Last night I checked the amp under a more organised scheme and sure enough, the hiss was not the amps fault. It was a combination of a noisy computer interface and pick up from unshielded inputs. The wiring is nothing short of bench-top mayhem at the moment so I should have known better. The firewire interface is an Apogee Ensemble, a high quality unit that was known to have noise issues in early models. I was aware of my noisy source, though I was sure I had checked the amp without an input, which is why I came here to ask about the hiss. The high sensitivity of the speaker/amp combo must have revealed far more of the interfaces hiss than I had become accustomed to.
linuxguru: Thanks for the list of worthy upgrades. There have been many, many iterations of components that I have seen applied to this amp but I decided to simply build the kit as is and see what I got first time around. I have another 2 monoblocks so I might build them with some modifications and your list is a great starting point.
jonclancy: Yeah, the leads were twisted and rather short, but moving them around the bench resulted in differing amounts of noise so they are picking up something. I'm using large AC wall warts as power right now while I decide on a toroidal and I can easily get hum and buzz (plus hiss I think) from them by moving input/output leads around nearby.
Long and short, I should have tested more thoroughly before coming here. I'll wire it all up properly, in a chassis, with a good transformer and listen from there.
Thanks everyone,
Tani.
linuxguru: Thanks for the list of worthy upgrades. There have been many, many iterations of components that I have seen applied to this amp but I decided to simply build the kit as is and see what I got first time around. I have another 2 monoblocks so I might build them with some modifications and your list is a great starting point.
jonclancy: Yeah, the leads were twisted and rather short, but moving them around the bench resulted in differing amounts of noise so they are picking up something. I'm using large AC wall warts as power right now while I decide on a toroidal and I can easily get hum and buzz (plus hiss I think) from them by moving input/output leads around nearby.
Long and short, I should have tested more thoroughly before coming here. I'll wire it all up properly, in a chassis, with a good transformer and listen from there.
Thanks everyone,
Tani.
"Long and short, I should have tested more thoroughly before coming here. I'll wire it all up properly, in a chassis, with a good transformer and listen from there."
Nonsense. Build it and comment. Everyones experiences are a learning tool for newbies monitoring silently.
I am glad you now have a reliable reference amp that can detect changes / issues upstream...
Nonsense. Build it and comment. Everyones experiences are a learning tool for newbies monitoring silently.
I am glad you now have a reliable reference amp that can detect changes / issues upstream...
That is certainly true. It is easy to forget that comments here are not just useful to the few who participate in the discussion. I guess I meant I expected myself to be more methodical in troubleshooting, given my engineering background (not electrical though), rather than searching for help straight away. But like you said, the comments help everyone.
Question on resistor types for RevC
Hi all,
I've been living with my dual mono fancy BOM build for a few months now. While it is really excellent it still seems a wee bit on the bright side for me. I am thinking about try replacing R1 & R4 like Dario and others suggest. I obtained the recommended carbon films, but I also picked up some mil spec 2w carbon compositions. I'd hate to bugger up the board by putting in and taking out parts to often. Any advice/thoughts on whether the carbon comps would yield better or worse results than the carbon films ?
Thx,
PJN
Hi all,
I've been living with my dual mono fancy BOM build for a few months now. While it is really excellent it still seems a wee bit on the bright side for me. I am thinking about try replacing R1 & R4 like Dario and others suggest. I obtained the recommended carbon films, but I also picked up some mil spec 2w carbon compositions. I'd hate to bugger up the board by putting in and taking out parts to often. Any advice/thoughts on whether the carbon comps would yield better or worse results than the carbon films ?
Thx,
PJN
Go right ahead, try out the carbon comps - I predict that it won't hurt the sonics; the equivalent noise source of the resistors is shunted by the dynamic impedance of the zeners (< 100 ohms) to ground, so the additional noise (if any) will be inaudible. The drift of the resistance is also not an issue for R1, R4.
I expect that the audible sonics will have some of the traditional attributes of carbon comps - less metallic, less sibilant, more body and aspiration.
Please post your impressions after you're done - I want to try this out too, but haven't yet found the right-sized carbon comps locally.
Last edited:
Finally, someone else willing to experiment!
Yes, please try the carbon comps and let us know the results. I am now using the carbon films, and they did reduce brightness a bit, just rounded off that annoying edge. I don't know what a carbon comp might do to the sound.
Don't worry too much about damaging the board there; it's an easy spot to get at, and if you destroy a pad, you can always jumper a small guage wire directly to the adjacent components.
Another thing you might try with far less risk to the board (but higher cost) is better RCA and binding posts. I've proven to myself time and again that they really do make an improvement, usually in high frequency smoothness and resolution.
Peace,
Tom E
Yes, please try the carbon comps and let us know the results. I am now using the carbon films, and they did reduce brightness a bit, just rounded off that annoying edge. I don't know what a carbon comp might do to the sound.
Don't worry too much about damaging the board there; it's an easy spot to get at, and if you destroy a pad, you can always jumper a small guage wire directly to the adjacent components.
Another thing you might try with far less risk to the board (but higher cost) is better RCA and binding posts. I've proven to myself time and again that they really do make an improvement, usually in high frequency smoothness and resolution.
Peace,
Tom E
I have used the Eichmann Silver Bullets. They do sound better than the inexpensive Switchcraft RCA's from Mouser. Expensive and one of the last things I would ever do to a system as the bang to buck ratio is not very impressive. Power supplies and passive components seem to be a better investment to me.
Solid single strand silver for jewelers is very very inexpensive and has a good return on investment. No, not 5 9's and nothing special about it, just cheap and sounds better than any other wire I have used.
Uriah
Solid single strand silver for jewelers is very very inexpensive and has a good return on investment. No, not 5 9's and nothing special about it, just cheap and sounds better than any other wire I have used.
Uriah
My current favorite connectors are solid copper RCA's and binding posts from Vampire. They have a full range of connectors, and all are very good, but the copper parts sound best. I would avoid Neutrik and cheap stuff sold by many vendors.
I don't care at all for Eichmann parts, as they use cheap plastic and charge an outrageous amount of money for them. I also don't like silver anywhere in my system except bass speaker cables. Others, such as Uriah mentioned, like silver. I always found it bright and forward to the point of not being able to listen, and I've tried it in many different places and configurations.
Furutech makes a very nice RCA plug with PCOCC central pin. WBT stuff also sounds good, but costs considerably more and I'm not sure it's worth it.
Don't forget about power cords, as these amps are very sensitive to them. You don't need to spend much money if you make your own with stranded 12 guage copper wire (nothing fancy or heavier needed) and Furutech plugs, or Wattgate minimum.
All of these parts are available from Parts Connexion, Take Five Audio, or Sonic Craft. Their prices are comparable, and service from any of them is excellent.
Peace,
Tom E
I don't care at all for Eichmann parts, as they use cheap plastic and charge an outrageous amount of money for them. I also don't like silver anywhere in my system except bass speaker cables. Others, such as Uriah mentioned, like silver. I always found it bright and forward to the point of not being able to listen, and I've tried it in many different places and configurations.
Furutech makes a very nice RCA plug with PCOCC central pin. WBT stuff also sounds good, but costs considerably more and I'm not sure it's worth it.
Don't forget about power cords, as these amps are very sensitive to them. You don't need to spend much money if you make your own with stranded 12 guage copper wire (nothing fancy or heavier needed) and Furutech plugs, or Wattgate minimum.
All of these parts are available from Parts Connexion, Take Five Audio, or Sonic Craft. Their prices are comparable, and service from any of them is excellent.
Peace,
Tom E
Update on latest mods to RevC
Hi guys,
I finally got around to modifying my RevC this weekend. I couldn't use the carbon comp resistors for R1/R4 like I had planned because they were very large and the leads would not fit through the holes on the board. So I took out the metal films and installed the KOA carbon films on Dario's BOM, I also put FKP's in C12 and C7. Since Dario recommended 10 - 22 pF for C7, I changed it to 15 pF. The change was certainly for the better but somewhat subtle. Since I'm certain my source is DC free I then removed the coupling cap C13, bigger change for the better and not so subtle. More open, bigger/deeper sound stage, more resolute bass, the little things that were bugging me are now gone. I'm very pleased with the result.
PJN
Hi guys,
I finally got around to modifying my RevC this weekend. I couldn't use the carbon comp resistors for R1/R4 like I had planned because they were very large and the leads would not fit through the holes on the board. So I took out the metal films and installed the KOA carbon films on Dario's BOM, I also put FKP's in C12 and C7. Since Dario recommended 10 - 22 pF for C7, I changed it to 15 pF. The change was certainly for the better but somewhat subtle. Since I'm certain my source is DC free I then removed the coupling cap C13, bigger change for the better and not so subtle. More open, bigger/deeper sound stage, more resolute bass, the little things that were bugging me are now gone. I'm very pleased with the result.
PJN
I'm guessing that you mean 15 nF, not pF (15 pF would be way too small for C7).
If you haven't already done so, also try a 150nF FKP2 at C32, a 10pF Silver Mica at C34, and a 22pF Silver Mica or C0G ceramic at C10. Individually, the changes are subtle, but they combine nicely to give a perceptible improvement.
Tom,
Thanks for all of your work with this amp. I have been following and studying and am making my choices for 'upgrades'. I am going to try a sonicap and A/B it with the obligato and see if my aging ears can detect a difference. I was looking at sonicraft's website and see they can a 200 V and a 600v 1 micro fahrad compacitors. I remember a poster, a long time ago, posting the obligato was pure audio porn. Part of that, I am sure, was the size of that 630V cap.
All,
Anyway, What are the advantages of a larger, higher voltage cap? It seems that bigger is better in some situations. Is this one of them?
I have been trying to get my head around AndrewT's discussion on the circuit design regarding the input cap and resistance. I think Dario designed in the 1 uF but does that jive with the calculations? Is 1 uF a good choice? Is there a likely sonic difference between a 200 v and 600 v cap?
TIA,
Rick
Removing Input cap
For as long as I have been following this amp project, people have been removing the input cap where they are sure of the lack of DC from their source. I generally agree that less in the signal path is a good thing but given the discussion on input capacitance/impedance I wonder if it is a good choice with this amp. It seems that simply removing the input cap without changing some appropriate resistors is a detriment to the design. I have not experimented with this at all so I can't make a claim either way.
Do resistors in the circuit need to be changed if removing the input cap entirely? I am still trying to understand that discussion. Anyone have resistor value suggestions regarding the absence of an input cap?
TIA,
rick
For as long as I have been following this amp project, people have been removing the input cap where they are sure of the lack of DC from their source. I generally agree that less in the signal path is a good thing but given the discussion on input capacitance/impedance I wonder if it is a good choice with this amp. It seems that simply removing the input cap without changing some appropriate resistors is a detriment to the design. I have not experimented with this at all so I can't make a claim either way.
Do resistors in the circuit need to be changed if removing the input cap entirely? I am still trying to understand that discussion. Anyone have resistor value suggestions regarding the absence of an input cap?
TIA,
rick
It's not my design... Penasa's original BOM indicated 1uF input cap.
I'm actually using a 1.2uF one.
Yes it is, without the input cap bass response is weaker.
If the Source is guaranteed to never have output offset then it must have a virtually leakage free DC blocking cap already in the output line or it must have a DC servo to minimise the offset and a detection/protection system for the few occasions when a DC coupled source may want to pass output offset to the receiver.
If the source does have a DC blocking cap then there is no advantage in adding a second DC blocking cap to the input of the receiver, none at all. However the DC blocking cap in the receiver is more likely to be sized to suit the receiver's input characteristics than what the source manufacturer would guess what might be needed.
In this situation one retains the better of the two DC blocking caps and bypasses the redundant one.
Now to the DC servo case.
An output with a DC servo that has been designed as a DC coupled Source will have a very low output impedance. Anywhere between 10r and 200r.
If this is connected to a DC blocked receiver, the output offset of the receiver does not change.
If the Receiver has no DC blocking cap on it's input then connected the DC coupled source will change the output offset of the receiver.
There's your design problem and the choices that MUST be made. How do you want to handle the variable and potentially damaging output offset.
Mixed AC and DC coupling does not improve this situation.
In my view you either have the amp DC coupled with a DC servo AND with detection/protection or you have the receiver AC coupled.
Choose, don't mix.
If the source does have a DC blocking cap then there is no advantage in adding a second DC blocking cap to the input of the receiver, none at all. However the DC blocking cap in the receiver is more likely to be sized to suit the receiver's input characteristics than what the source manufacturer would guess what might be needed.
In this situation one retains the better of the two DC blocking caps and bypasses the redundant one.
Now to the DC servo case.
An output with a DC servo that has been designed as a DC coupled Source will have a very low output impedance. Anywhere between 10r and 200r.
If this is connected to a DC blocked receiver, the output offset of the receiver does not change.
If the Receiver has no DC blocking cap on it's input then connected the DC coupled source will change the output offset of the receiver.
There's your design problem and the choices that MUST be made. How do you want to handle the variable and potentially damaging output offset.
Mixed AC and DC coupling does not improve this situation.
In my view you either have the amp DC coupled with a DC servo AND with detection/protection or you have the receiver AC coupled.
Choose, don't mix.