hi,
I've built the P3A and listen to it for the past few months. I must say this is an exceptionally good amp. I've pushed this amp thru some torturing test, and still able to survive. But the trouble is that, this particular amp sounds too bright and trashy for my taste. i've tried using several mini monitors on this amp but the same result.
i am running the amp with the 2sc3281 and 2sa1302 pairs(where i obtained sometime ago), the rest of the transistors are the BD and BC sets(as stated in Rod's recommendation), +-38V after rectifier. Source is direct from the cd player thru a passive pre-amp.
Any suggestions? And anyone knows where to get reliable parts in Malaysia?
thanks
I've built the P3A and listen to it for the past few months. I must say this is an exceptionally good amp. I've pushed this amp thru some torturing test, and still able to survive. But the trouble is that, this particular amp sounds too bright and trashy for my taste. i've tried using several mini monitors on this amp but the same result.
i am running the amp with the 2sc3281 and 2sa1302 pairs(where i obtained sometime ago), the rest of the transistors are the BD and BC sets(as stated in Rod's recommendation), +-38V after rectifier. Source is direct from the cd player thru a passive pre-amp.
Any suggestions? And anyone knows where to get reliable parts in Malaysia?
thanks
Hi!
Notice that the input cap supposedly has a big influence on the sound.
Try using a non-standart (MKT / MKS) type. I use a MKP from IT (Audyn), you could also try Mundorf etc.
I don't know where to find it right now, but I think I've read somewhere on his site that Rod E. himself even uses paper-in-oil caps (verrry expensive) like Jensen for his own setup, resulting in the best sound he ever produced on his system. But I could be wrong.
Simply try it, and also try a different speaker system, since often it is the speaker that sounds harsh, not the amp... you could try reducing the volume of the tweeter a little bit, look at ESP loudspeaker crossover articles for that...
Bye,
Arndt
Notice that the input cap supposedly has a big influence on the sound.
Try using a non-standart (MKT / MKS) type. I use a MKP from IT (Audyn), you could also try Mundorf etc.
I don't know where to find it right now, but I think I've read somewhere on his site that Rod E. himself even uses paper-in-oil caps (verrry expensive) like Jensen for his own setup, resulting in the best sound he ever produced on his system. But I could be wrong.
Simply try it, and also try a different speaker system, since often it is the speaker that sounds harsh, not the amp... you could try reducing the volume of the tweeter a little bit, look at ESP loudspeaker crossover articles for that...
Bye,
Arndt
Tweakers Heaven......
Hello Hazard,
It sounds like you are new to this, so.....
There are different kinds of "bright and trashy, and a heap of different causes.
In the case of this
P3A amplifier design, the first thing to do is to make sure that C1, C3 and C5 are of good quality.
Good quality caps are low ESR caps as used in SMPS, and often come in covering of black shrinkwrap with gold lettering.
These do not have to be expensive - I regularly use these Hitano low ESR XLR series capacitors, and so far with excellently good results, and they are only 10% or so more expensive than standard types.
Different electrolytic caps will impart different distortions and consequent sonic characters, and some swear by the very expensive 'Black Gates' caps, but lesser priced good quality caps are a good starting point.
C2, C4, C6, C7, C+ and C- should NOT be ceramic, for these are measurably distorted, and cause harsh, smashing and clangy sounds to my ear.
Polypropylene is usually a good sonic choice (especially C1), avoid polyester (can sound 'dead' and coloured), and polystyrene caps have another character (like squeaking styrofoam to my ear).
VR1 should be a 'select on test' fixed resistor.
R13 and R14 can sound easier if multiple 1W or 2W carbon or MF resistors, instead of the ceramic bodied 5W types.
The BD139/140 are not very stout nor overly good sounding transistors ime/imo.
I have found in servicing experience that amplifiers with high power driver stages sound solider, more relaxed and overall more comfortable sounding.
I would even try replacing the driver transistors to those the same as trhe outputs, and adjust R11 and R12 accordingly.
There has been discussion about different coloured LED's causing different sounds, so changing D1 to different colour and changing R7 accordingly may may be beneficial (blue Led's have been mentioned favourably).
Be prepared to accept that different component types (even resistors) may cause differing sonic characters, and be prepared to change components in stages in order to guage the resulting changes.
You can 'tune' an amplifier like this for sound that you prefer.
Cables and speakers are a further subject.
If youy need more info, just say.
Eric.
Hello Hazard,
It sounds like you are new to this, so.....
There are different kinds of "bright and trashy, and a heap of different causes.
In the case of this
P3A amplifier design, the first thing to do is to make sure that C1, C3 and C5 are of good quality.
Good quality caps are low ESR caps as used in SMPS, and often come in covering of black shrinkwrap with gold lettering.
These do not have to be expensive - I regularly use these Hitano low ESR XLR series capacitors, and so far with excellently good results, and they are only 10% or so more expensive than standard types.
Different electrolytic caps will impart different distortions and consequent sonic characters, and some swear by the very expensive 'Black Gates' caps, but lesser priced good quality caps are a good starting point.
C2, C4, C6, C7, C+ and C- should NOT be ceramic, for these are measurably distorted, and cause harsh, smashing and clangy sounds to my ear.
Polypropylene is usually a good sonic choice (especially C1), avoid polyester (can sound 'dead' and coloured), and polystyrene caps have another character (like squeaking styrofoam to my ear).
VR1 should be a 'select on test' fixed resistor.
R13 and R14 can sound easier if multiple 1W or 2W carbon or MF resistors, instead of the ceramic bodied 5W types.
The BD139/140 are not very stout nor overly good sounding transistors ime/imo.
I have found in servicing experience that amplifiers with high power driver stages sound solider, more relaxed and overall more comfortable sounding.
I would even try replacing the driver transistors to those the same as trhe outputs, and adjust R11 and R12 accordingly.
There has been discussion about different coloured LED's causing different sounds, so changing D1 to different colour and changing R7 accordingly may may be beneficial (blue Led's have been mentioned favourably).
Be prepared to accept that different component types (even resistors) may cause differing sonic characters, and be prepared to change components in stages in order to guage the resulting changes.
You can 'tune' an amplifier like this for sound that you prefer.
Cables and speakers are a further subject.
If youy need more info, just say.
Eric.
The brigth sound...or why people don't like passives preamps...
Hi Hazard!
The problem you have is that your amp has a input low pass filter (R1-C2) and the pot feeding your amp has a maximal output resistence of 12K5 Ohms
This 12K5 Ohms add to the 2K2 (R1) and the input filter become in action to low in the audio band...
So you must change the 220 pF (C2) capacitor to 33pF...for the same cut of frequency in the midle point of the input pot!!!
Put the pot near the input of the amp...and use from the pot to the amp low capacity interconects...
After that change i believe that you will enjoy the sound of your amp...good design is much more important that the brands of parts you are using...
Cheers
PS: this problem is the cause of the bad name of passive amps in certain circles...
Hi Hazard!
The problem you have is that your amp has a input low pass filter (R1-C2) and the pot feeding your amp has a maximal output resistence of 12K5 Ohms
This 12K5 Ohms add to the 2K2 (R1) and the input filter become in action to low in the audio band...
So you must change the 220 pF (C2) capacitor to 33pF...for the same cut of frequency in the midle point of the input pot!!!
Put the pot near the input of the amp...and use from the pot to the amp low capacity interconects...
After that change i believe that you will enjoy the sound of your amp...good design is much more important that the brands of parts you are using...
Cheers
PS: this problem is the cause of the bad name of passive amps in certain circles...
I think I did everything wrong by u guys lolI used ceramics for most of the low capacitance ones.
Tho i'm getting decent (although still wirewound) resistors and a replacement C1.
Mines only a car amplifier though (running off rod's SMPS) and unregulated. Sofare running off a computer PSU I can hear no switching noise at all but I've order input chokes for the +- rails to get rid of alternator him.
I've got a bit of extra money to throw round playing so I'll see what happens.
Any suggestions?
First, what is the output stage idling current? You may have too much crossover notch. Check the voltages across R13 and R14, with no input. You should see around 20mV to 50 mV across each.
In addition to the other posters' good reccommendations, you can modify the circuit for much better subjective performance quite cheaply, since you say that you haven't done a final layout yet:
Reduce the loading effect of Q4 on Q1 by adding an NPN emitter follower between Q1 collector and Q4 base. With teh new transistor's base to junction of Q1 collector and R6, it's emitter to Q4 base, with a 2K or so resistor to + rail, and it's collector to gound.
Keep C4 between Q1 and Q4 collectors.
You'd have to increase R6 to 1K or so, then also add emitter degeneration resistors, 100 ohm should be just right, in series with each of Q1 and Q2 emitters to collector of Q3.
VAS current is a little low, and teh driver output stage may be loading it at highr output levels, but try the above first.
I'd also consider R9, R10 and C5 with a proper constant current source if the previously mentioned mods don't improve things enough for you.
Anoter question would how youv'e managed teh power and ground distribution, and where do you take feedback resitor R5 from? Are all these connections wired up in star fashion? Where do you obtain the ground sides of the input bias and feedback down components? That can also make a big difference in subjective sound quality.
Have fun,
Lukas
First, what is the output stage idling current? You may have too much crossover notch. Check the voltages across R13 and R14, with no input. You should see around 20mV to 50 mV across each.
In addition to the other posters' good reccommendations, you can modify the circuit for much better subjective performance quite cheaply, since you say that you haven't done a final layout yet:
Reduce the loading effect of Q4 on Q1 by adding an NPN emitter follower between Q1 collector and Q4 base. With teh new transistor's base to junction of Q1 collector and R6, it's emitter to Q4 base, with a 2K or so resistor to + rail, and it's collector to gound.
Keep C4 between Q1 and Q4 collectors.
You'd have to increase R6 to 1K or so, then also add emitter degeneration resistors, 100 ohm should be just right, in series with each of Q1 and Q2 emitters to collector of Q3.
VAS current is a little low, and teh driver output stage may be loading it at highr output levels, but try the above first.
I'd also consider R9, R10 and C5 with a proper constant current source if the previously mentioned mods don't improve things enough for you.
Anoter question would how youv'e managed teh power and ground distribution, and where do you take feedback resitor R5 from? Are all these connections wired up in star fashion? Where do you obtain the ground sides of the input bias and feedback down components? That can also make a big difference in subjective sound quality.
Have fun,
Lukas
Aaaargh!!!!
Make that a PNP.......
Reduce the loading effect of Q4 on Q1 by adding an NPN emitter follower between Q1 collector and Q4 base. With teh new transistor's base to junction of Q1 collector and R6, it's emitter to Q4 base, with a 2K or so resistor to + rail, and it's collector to gound.
Make that a PNP.......
Reduce the loading effect of Q4 on Q1 by adding an NPN emitter follower between Q1 collector and Q4 base. With teh new transistor's base to junction of Q1 collector and R6, it's emitter to Q4 base, with a 2K or so resistor to + rail, and it's collector to gound.
k guys,
I think i'm gonna visit the store to top up those caps.
I'll try changing those caps first then i'll add on the emitter follower later on.
About the emitter follower, does it provide any significant improvement overall?
The main purpose of building this amp is purely... out of curiosity, due to the praises of many owner and simplicity of the circuit. This is the first time I took up the initiative to build a PCB board of my own(I think Rod will be cursing me for this),so getting the amp to work on a piece of vero is the first thing on my mind. Stupid me, that i totally neglected the quality parts used. But anyway it's part of learning. Right?
I think i'm gonna visit the store to top up those caps.
I'll try changing those caps first then i'll add on the emitter follower later on.
About the emitter follower, does it provide any significant improvement overall?
The main purpose of building this amp is purely... out of curiosity, due to the praises of many owner and simplicity of the circuit. This is the first time I took up the initiative to build a PCB board of my own(I think Rod will be cursing me for this),so getting the amp to work on a piece of vero is the first thing on my mind. Stupid me, that i totally neglected the quality parts used. But anyway it's part of learning. Right?
since you are from M'sia
octave electronics sells polyprop metal or film and jensen, Blackgate series(normal to BG-NX) great for feedback caps,
RS and Farnell are realiable for transistor and wima polyprop cap the small value like 330p. transistor i see you are using sc and sa series which is out of production. If bought it from Jalan PASAR good luck then. NEw codes are on ROD site. Motorola's distinct sound is more laid back. I am a huge motorola fan so i tend to promote their products. Go with MJL transistor. For BD140 and BD139 and BC559or 549 check the Japanese transistors for ite. 2SA970 for pnp. I have to get back to u on the sub for bd139 and bd140 forgot the code
octave electronics sells polyprop metal or film and jensen, Blackgate series(normal to BG-NX) great for feedback caps,
RS and Farnell are realiable for transistor and wima polyprop cap the small value like 330p. transistor i see you are using sc and sa series which is out of production. If bought it from Jalan PASAR good luck then. NEw codes are on ROD site. Motorola's distinct sound is more laid back. I am a huge motorola fan so i tend to promote their products. Go with MJL transistor. For BD140 and BD139 and BC559or 549 check the Japanese transistors for ite. 2SA970 for pnp. I have to get back to u on the sub for bd139 and bd140 forgot the code
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.