Input Coupling Cap Help
In regards to replacing an input coupling cap, someone mentioned that you DON'T want to use a cap that is physically LARGER than the original (HF instability and oscillation).
Is the nature of this problem here due to the cap being "physically larger", or is the problem using "larger values" (or both?).
If someone uses the original values same as the manufacturer, but they use a cap that is slightly larger in size, can still still cause problems?
I would like to keep the original stock value of 330nF, but the polypropylene i found is physically a bit larger (7mm more in height).
In regards to replacing an input coupling cap, someone mentioned that you DON'T want to use a cap that is physically LARGER than the original (HF instability and oscillation).
Is the nature of this problem here due to the cap being "physically larger", or is the problem using "larger values" (or both?).
If someone uses the original values same as the manufacturer, but they use a cap that is slightly larger in size, can still still cause problems?
I would like to keep the original stock value of 330nF, but the polypropylene i found is physically a bit larger (7mm more in height).
No direct experience with the 909, but I did find some problems replacing electrolytic input coupling caps with physically large polypropylenes, because the larger cap had to float above other circuit elements, and the coupling from higher level / gain circuit nodes to the outer foil of the PP input cap caused unwanted HF coupling and instability.
So, it really depends on the layout and what the larger cap can couple to. I looked up some images of the 909, and it doesn't seem too cramped - it might work? But, perhaps the cap could couple to a nearby output transistor?
Another possibility is that if it's a wound capacitor, the outer foil will be attached to one lead or the other. If the outer foil attaches to the node closer to the input (which will be lower impedance than the higher Z load node) then the coupling will be minimized. The outer foil of a wound cap is usually identified with a stripe next to the lead attached to outer foil. A stacked film cap may or may not have outer foil attached to both leads - not sure how this is commonly done, or if 'outer foil' applies to stacked caps.
A third idea is to add a grounded screen to shield the cap from external fields. This could add unwanted shunt capacitance from the input to ground, but this is relatively benign compared to HF oscillation.
So, it really depends on the layout and what the larger cap can couple to. I looked up some images of the 909, and it doesn't seem too cramped - it might work? But, perhaps the cap could couple to a nearby output transistor?
Another possibility is that if it's a wound capacitor, the outer foil will be attached to one lead or the other. If the outer foil attaches to the node closer to the input (which will be lower impedance than the higher Z load node) then the coupling will be minimized. The outer foil of a wound cap is usually identified with a stripe next to the lead attached to outer foil. A stacked film cap may or may not have outer foil attached to both leads - not sure how this is commonly done, or if 'outer foil' applies to stacked caps.
A third idea is to add a grounded screen to shield the cap from external fields. This could add unwanted shunt capacitance from the input to ground, but this is relatively benign compared to HF oscillation.
In regards to replacing an input coupling cap, someone mentioned that you DON'T want to use a cap that is physically LARGER than the original (HF instability and oscillation).
Is the nature of this problem here due to the cap being "physically larger", or is the problem using "larger values" (or both?).
If someone uses the original values same as the manufacturer, but they use a cap that is slightly larger in size, can still cause problems?
I would like to keep the original stock value of 330nF, but the polypropylene i found is physically a bit larger (7mm more in height).
Is the nature of this problem here due to the cap being "physically larger", or is the problem using "larger values" (or both?).
If someone uses the original values same as the manufacturer, but they use a cap that is slightly larger in size, can still cause problems?
I would like to keep the original stock value of 330nF, but the polypropylene i found is physically a bit larger (7mm more in height).
@Sonic77 - starting multiple threads on the same/similiar topic is against the Forum Rules, as a result, your threads have been merged.
Hi Sonic77,
7 mm may be fine, but as it has been pointed out by other members, it may also be too large for your application. If it doesn't come close to other components, you should be fine. It is a judgement call that an experienced person who had the unit in front of them could make more easily. Just make sure that the component will sit on the PCB like the original did. If you have to "fly" the component above the other parts, it is too big.
-Chris
7 mm may be fine, but as it has been pointed out by other members, it may also be too large for your application. If it doesn't come close to other components, you should be fine. It is a judgement call that an experienced person who had the unit in front of them could make more easily. Just make sure that the component will sit on the PCB like the original did. If you have to "fly" the component above the other parts, it is too big.
-Chris
Thankyou everyone really appreciate the feedback.
Can this issue also arise when "flying", or raising an "electrolytic capacitor" slightly above the PCB? Or is this only a problem with input coupling caps ect?
I would also love to replace C7 with Nichicon KZ, but they are only available in 100V (original voltage is 63V) and the Nichicon may be too large to sit flush on the board. If i install the Nichicon and have a few mm's of space between the cap and the board can that also cause issues?
Can this issue also arise when "flying", or raising an "electrolytic capacitor" slightly above the PCB? Or is this only a problem with input coupling caps ect?
I would also love to replace C7 with Nichicon KZ, but they are only available in 100V (original voltage is 63V) and the Nichicon may be too large to sit flush on the board. If i install the Nichicon and have a few mm's of space between the cap and the board can that also cause issues?
The issue arises when you attempt to fit almost any oversize component. Because of their mass and potential to tear out if the amplifier is mishandled, larger electrolytics are fitted with multiple or stronger "snap-lock" pins to secure them so they remain firmly on the PCB at all times. Many manufacturers also glue them in with large beads and gobs of silicone mastic to ensure they will remain in position both in use and transit.
If by "install" you mean to raise the capacitor to the extent of its snap-in pins, then that is not secure and a bad idea because any vibration or accidental drop of the amplifier will peel the copper foil from the underside of the PCB or loosen the pins. That could be disastrous with today's minimal trace and pad size PCBs. If you make a thin ring of as large a diameter as the original cap. from insulating material such as PVC tube and place this in the gap between cap. and PCB for support, it may help, as long as the cap still sits squarely and securely and the pins are both mechanically secured before being very well tinned and soldered.
Don't just "float" parts precariously. They were not intended to just hang anywhere on big, stiff wires as in the the old days of steam radio; prone to movement, susceptible to noise and physical damage etc. Large, unshielded foil caps at the input are also a bad idea in many amplifiers as they make ideal aerials for RF and EMI reception. You cannot shield a foil input cap by connecting the outer foil layer to ground, so you either have to wrap it all in conductive foil and ground that or just get used to the compromised audio.
If by "install" you mean to raise the capacitor to the extent of its snap-in pins, then that is not secure and a bad idea because any vibration or accidental drop of the amplifier will peel the copper foil from the underside of the PCB or loosen the pins. That could be disastrous with today's minimal trace and pad size PCBs. If you make a thin ring of as large a diameter as the original cap. from insulating material such as PVC tube and place this in the gap between cap. and PCB for support, it may help, as long as the cap still sits squarely and securely and the pins are both mechanically secured before being very well tinned and soldered.
Don't just "float" parts precariously. They were not intended to just hang anywhere on big, stiff wires as in the the old days of steam radio; prone to movement, susceptible to noise and physical damage etc. Large, unshielded foil caps at the input are also a bad idea in many amplifiers as they make ideal aerials for RF and EMI reception. You cannot shield a foil input cap by connecting the outer foil layer to ground, so you either have to wrap it all in conductive foil and ground that or just get used to the compromised audio.
Last edited:
Hi Sonic77,
I couldn't agree with Ian any more than I do. Aside from the mechanical issues which Ian pointed out, there are also electrical issues. Consider that a capacitor is a big cylinder of metal like an antenna. It's either going to transmit or receive signal from other places in the circuit. You wouldn't stick antennas throughout your unit, so why would you fly an oversized component?
Basically, the only upside to mounting a larger component in the air is that you can connect it to the circuit. Every other consideration is a strong negative. Never mind the safety aspect if you decide to fly filter capacitors.
Nope, if a component doesn't physically fit where the old part was, you shouldn't use it. Basic, simple, common sense.
-Chris
I couldn't agree with Ian any more than I do. Aside from the mechanical issues which Ian pointed out, there are also electrical issues. Consider that a capacitor is a big cylinder of metal like an antenna. It's either going to transmit or receive signal from other places in the circuit. You wouldn't stick antennas throughout your unit, so why would you fly an oversized component?
Basically, the only upside to mounting a larger component in the air is that you can connect it to the circuit. Every other consideration is a strong negative. Never mind the safety aspect if you decide to fly filter capacitors.
Nope, if a component doesn't physically fit where the old part was, you shouldn't use it. Basic, simple, common sense.
-Chris
Thankyou for the feedback guys, very informative and much appreciated.
For the input coupling cap (C2) I actually managed to find a nice polypropylene with the exact same dimensions as the original (and using same values), literally a drop in replacement so very happy with that.
As for the electrolytics, i would really like to replace C9 & C11 with Nichicon KZ, as i have heard a lot of positive feedback for these caps and genuinely believe they will improve the sound quality by a noticeable degree. These positions (C9 & C11) are on the side boards and the caps are originally positioned horizontally by Quad.
Unfortunately the Nichicon KZ are rather large, but perhaps if i were to place something in the gap between the cap and the pcb as Ian suggested this might be workable?
Trying to flesh this out because i would really love to install the KZ if possible.
For the input coupling cap (C2) I actually managed to find a nice polypropylene with the exact same dimensions as the original (and using same values), literally a drop in replacement so very happy with that.
As for the electrolytics, i would really like to replace C9 & C11 with Nichicon KZ, as i have heard a lot of positive feedback for these caps and genuinely believe they will improve the sound quality by a noticeable degree. These positions (C9 & C11) are on the side boards and the caps are originally positioned horizontally by Quad.
Unfortunately the Nichicon KZ are rather large, but perhaps if i were to place something in the gap between the cap and the pcb as Ian suggested this might be workable?
Trying to flesh this out because i would really love to install the KZ if possible.
Hi Sonic77,
I have to repair a lot of equipment where large parts were installed. Once it is returned to the customer with a proper, improved part that fits, they are normally very pleased with the result. Remember, the product has to operate properly once the work is done and you have to live with it. Large parts can also kill air circulation which may cause some other parts to run at a higher temperature. So if a part doesn't fit, don't use it!
-Chris
You will find that reports of one capacitor's high performance over another is exaggerated. While it may sound slightly better, and I will give this one the benefit of a doubt, it is still too large and that closes the matter.
I have to repair a lot of equipment where large parts were installed. Once it is returned to the customer with a proper, improved part that fits, they are normally very pleased with the result. Remember, the product has to operate properly once the work is done and you have to live with it. Large parts can also kill air circulation which may cause some other parts to run at a higher temperature. So if a part doesn't fit, don't use it!
-Chris
Thankyou for that, i'll heed your advice and appreciate your feedback.
Just on a separate note regarding the input coupling cap, some people recommended changing from the original value of 330nF up to 470nF, and i just wanted to ask the question, by raising the value to 470nF does this chop off some of the high frequency information?
Just on a separate note regarding the input coupling cap, some people recommended changing from the original value of 330nF up to 470nF, and i just wanted to ask the question, by raising the value to 470nF does this chop off some of the high frequency information?