them up some. Or I'd double them up, 2 in parallel with 2*R each, or 2 in series w/ R/2 each, and 1/2 the power each.

How hot do they get to your finger ?

[David Aurora]

Measuring the average air temp was simply to establish approximately how much heat the resistor was creating,I know they get hot! The trace side of the PCB has scorch marks on it,a diode was damaged and parts of the copper trace had lifted from the board.

How do you suppose that heat gets from the 2 big resistors to the far side of the board, when the non-trace side of the PCB does not seem to have suffered. I see those resistors have been mounted well off the board, presumably to avoid them seriously heating the PCB. You don't usually get a huge amount of heat conducted down a resistor's leads to cook the other side of the board. Many high power resistors raised well above a PCB have little impact on the PCB temperature when run at up to 150C (the typical maximum recommended temp for many power resistors). Maybe its something else which is cooking the board. Maybe its something you aren't seeing, as it only gets hot under particular operating conditions. Most amps run cool until they are thrashed.

Like the guy said, this is a known and common issue. Fender don't give a shit as long as things limp across the warranty finish line, they're STILL releasing amps with these same design flaws after decades.

The simple solution for these and many other Fenders (Hot Rod series in particular) is to raise the resistors off the boards SECURELY. Not just long skinny legs so that the body of the resistor bounces around and cracks the solder joints, you need to keep them in place as if the legs get hot and the part is unsecured then your joints will fail. Depending on the model there are different ways to do it best, often it's also worth scraping back some solder mask at the joints to get a sturdier connection.

[John B]

I can't confirm the issue in this instance, but if you need or want to replace the resistor, another option may be aluminium cased resistors bolt to the chassis, then connected to the board with flying leads. Chassis provides heatsinking, whole setup is still silent, and the heat is generated off board.

[james_s]

I've also seen metal brackets for mounting the ceramic resistors to the chassis. They're not meant for transferring heat to the chassis but hold the resistors securely in a location where the heat they produce won't hurt anything.

[Old Printer]

To the OP, if he hasn't left town... power supplies and motors make electrical noise, putting them near an amp should be avoided if possible. Everybody is right, this is not a major problem. If it bugs you, a higher watt resistor(s) mounted on a heat sink type aluminum plate should be more than enough. Guitar amps are pretty rugged, more so in the old days and not terribly efficient. Nature of the beast. Welcome to the EEV.

[bobbydazzler]

Get power resistors you can mount to the chassis.  But I don't see the problem really if they're only dissipating 2w, board scorching can be prevented by raising the resistors off the board slightly.