I recently finished a Rod Elliott P101 MOSFET amp project. Pics on the solid-state project pics #6636
After some listening and getting new speakers Troels Gravesen DTQWT-mkII, something was not right with the sound. Like some odd equaliser effect.
I took a guess at C1 the DC blocking input cap. I had gone with 10 micro f, Elna CE-BP 100v.
I removed that and linked the PCB, turned on. 110mV DC at speaker terms. OK for a quick test. Straight away sounds is normal. Weird EQ effect gone. In fact the sound was really good. Test done power down.
Quick look back at Rods site. 1mf mtk or 10mf caps for C1. Looking in my stores (card board box of bits) 1mf mtk cap. Solder that in. 25mV at speaker terms. But sound slightly down from no cap. Obviously running with no cap not an option. So any help on what caps sound good in this amp input line role?
Current cap would look like a Red Wima 63V.
Thanks
After some listening and getting new speakers Troels Gravesen DTQWT-mkII, something was not right with the sound. Like some odd equaliser effect.
I took a guess at C1 the DC blocking input cap. I had gone with 10 micro f, Elna CE-BP 100v.
I removed that and linked the PCB, turned on. 110mV DC at speaker terms. OK for a quick test. Straight away sounds is normal. Weird EQ effect gone. In fact the sound was really good. Test done power down.
Quick look back at Rods site. 1mf mtk or 10mf caps for C1. Looking in my stores (card board box of bits) 1mf mtk cap. Solder that in. 25mV at speaker terms. But sound slightly down from no cap. Obviously running with no cap not an option. So any help on what caps sound good in this amp input line role?
Current cap would look like a Red Wima 63V.
Thanks
m = milli
f = femto
µ or u = micro
F = farad
I presume you mean 1 and 10 micro farads? The use uF or µF or you'll spread
no end of confusion.
What's the time constant, that's the key decision. For film 5 Hz (30ms) is reasonable, for electrolytics you want to swamp any distortion and 1Hz (160ms time constant) is more like it.
The input blocking capacitor is not necessary, the DC offset must be from the source.
Yes 1µf, sorry to lazy with phone to work out how to make micro symbol.
Thanks, I will learn about time constant, as I don't understand.
1µf is stated in the instructions, 1µf input cap gives a low frequency -3dB frequency of 7Hz.
Regarding DC off set at speaker terminals. My source is a Cyrus VS2pre, which I measure 0.04mVDC at pre out, Fluke 179.
The 100mV at Speaker terminals with no input cap, was with no source connected. Put 1uf or 10uf input cap in the circuit I have around 27mVDC.
Thanks for the help guys. I had thought to run with no caps would be fine. As I have DC detection and cut out at speaker terms.
Guess I need find out why I get the DC offset with no cap. Check cable, RCA socket, PCB. Both left and right channels have same results currently.
Thanks, I will learn about time constant, as I don't understand.
1µf is stated in the instructions, 1µf input cap gives a low frequency -3dB frequency of 7Hz.
Regarding DC off set at speaker terminals. My source is a Cyrus VS2pre, which I measure 0.04mVDC at pre out, Fluke 179.
The 100mV at Speaker terminals with no input cap, was with no source connected. Put 1uf or 10uf input cap in the circuit I have around 27mVDC.
Thanks for the help guys. I had thought to run with no caps would be fine. As I have DC detection and cut out at speaker terms.
Guess I need find out why I get the DC offset with no cap. Check cable, RCA socket, PCB. Both left and right channels have same results currently.
When checking DC offset, short the input to ground. I just use an RCA plug with a link between hot and cold.
I built a P101 over a decade ago and the DC offset was low (< 10mV) but can't remember the value.
I built a P101 over a decade ago and the DC offset was low (< 10mV) but can't remember the value.
Thanks, everyday is a school day. I will try that out next week, when I get some play time.
Update on input capacitor on my P101 amp. After trying caps from Wima, Elna and Nichicon muse.
I have been trying some crossover type caps. Clarity cap CSA, Jantzen silver Z cap and Jantzen Alumen Z cap all at 5.6uf. Alumen Z is the answer for my ears. All frequencies seem balanced. Vocal clear. Odd time smearing is gone. Now the P101 can play how I hoped the project would turn out.
Quite amazing, interesting and frustrating at times how much sound signature the input cap, gives an amplifier.
I have been trying some crossover type caps. Clarity cap CSA, Jantzen silver Z cap and Jantzen Alumen Z cap all at 5.6uf. Alumen Z is the answer for my ears. All frequencies seem balanced. Vocal clear. Odd time smearing is gone. Now the P101 can play how I hoped the project would turn out.
Quite amazing, interesting and frustrating at times how much sound signature the input cap, gives an amplifier.
Thanks. I tried all of those caps on my speakers also. As 5.6uf is tweeter size. Alumen Z is King there to. To more in post for there now also.
I spent a lot of time reading Humble Homemade HiFi cap test ��
I spent a lot of time reading Humble Homemade HiFi cap test ��
Too bad that article only addresses passive loudspeaker crossovers. The low impedance and vastly higher currents there can result in mechanical forces that in turn, could modulate the audio currents flowing within the cap. I doubt that the minuscule currents in an input cap. will have such an affect, if any, apart from the fact that being large, they may form a significant aerial for picking up noise and out-of phase rail interference which may well confuse informal "cap rolling" tests.
I'm not saying that you won't hear any difference between caps in your sighted, delayed comparisons but you need to look elsewhere for rational explanations at signal level audio. Regarding MKT caps, I think it is well known that they are no-nos for audio. Cheap but any good for audio? Generally, no because they do introduce measureable distortion according to D. Self's articles, for one reference. This is a function of the PET (polyester) film used in their construction which is not neutral to electrical fields in the way PP, PS, PE etc films are.
The alternatives are much larger, more expensive etc. but you probably get what you pay for and the best popular input types are polypropylene film/foil type (FKP) followed by standard PP film types (MKP, MKS etc) in commercial products. This EPCOS (TDK) article provides the facts on industry standard type caps, though some of the abbreviations used may differ from those used in textbooks and by other manufacturers.
There are are also amazingly expensive "boutique" class components made by specialists catering primarily to audiophiles but I'm not going to discuss choices in that highly contentious and personal area of audio experience.
I'm not saying that you won't hear any difference between caps in your sighted, delayed comparisons but you need to look elsewhere for rational explanations at signal level audio. Regarding MKT caps, I think it is well known that they are no-nos for audio. Cheap but any good for audio? Generally, no because they do introduce measureable distortion according to D. Self's articles, for one reference. This is a function of the PET (polyester) film used in their construction which is not neutral to electrical fields in the way PP, PS, PE etc films are.
The alternatives are much larger, more expensive etc. but you probably get what you pay for and the best popular input types are polypropylene film/foil type (FKP) followed by standard PP film types (MKP, MKS etc) in commercial products. This EPCOS (TDK) article provides the facts on industry standard type caps, though some of the abbreviations used may differ from those used in textbooks and by other manufacturers.
There are are also amazingly expensive "boutique" class components made by specialists catering primarily to audiophiles but I'm not going to discuss choices in that highly contentious and personal area of audio experience.
is this really what is happening when you used 10uF elna cap and the film cap?
Capacitor Burn-In/Break-In
xrk971 had reported and measured increased distortion when new amp was built ( with a new input coupling cap) and then after playing for hours , it vanished...
Yeah! I had the Elna in place for around 50/70hrs of listening. I have also been back to them to try again. I am sure they are genuine, I purchased from HiFi collective.
To do my test, I spilt the output from my Pre amp, in a small electrical enclosure, with multiple RCA socket out. Each set of sockets I solder the caps for test. So I can get up from chair, swap lead to next set of sockets. So very little delay in one cap to the next.
I wanted small cap to sound the best, so I could solder it in to the amp PCB. As it is I have the caps soldered into a lead outside of my amp.
I wasn't out to try and use this sort of cap, it is just what I am hearing and thought it worth sharing what I hear.
It is all IMHO and others hearing or testing may/probably will vary.
It seems to be something in the make up of the Alumen Z that makes all the difference. Annoyingly as they are kinda costly.
To do my test, I spilt the output from my Pre amp, in a small electrical enclosure, with multiple RCA socket out. Each set of sockets I solder the caps for test. So I can get up from chair, swap lead to next set of sockets. So very little delay in one cap to the next.
I wanted small cap to sound the best, so I could solder it in to the amp PCB. As it is I have the caps soldered into a lead outside of my amp.
I wasn't out to try and use this sort of cap, it is just what I am hearing and thought it worth sharing what I hear.
It is all IMHO and others hearing or testing may/probably will vary.
It seems to be something in the make up of the Alumen Z that makes all the difference. Annoyingly as they are kinda costly.