I have an older receiver that is humming continuosly and the filter caps are suspect. I have tried a number of different outlets and plug in configurations, no change. Let's say I don't have any capacitors of the proper voltage rating on hand and want to rule out something else being the problem. Is it madness to simply disconnect the filter caps and jump across them for a short test to see if the hum goes away? Is there any other tried and true method to confirm this is the problem without actually replacing the suspect caps? The unit is a Carver MXR-130, but my question is general in nature.
Yes, that's pretty close to madness. You could as well plug your speakers directly into mains, sort of.
Only applicable rule of thumb for caps is, if they're older than say 10-15 years, replace them.
If you want to test the caps, remove them from the amp, charge them and measure how long it takes until they significantly drop in voltage (hopefully several seconds). Or measure their capacitance.
Have fun, Hannes
Hi DreadPirate,
I am very familiar with the Carver receivers. For starters, the "mag coil" type power regulation does put an extra strain on the filter capacitors. However, I have also found many that were fine and in no need of replacement. Now, after this statement we can look at the general question you have.
The very first question for you is this: Is it a 60 Hz hum, or a 120 Hz buzz? That will tell you whether you are dealing with a power supply issue, or some other type of problem. You haven't even used any equipment yet!!
Assuming you hear a buzz ...
The only way you can test a capacitor by placing another across it (make sure the rated voltage is the same or greater than the unit you are testing) is to turn the power off and discharge the existing capacitors. You don't want to create a loud snap or bang, the current surge can cause damage to the capacitors - even new ones. Once everything is discharged, place the test capacitor terminals across the suspect unit. You can use clip leads, or "tack solder" the replacement in place. Apply power and check for any change in operation. Turn off and discharge before you touch any connections!
The easier way to check the filter capacitors is to use an oscilloscope. This is exactly the right tool for this job. While the unit is running, connect the scope ground to the chassis ground (unless you have a hot chassis, you will need an isolation transformer for that). Set your 'scope for AC coupling and the least sensitive range and connect to each supply capacitor in turn. You can decrease your volts / div setting until you get a waveform that is easy to see. Increase back to the highest volts / div setting before removing the probe and hooking on to the next capacitor. If you don't do this, you may damage the input amplifier in your 'scope. The capacitor showing a large AC component is bad. A rectified AC wave means that the capacitor is completely open, this is actually rare except in really old tube radios and other ancient gear.
I was hoping I had taken some pictures to illustrate what constitutes a capacitor going bad, but I can't find any. Oh well, you are looking at the leading edge of the ripple. As a capacitor gets older, you will see a "pip" on the leading edge. This can get quite pronounced as the condition of a capacitor gets worse over time. I never replace a capacitor that tests well unless a customer insists, and has possibly supplied new parts. Good filter caps are not cheap and the money is better spent solving real problems.
For those of you who only have a meter, there is an iffy way to check the condition of a filter capacitor. Measure the AC voltage across the capacitor in operation, it will read higher as a capacitor fails. This will probably miss showing a capacitor that is in a failure mode, but will definitely show up a bad one - as in your case here, DreadPirate. Having the right tools makes life so much easier.
Now, for those who do have a proper capacitor tester (a bridge or an LCR meter - possibly the Fluke and Agilent hand held meters). Testing with these is time consuming (you've got to pull the suspect part out of circuit), and provides too much information. All you need to know is okay, or not okay. I don't use an ESR meter, but I suspect it also will not tell you the whole story. I use an HP 4263A LCR meter normally, but not to check filter caps for total failure unless I can not power up a unit for some reason. This is then the next best thing and I'm looking at the rough capacitance indication and more at the losses displayed. Even then, a visual inspection for popped seals or vents, or leaking electrolyte, is faster and a great pre-check.
-Chris
I am very familiar with the Carver receivers. For starters, the "mag coil" type power regulation does put an extra strain on the filter capacitors. However, I have also found many that were fine and in no need of replacement. Now, after this statement we can look at the general question you have.
The very first question for you is this: Is it a 60 Hz hum, or a 120 Hz buzz? That will tell you whether you are dealing with a power supply issue, or some other type of problem. You haven't even used any equipment yet!!
Assuming you hear a buzz ...
The only way you can test a capacitor by placing another across it (make sure the rated voltage is the same or greater than the unit you are testing) is to turn the power off and discharge the existing capacitors. You don't want to create a loud snap or bang, the current surge can cause damage to the capacitors - even new ones. Once everything is discharged, place the test capacitor terminals across the suspect unit. You can use clip leads, or "tack solder" the replacement in place. Apply power and check for any change in operation. Turn off and discharge before you touch any connections!
The easier way to check the filter capacitors is to use an oscilloscope. This is exactly the right tool for this job. While the unit is running, connect the scope ground to the chassis ground (unless you have a hot chassis, you will need an isolation transformer for that). Set your 'scope for AC coupling and the least sensitive range and connect to each supply capacitor in turn. You can decrease your volts / div setting until you get a waveform that is easy to see. Increase back to the highest volts / div setting before removing the probe and hooking on to the next capacitor. If you don't do this, you may damage the input amplifier in your 'scope. The capacitor showing a large AC component is bad. A rectified AC wave means that the capacitor is completely open, this is actually rare except in really old tube radios and other ancient gear.
I was hoping I had taken some pictures to illustrate what constitutes a capacitor going bad, but I can't find any. Oh well, you are looking at the leading edge of the ripple. As a capacitor gets older, you will see a "pip" on the leading edge. This can get quite pronounced as the condition of a capacitor gets worse over time. I never replace a capacitor that tests well unless a customer insists, and has possibly supplied new parts. Good filter caps are not cheap and the money is better spent solving real problems.
For those of you who only have a meter, there is an iffy way to check the condition of a filter capacitor. Measure the AC voltage across the capacitor in operation, it will read higher as a capacitor fails. This will probably miss showing a capacitor that is in a failure mode, but will definitely show up a bad one - as in your case here, DreadPirate. Having the right tools makes life so much easier.
Now, for those who do have a proper capacitor tester (a bridge or an LCR meter - possibly the Fluke and Agilent hand held meters). Testing with these is time consuming (you've got to pull the suspect part out of circuit), and provides too much information. All you need to know is okay, or not okay. I don't use an ESR meter, but I suspect it also will not tell you the whole story. I use an HP 4263A LCR meter normally, but not to check filter caps for total failure unless I can not power up a unit for some reason. This is then the next best thing and I'm looking at the rough capacitance indication and more at the losses displayed. Even then, a visual inspection for popped seals or vents, or leaking electrolyte, is faster and a great pre-check.
-Chris
Hi Anatech,
Thorough response, thanks. I do have a Tek oscilloscope, but so far have not developed the necessary skills, so it sits. One day I will, though. I did place a meter across both the 10,000/63V and 8200/42V caps. In DC mode, I'm reading 60V on the former, 39V on the latter. Not much wiggle room here, I will opt for larger value voltage caps upon replacement. What did surprise me are the readings I obtained in AC mode: 135VAC across the 10K and 89VAC across the 8.2K! Do I need to check my rectification diodes at this point? I also looked at the suspect power resistors that cook the boards, I will be replacing those for sure, they heat up quite a bit. I am getting a very loud hum, but can hear the source in the background. Unsure as to 60Hz or 120Hz, though. By the way, this is definitely a hum, not a buzz like is in most of my other Carver stuff to varying degrees.
Thorough response, thanks. I do have a Tek oscilloscope, but so far have not developed the necessary skills, so it sits. One day I will, though. I did place a meter across both the 10,000/63V and 8200/42V caps. In DC mode, I'm reading 60V on the former, 39V on the latter. Not much wiggle room here, I will opt for larger value voltage caps upon replacement. What did surprise me are the readings I obtained in AC mode: 135VAC across the 10K and 89VAC across the 8.2K! Do I need to check my rectification diodes at this point? I also looked at the suspect power resistors that cook the boards, I will be replacing those for sure, they heat up quite a bit. I am getting a very loud hum, but can hear the source in the background. Unsure as to 60Hz or 120Hz, though. By the way, this is definitely a hum, not a buzz like is in most of my other Carver stuff to varying degrees.
you may also keep in mind that even though capacitors may old but presumably not totally dead meaning that 60 hz noise will not come out of them ..... if they are half dead they wiil present their problem while operating in some ammount of power ..... using it at no play volume will not produce noise and at full or high power expect very distorted sound and much lower levels of power than expected .
you may wana check that your problem is "mechano" meaning some connection is loose or broken , some plug . or generally somewhere earth needs to be provided and its not .
also ( even though i am not familiar with the circuit ) you might have a problem with the triac area. but others should advice you there .... some leak there in the triac drive might create such an issue ....but still i cannot help you there ... rectifier board might also be a nice place to look
(((( hello Chris ....nice to see you very active !!! hope you are doing just fine ))))
you may wana check that your problem is "mechano" meaning some connection is loose or broken , some plug . or generally somewhere earth needs to be provided and its not .
also ( even though i am not familiar with the circuit ) you might have a problem with the triac area. but others should advice you there .... some leak there in the triac drive might create such an issue ....but still i cannot help you there ... rectifier board might also be a nice place to look
(((( hello Chris ....nice to see you very active !!! hope you are doing just fine ))))
Last edited:
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.