Well , this will be a part of the "slewmaster" multiple input/linestage/DAC
project.
I saw all the lehmann clones , the one that caught my eye was the class
A diamond buffer. I added the CCS's for 2 reasons , better psrr and a thermal
offset to the main OP.
This would have to be built and tested thermally.
(below) would be an even better assurance of that stability. Notice
Q5/6 and D3/4.
D3/4 Coupled with Q8/7 would lock the buffers output at 120ma.
THD is still 10ppb.
OS
Attachments
I should be testing the SMT version of yours in a couple weeks. I've added an input selector to it as well. Still needs remote control and a DAC.
I'm not really finding anything spectacular looking for the VAS transistors. It seems VBE jumps to 6V with anything high speed. Is this possible to work with or would MJD340/350 be a better option?
http://www.mouser.com/ds/2/308/NSS1C301E-D-278477.pdf
https://www.fairchildsemi.com/datasheets/MJ/MJD340.pdf
http://www.mouser.com/ds/2/308/NSS1C301E-D-278477.pdf
https://www.fairchildsemi.com/datasheets/MJ/MJD340.pdf
Last edited:
If current is fairly light you make an oversized copper pad to cool it. If current is higher you solder a heat sink to that pad. The heatsinks look pretty cool too!
http://www.farnell.com/datasheets/1756551.pdf
http://www.farnell.com/datasheets/1756551.pdf
A preamp/headphone amp???? is it tested? i want one..... are there any schematics and files availiable for etching? OS maybe its time for a preamp/headphone amp...
That - looks like a pain in the butt. Why not keep the main VAS device
through hole ?
What fun would that be ?
I also intend on porting mine to a IRFP based 50W class A stage for
the slewmaster IPS's.
Also ...thinking of porting the line stage/headphone amp for IPS compatability.
At 24V rails , most of my IPS's can run with minor changes to the passives
and give sub PPM performance.
The op-amp is better , Jfet inputs - low noise.. might beat the BCxxx/Ksa/c
inputs of the slew modules.
OS
I thought about it when I did my own SMD power amp layout, and decided it wasn't worth trying as nothing from SMD can touch A1381/C3503 for a VAS transistor. Besides one would want to keep the parasitic capacitance at the collector of a VAS transistor as small and as predictable as possible, as that is one of the most critical nodes, if not THE most critical node, within an amp, using a big chunk of PCB space for heatsinking is definitely not a best decision for me in this context. I would just stick with TO-126 for VAS until the supply has exhausted.
I also read in a previous post about tinning the traces on a home etched SMD layout. If that is the case, one might not want to run a trace so that an SMD part sits right on it, as the part would then be sitting on the solder instead of sitting leveled on the pads. In case of reflow soldering being used, the molten solder under an SMD package may float the part causing unexpected positioning error.
Just a couple nickels, pennies are out of circulations nowadays
Last edited:
I was going to try to stay all SMD transistors on this input but I've given up on the VAS transistors. They will be through hole.
What do you use for film caps? Do you use through hole or SMD?
I was referring to liquid tin so there's no height difference.
Last edited:
Same idea as yours , a cheap pga2311 remote with one input hooked to
(below) a cheap dac. PGA running to the buffer - one could even switch
the buffer between line out or headphone (dual use).
Time to diverge from the usual class AB stuff.
OS
Attachments
I'm trying figure out remote control through Arduino. It doesn't look terrible to do. That Chinese one I've got works fairly well but it defaults to on and 25% volume on power failure.
I've got an input selector board designed. 3 stereo RCA input, DAC input and an output relay on it. I've got a cheap DAC from China. I'm getting some 60Hz from it now. The design sucks pretty badly. I think I want to start from scratch and do my own DAC as well. I just need to figure out what DAC to use and where to find it.
Attachments
That's standard for all of em' (-46db fault setting).
That must be set in the controller , because the pga2311 http://www.ti.com/lit/ds/symlink/pga2311.pdf
defaults to full mute (-96db) on power failure (datasheet says so).
OS
Seems dumb. That's pretty loud on my speakers. Why wouldn't it default to off? I could use a latching relay on the power circuit to get around it but it's more fun to design my own controller.
One approach would be to differentiate between normal power-off and a mains-fail condition. On a mains-fail state, the last volume setting can be saved to the ATMega's EEPROM in a matter of milliseconds. On mains restore, the last value can be read and reinstated, either instantaneously or gently ramped from zero volume.
The EEPROM memory has a specified life of 100,000 write/erase cycles. If this is of any concern, the non-volatile RAM (56 bytes) of a DS1307 RTC can be used to hold this information.
Why don't we go over to the "digital line level" forum and do a combo
DAC/line stage/input selector.
Either all on one board (PS/ circuitry) or a separate PS board.
PS - you are getting the 60hz from the chinese deal because
of the "cheepness" of their power supplies - the chips are not at fault.
OS
I'm in for an all in one board with a separate supply board. I want to do a chassis for it with the supply in a Mu-Metal can.
Yes the Chinese supply sucks a$$. We did better on the crossover. 3 LM317s and a couple cheap caps.