Filter caps at IEC mains inlet.
There is a filter network at the IEC inlet considting of 2 caps from Earth to L and earth to N. And a large cap across L and N. Is there any advantage to be had in replacing these with better quality caps? Or even removing them?
I have some nice varistors in my kit box which I could fit in place of the 2 Ceramics from earth to L and N for spike protection. Also I have a 1uF Mundorf MKP rated at 300v so would there be any benefit by using this as a bypass on the large cap across L and N?
Mike.
There is a filter network at the IEC inlet considting of 2 caps from Earth to L and earth to N. And a large cap across L and N. Is there any advantage to be had in replacing these with better quality caps? Or even removing them?
I have some nice varistors in my kit box which I could fit in place of the 2 Ceramics from earth to L and N for spike protection. Also I have a 1uF Mundorf MKP rated at 300v so would there be any benefit by using this as a bypass on the large cap across L and N?
Mike.
Hi Mike,
The voltage rating printed on most caps is for DC, so the Mundorf is not really suitable. It's not a good idea to put this across the mains. 400vdc caps might be ok across the mains though, but others on here would definitely recommend "x" rated caps, which are certified for safe fitting across live and neutral (and y-rated caps for live-earth and neutral-earth).
I have an 8uF "lighting" cap across L&N and I like what it does for my sound.
Simon
The voltage rating printed on most caps is for DC, so the Mundorf is not really suitable. It's not a good idea to put this across the mains. 400vdc caps might be ok across the mains though, but others on here would definitely recommend "x" rated caps, which are certified for safe fitting across live and neutral (and y-rated caps for live-earth and neutral-earth).
I have an 8uF "lighting" cap across L&N and I like what it does for my sound.
Simon
Hi,
only use X or Y rated caps on the mains.
250Vac and/or 400/450Vac Motor run caps can be used, but do not connect them to Earth, they are not rated for that duty.
Motor start caps are only rated for intermittent duty.
John Curl recommends 10uF across Live to Neutral, either at the distribution board or behind the outlet socket. It is far too big to fit inside an IEC filter casing.
If you plan to fit anything that is not X or Y rated then maybe you should ask your house insurance company to confirm you are not invalidating your insurance cover. They would just love to discover an excuse to renege on paying out.
only use X or Y rated caps on the mains.
250Vac and/or 400/450Vac Motor run caps can be used, but do not connect them to Earth, they are not rated for that duty.
Motor start caps are only rated for intermittent duty.
John Curl recommends 10uF across Live to Neutral, either at the distribution board or behind the outlet socket. It is far too big to fit inside an IEC filter casing.
If you plan to fit anything that is not X or Y rated then maybe you should ask your house insurance company to confirm you are not invalidating your insurance cover. They would just love to discover an excuse to renege on paying out.
not necessarily, there is very little choice in X & Y rated and if they also need to be physically similar that reduces choice even more.mikesnowdon said:
Most X & Y rated are polypropylene, but some are ceramic or polyester.
All must survive very high voltage pulses and be self healing.
The Y rated must never short to Earth. These are onerous specifications. Ignore them at your peril.
mikesnowdon said:
Yes, I use something a bit like that. More elaborate and well-researched commercial items (plug-in filters) might be even more effective, as values and the circuit used are more complex than putting one straight across L & N. They may also contain better (faster response and more current handling) spike protection than a simple VDR.
That said, just putting ~8uF across the mains is likely to be audible. I find things sound warmer with improved imaging. A little ~0.33uF x-rated cap alone won't do anything audible IME.
Simon
Hi Mike!
What about these caps:
http://cgi.ebay.co.uk/CBB60-Washing...6743761QQihZ011QQcategoryZ66994QQcmdZViewItem
A little pricey, but suitable... Try searching them in another place , maibe in your country
What about these caps:
http://cgi.ebay.co.uk/CBB60-Washing...6743761QQihZ011QQcategoryZ66994QQcmdZViewItem
A little pricey, but suitable... Try searching them in another place , maibe in your country
Hi Mike,
Look at this from a realistic viewpoint. The inductance of that coil is very low. It will only affect current at higher frequencies where it's reactance becomes significant compared to that resistor. Up to that point, it "looks" like a wire to audio frequencies, shorting out that resistor. So that resistor isn't effective at audio frequencies really.
Concerning foil trace length ... Connection resistance is very likely higher than a short length of foil trace in your output path. Always reduce the number of connection points! The foil resistance is essentially zero compared to the impedance levels that your signal circuitry runs at. You do want to route signal level wiring away from electrical noise sources.
What I see here is that you might be over thinking things again. Try to keep everything in perspective to the actual environment that your possible defect is located.
With regard to your mains system concerns. John Curl lives in a country where the mains voltage is maybe 125 VAC, so your power dissipation will be twice what it is in North America. You may want to cut capacitances in half. At the very least, calculate your peak currents and make sure they are well within your capacitor's specifications. Do not fail to consult an electrician if you are considering any mains type stuff. Your capacitors absolutely must be rated for mains use as well. This will include safety ratings for the exposed parts, like the case.
I'm not trying to rain on your parade, just to focus your attention on things that really matter.
-Chris
Absolutely nothing wrong with that at all! Nothing will become non-linear and there will be no effect on the sound quality from doing this. That's assuming you use normal resistors and not some made-up iron thingy.
Look at this from a realistic viewpoint. The inductance of that coil is very low. It will only affect current at higher frequencies where it's reactance becomes significant compared to that resistor. Up to that point, it "looks" like a wire to audio frequencies, shorting out that resistor. So that resistor isn't effective at audio frequencies really.
Concerning foil trace length ... Connection resistance is very likely higher than a short length of foil trace in your output path. Always reduce the number of connection points! The foil resistance is essentially zero compared to the impedance levels that your signal circuitry runs at. You do want to route signal level wiring away from electrical noise sources.
What I see here is that you might be over thinking things again. Try to keep everything in perspective to the actual environment that your possible defect is located.
With regard to your mains system concerns. John Curl lives in a country where the mains voltage is maybe 125 VAC, so your power dissipation will be twice what it is in North America. You may want to cut capacitances in half. At the very least, calculate your peak currents and make sure they are well within your capacitor's specifications. Do not fail to consult an electrician if you are considering any mains type stuff. Your capacitors absolutely must be rated for mains use as well. This will include safety ratings for the exposed parts, like the case.
I'm not trying to rain on your parade, just to focus your attention on things that really matter.
-Chris
Thanks again Chris. Your advice is always appreciated.
RE: signal traces. I have posted an image below to better describe the idea. As you can see adding this new link not only shortens the signal path considerably - it also eliminates 2 solder joints and a switch from each channel. So based on your data concerning the resistance of a 'joint' its a win-win situation!
Below image to illustrate a simmilar mod on the output to the speaker terminals. Im not sure about this one. I suspect that a better soloution would be to bypass the whole lot with some good wire - soldered directly between the output coil/resistor and speaker terminal.
Im pleased there is no problem with resistors inside coils. This means I can move the resistor and do the wire mod above very neatly and with even less pcb trace.
I know were talking in terms of minimal improvements to SQ but every little bit adds up. And I can re-route the output wiring away from the Trafo. My intention is to squeeze every last bit of extra performance from the amp.
Mike.
RE: signal traces. I have posted an image below to better describe the idea. As you can see adding this new link not only shortens the signal path considerably - it also eliminates 2 solder joints and a switch from each channel. So based on your data concerning the resistance of a 'joint' its a win-win situation!
Below image to illustrate a simmilar mod on the output to the speaker terminals. Im not sure about this one. I suspect that a better soloution would be to bypass the whole lot with some good wire - soldered directly between the output coil/resistor and speaker terminal.
Im pleased there is no problem with resistors inside coils. This means I can move the resistor and do the wire mod above very neatly and with even less pcb trace.
I know were talking in terms of minimal improvements to SQ but every little bit adds up. And I can re-route the output wiring away from the Trafo. My intention is to squeeze every last bit of extra performance from the amp.
Mike.
Well there not realy concerns, just curiositys or considerations. Some have said that these filter caps do affect SQ so I'm trying to find out why and what works best. If a cap change on the mains filtering can remove just a little bit of grain from the sound (for example) then its worth it IMO. As long as its not too expensive that is, those motor caps are chap enough as are the lighting caps.
Mike.
Hi Mike,
Your plan sounds good. I wasn't being critical of your plans, but rather just trying to give you the tools to determine what is worthwhile doing for you. Wiring changes in your speaker output circuit sound like they will give you an increase in performance.
I'm not sure yet if bypassing the tone controls is okay or not. There will be a gain element (op amp or transistors) to compensate for the loss in the tone circuitry. They are normally part of the circuit and you need to make sure you are bypassing everything, not just the controls. I've been a bit short of time, so I am giving you my concerns without examining the circuit. Stuff to watch for.
-Chris
Your plan sounds good. I wasn't being critical of your plans, but rather just trying to give you the tools to determine what is worthwhile doing for you. Wiring changes in your speaker output circuit sound like they will give you an increase in performance.
I'm not sure yet if bypassing the tone controls is okay or not. There will be a gain element (op amp or transistors) to compensate for the loss in the tone circuitry. They are normally part of the circuit and you need to make sure you are bypassing everything, not just the controls. I've been a bit short of time, so I am giving you my concerns without examining the circuit. Stuff to watch for.
No problem. My response is only "within reason".
-Chris
There is of course an opamp in the tone circuit. However, with the 'direct' switch in the on position the whole tone circuit is bypassed. The only problem with this is that the traces are very long and pass through a switch. If those who have the schematics can take a look you should see what I mean. The 'input buffer' provides gain of about 6dB into the LM3886. As it is this is too high for me - the amp goes into clipping at the 12'oclock position - wasting useable range in the pot. In an earlier post I mentioned a suggestion by Greg Ball to reduce the gain in the power stages by about half. This also has a benefit of improving PSSR apparently. Regarding the coils on the output I'll be mounting the resistors inside the coil - making sure to insulate the leads with heat shrink. Then the mounting holes where the resistors were can be used to wire the output direct to the binding posts. This will be done with good quality speaker wire and new binding posts. The stock ones are very cheap and have no plating.
Hi all!
One thing I have learnt from my research into CD player mods is that its beneficial to seperate power rails, especially digital ones. Apparently digital electronics can put alot of crap onto the power rails which we dont want on our delicate output stages do we?
For this reason I have decided to purchase a suitable small transformer for the +5v rail feeding the microprocessor ic on the front panel - the area that controls the input selection and motor driver etc. It will have its own dedicated PSU mounted in the front right corner as far away as possible from the preamp sections.
regarding my buzzing problem:
Can anyone tell me what the "DC offset detect" circuit is doing? This is one area I havent investigated as a possible cause of buzzing. When I did the first round of mods I had a loud buzz on the first powerup. I shut it down and tried again with vol at zero - same thing happened. On the third attempt everything seemed fine until later that night I noticed this faint buzzing form one speaker. Could something have gone in the DC detect circuit which would cause a buzzing sound? (Details of what I have done so far to try and solve the buzzing can be found-here)
Also I would like to know if Im ok to disable the DC detect and how? This way I can easilly test if its the cause of my problem.
Thanks.
Mike.
(Reminder: I have the Service Maual if anyone would care to take a look. Thanks, Mike.)
One thing I have learnt from my research into CD player mods is that its beneficial to seperate power rails, especially digital ones. Apparently digital electronics can put alot of crap onto the power rails which we dont want on our delicate output stages do we?
For this reason I have decided to purchase a suitable small transformer for the +5v rail feeding the microprocessor ic on the front panel - the area that controls the input selection and motor driver etc. It will have its own dedicated PSU mounted in the front right corner as far away as possible from the preamp sections.
regarding my buzzing problem:
Can anyone tell me what the "DC offset detect" circuit is doing? This is one area I havent investigated as a possible cause of buzzing. When I did the first round of mods I had a loud buzz on the first powerup. I shut it down and tried again with vol at zero - same thing happened. On the third attempt everything seemed fine until later that night I noticed this faint buzzing form one speaker. Could something have gone in the DC detect circuit which would cause a buzzing sound? (Details of what I have done so far to try and solve the buzzing can be found-here)
Also I would like to know if Im ok to disable the DC detect and how? This way I can easilly test if its the cause of my problem.
Thanks.
Mike.
(Reminder: I have the Service Maual if anyone would care to take a look. Thanks, Mike.)
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.