I can only repeat my initial impressions with an off-the-shelf PSU and average soundcard: very very neutral, clean, and natural. Bass is thundereous and bottomless.
I will soon be able to give a more acurate description of the performance: a ES9018 dac is in the works, and I am finally close to complete the amplifier by giving it a power supply it deserves.
I have chosen a fully discrete design to complement it nicely, see here: http://www.diyaudio.com/forums/power-supplies/247142-high-performance-psu-voltage-regulator.html
PCBs will be produced soon.
cheers!
I will soon be able to give a more acurate description of the performance: a ES9018 dac is in the works, and I am finally close to complete the amplifier by giving it a power supply it deserves.
I have chosen a fully discrete design to complement it nicely, see here: http://www.diyaudio.com/forums/power-supplies/247142-high-performance-psu-voltage-regulator.html
PCBs will be produced soon.
cheers!
Last edited:
Just spend an evening working through 700 posts. Quite lengthy and distracting at times, but some valuable info here. After the first 150 posts I finally made a donation to the forum... Oh, and I have all the LA issues; ordering them regularly when they come out since vol. 0! I'm a fan of some good audio-related magazine, too, so this is my first choice and that has to be supported.
I am going to build this amp on perfboard and see how well that will work out. Unfortunately I can't find any of the transistors at my favourite suppliers, so I will go with all substitutes. In this thread BD139/140 have been mentioned. I wondered, however, if the BD131/132 might not be a better fit? They have an fT of 60MHz like the MJEs, not 3MHz like the other BDs. Current capability is also a bit higher. Any comments on this?
Will have to order the transistors and some passives, so it will take a while. Layout is ready though
I am going to build this amp on perfboard and see how well that will work out. Unfortunately I can't find any of the transistors at my favourite suppliers, so I will go with all substitutes. In this thread BD139/140 have been mentioned. I wondered, however, if the BD131/132 might not be a better fit? They have an fT of 60MHz like the MJEs, not 3MHz like the other BDs. Current capability is also a bit higher. Any comments on this?
Will have to order the transistors and some passives, so it will take a while. Layout is ready though
You might want to take Richard's advice and substitute a 500 ohm trimpot for R10 (249 ohms). This will help you tweak the THD% The JFETS are available as SO23, but the TO-92 are difficult to source from distributors. You could also use 2SK170/2SJ74 in the GR Idss rating, but the pinout is different from the devices Richard had specified.
The output transistors are best matched for Vbe +/- 5% so buy a bunch!
The output transistors are best matched for Vbe +/- 5% so buy a bunch!
getting close
I finally found the 249R dale resistors today. I had to buy 50 of them but they weren't to expensive. I only have a computer o-scope right now and I'm not sure how to use it to tweak the distortion. I think I would need a distortion analyzer? I have the transistors and FETs already I just need to sit down and match them.
I will have the output transistors in a couple of days. I ordered 100 output transistors, 50 MJE200 and 50 MJE210.
Thanks for the reply,
Bob
I finally found the 249R dale resistors today. I had to buy 50 of them but they weren't to expensive. I only have a computer o-scope right now and I'm not sure how to use it to tweak the distortion. I think I would need a distortion analyzer? I have the transistors and FETs already I just need to sit down and match them.
I will have the output transistors in a couple of days. I ordered 100 output transistors, 50 MJE200 and 50 MJE210.
Thanks for the reply,
Bob
MJE200/210
I received them a couple of days ago. I have no idea how to check if their genuine?
I received them a couple of days ago. I have no idea how to check if their genuine?
Attachments
A first test was really promising. Offsets were easily trimmed out and the scope showed no signs of oscillation with the inputs shorted and 330R load resistors.
The only thing I'm not quite sure about is the idle current. It's more than 80mA for each output transistor, for a total supply current of ~360mA per rail for a stereo amp. I was expecting something like half of that. The article was talking about >200mA for the power supply - is that meant for a single (=mono) buffer only?
The only thing I'm not quite sure about is the idle current. It's more than 80mA for each output transistor, for a total supply current of ~360mA per rail for a stereo amp. I was expecting something like half of that
. The article was talking about >200mA for the power supply - is that meant for a single (=mono) buffer only?
The turnover in this type is probably too low to make it attractive for fakers.
Jan
A first test was really promising. Offsets were easily trimmed out and the scope showed no signs of oscillation with the inputs shorted and 330R load resistors.
The only thing I'm not quite sure about is the idle current. It's more than 80mA for each output transistor, for a total supply current of ~360mA per rail for a stereo amp. I was expecting something like half of that
Looking at the circuit in fig 8, and assuming that the diode drops and the Vbe's match, you're left with 600mV across 9.4 ohms which is a nominal 64mA through each output stage. So even with a large margin of error, 180mA per channel seems excessive.
What's the voltage across the Re's in the output stage?
Jan
So the bias current is 390/4.7=83mA, a bit more than my very rough guestimate. Still is only 166mA for two channels.
Where does the 360mA number come from you mentioned above?
Edit are the bias voltages across the Re's the same (sort of) in each channel?
Jan
Total current draw for a stereo setup (2 channels) on the positive rail (15V lab psu).
Yes.
There seems a disconnect there somewhere. Can you insert an ampmeter in the pos and neg psu cabling?
Wait! At which Re did you measure that 390mV, R11/R12 or R16/R17? For the bias the two last would determine the bias current.
Jan
Last edited: