I have a nice Velodyne sub that is misbehaving. In summary, a period of hum/rustling sounds after switching on that gradually resolves. The noise is unaffected by adjusing any of the preamp controls (including volume knob) or shorting inputs so I'm assuming fault is after the preamp.
Having checked the obvious candidate caps on the PSU (the bulk decoupling) and noted no significant supply ripple, I wonder if the issue might instead by to do with dried out AC coupling caps at the power amp input or on the pre-amp output.
The only issue with this idea (to me at least) is that I'd assume dried out caps would present consistently rather than, somehow, recovering after a few minutes.
The only possible exception to this might be if the cap includes some kind of self healing behaviour that occurs once a potential is applied. I am far from expert on this but I gather tntulum caps do present some type of healing that can occur. Is it nonsense to go hunting for tantalum cpas in the signal path?
The amp module takes some dismantling so any other ideas gratefully received.
Having checked the obvious candidate caps on the PSU (the bulk decoupling) and noted no significant supply ripple, I wonder if the issue might instead by to do with dried out AC coupling caps at the power amp input or on the pre-amp output.
The only issue with this idea (to me at least) is that I'd assume dried out caps would present consistently rather than, somehow, recovering after a few minutes.
The only possible exception to this might be if the cap includes some kind of self healing behaviour that occurs once a potential is applied. I am far from expert on this but I gather tntulum caps do present some type of healing that can occur. Is it nonsense to go hunting for tantalum cpas in the signal path?
The amp module takes some dismantling so any other ideas gratefully received.
Drop type tantalums? Older than 15 years? If both answers are "yes" than you have the both famous and notorious bad tantalum caps. Replace them all before they decide for you.
Modern ones are fine.
BTW technical forum so brand, type and pictures are needed to solve your problem. Otherwise the "car for sale, color red" syndrome.
Modern ones are fine.
BTW technical forum so brand, type and pictures are needed to solve your problem. Otherwise the "car for sale, color red" syndrome.
Attachments
Thanks, just to make sure I know what you mean by 'drop type'. If you mean ones that look like this then, yes, they are drop type and yes they are over 15 years old.
To be clear then, this type of cap can fail in the manner I described? And, most critically, somehow recover after a few minutes of having a voltage applied (in the correct polarity). I'm aware of the usual catastrophic type of failure (dead short, blown) but not one that would cause the symptoms I am seeing.
Cheers.
Just don't overthink, replace them as 9 out of 10 time these are exactly the notorious ones. Be careful with polarity. If modern ones with higher voltage ratings fit physically please choose those. Let's say you have 4.7 µF 16V then use 4.7 µF 25V or even 4.7 µF 35V if they have the same footprint. The modern ones in resin casing are mechanically more sturdy certainly in vibrating environments. For many small values today film caps exist and these are to be preferred. These are made by for instance Wima till about 10 µF 50V in 5 mm pitch. Never fail, never need replacement either. From 1 to 4.7 µF 50V Wima MKS2 are pretty small.
Power supply caps may need replacment too after so many years. Don't overthink, replace for something known to be good.
Now comments will follow telling you to use normal electrolytic, polymer caps etc.
Power supply caps may need replacment too after so many years. Don't overthink, replace for something known to be good.
Now comments will follow telling you to use normal electrolytic, polymer caps etc.
Last edited:
Good advice, thanks. I don't intend to 'overthink' but there will probably be several hours of delicate work to dissect the amp module to gain access to the caps in question. It is a remarkable construction of interconnected boards and heatsinking all of which needs to be desoldered and carefully taken apart. Just trying to evaluate how likely these caps are to be the issue before embarking on the task....
Cheers.
Cheers.
Typically any electrolytic cap (even solid tantalum) would have to reform if it was not used for many years (or even brand new) but forming does not take hours it may take minutes depending on the voltage and current applied. Once reformed they will be good to use again.
You may have come across a term called burn in. And yes many electronic products are burned in for some of these reasons. You may buy a new amplifier and when turned on sounds thin but after a while it sounds good. This is particularly evident when manufacturers are high volume and not capable of setting up each product they make to be burned in because of time, effort, and cost involved, they expect the user to perform this duty.
My advice, leave it turned on for an hour. If it does not perform as expected then start looking for a faulty component. Hope this helped.
You may have come across a term called burn in. And yes many electronic products are burned in for some of these reasons. You may buy a new amplifier and when turned on sounds thin but after a while it sounds good. This is particularly evident when manufacturers are high volume and not capable of setting up each product they make to be burned in because of time, effort, and cost involved, they expect the user to perform this duty.
My advice, leave it turned on for an hour. If it does not perform as expected then start looking for a faulty component. Hope this helped.
Classic drop type tantalum caps of the known notorious quality can have a spectacular burn out effect. Especially when used directly on power rails.
Sure, I understand they are likely to need to be replaced, I'm just trying to figure out if they are probable cause for the issue I am seeing.
If I dismantle the amp to repair it, I'll definitely replace them all. I'm just less excited about doing all that work without understanding if it is likely to fix the issue at hand.
Does that make more sense?
If I dismantle the amp to repair it, I'll definitely replace them all. I'm just less excited about doing all that work without understanding if it is likely to fix the issue at hand.
Does that make more sense?
Who knows? It is your device and it will likely not solve itself. Only you can verify defects.
Tantalums, electrolytic caps are the usual suspects with age. You can wait but chances are things will not repair themselves. Never had that luck except when working on others stuff 😉
Now we could continue with the "car for sale, color red" syndrome or you could mention the type name of the device. Maybe you could also try to attach its schematics/service manual.
Tantalums, electrolytic caps are the usual suspects with age. You can wait but chances are things will not repair themselves. Never had that luck except when working on others stuff 😉
Now we could continue with the "car for sale, color red" syndrome or you could mention the type name of the device. Maybe you could also try to attach its schematics/service manual.
Last edited:
electronic components can get noisy, especially transistors and can heal during the period they reach operating temperature, but it would get worse over time. Humming on the other hand could be caused by many things, but mostly a dry soldering joint, or some grounding disconnect. I would say, short the inputs and listen. Tap components with the handle of your screwdriver. If there are noises then we are looking at a dry joint or similar.
Transistor noise sounds like rustling leaves in a breeze but also erratic pops and clicks. A capacitor that went faulty but not physically broken can change tonality i.,e, no bass or low volume, sometimes noise.
Sorry, I did not read your experience properly, it could be a transistor going noisy. This you will find using "Freeze-It"spray. You start spraying every transistor in the amp until the noise disappears, or if quiet re-appears and that is the one.
Yes, I'm out of freezer spray which is why I opted to do the opposite and just heat sections of the board instaed. Moving sections of the PCB from room to hot temp had no effect on the sound. I have a can of freezer spray on order.
The dominant charcacteristic of the sound is hum. The 'rustling sound' is more like the effect of the hum cutting in and out. It does not sound like transitor failure to my ear (from previous experience of this failure type). To that end, I agree with your suggestion that a faulty earth connection could be a contributor. I've tried 'percussive debug' already but did not yield any results. I'll maybe have an other bash (pun intended' this evening.
The more I think about it, the less convinced I am that a cap could be the source of this issue. Supply decoupling seems unlikely to be the cause since supplies appear clean (at least to the resolution of my scope). I'd expect failing AC coupling to present, as you say, as a tonal shift or, worst case, a loud bang as the cap shorted open. Instead, I get hum that gradually cuts in out until it somehow resolves.
Let's go occam's razor and assume a solder or socket issue and take it from there.
The dominant charcacteristic of the sound is hum. The 'rustling sound' is more like the effect of the hum cutting in and out. It does not sound like transitor failure to my ear (from previous experience of this failure type). To that end, I agree with your suggestion that a faulty earth connection could be a contributor. I've tried 'percussive debug' already but did not yield any results. I'll maybe have an other bash (pun intended' this evening.
The more I think about it, the less convinced I am that a cap could be the source of this issue. Supply decoupling seems unlikely to be the cause since supplies appear clean (at least to the resolution of my scope). I'd expect failing AC coupling to present, as you say, as a tonal shift or, worst case, a loud bang as the cap shorted open. Instead, I get hum that gradually cuts in out until it somehow resolves.
Let's go occam's razor and assume a solder or socket issue and take it from there.
To be clear, the sub is dismantled and running on my desk in bits. I have the enclosure and the amp panel separated with the power board removed from the preamp and jumpered accross. The issue is the power board (the most interesting part of the assembly) is a sandwhich of several boards bolted and soldered together. Gaining access to the innards of this assembly carries a high risk of damaging the boards during desoldering as they are cheaply made.
In my experience a capacitor does not hum, unless it is used in the power supply and is faulty. You said that scoping it shows a stable supply. No in the past people did use tantalum caps for decoupling signals between stage, but later used Polystyrene or Polyethylene and so on because of better stability. Now I will ask you a weir question, does this hum appear every time your fridge turns on, or some other electrical commutation power turns on, a hair dryer, clothes dryer, neon tube ballast. It sound ridiculous but it can happen.