VSSA PCB gerbers already prepared. This week I'll visit three PCB+SMD assembly companies and choose the one to produce VSSA.
Two MUR160 transformed to SMD, now ES1D, Vishay, also preinstalled. I'll also try to set faston terminals to be preinstalled
There will be only 6 electrolytic capacitors, two MKT capacitors, two BCs, KSA/KSC and ALF mosfet to solder manually.
Thanks LC,
That will solve my worry on installing SMD parts!
chunek
Nice Linear PSU schematic with good care of diodes switching parasitic harmonics. Can be improved (may-be ?) with dumping network at the trasfo coils, and, ('i'm sure of that) just paralelling Zener or voltage regulators to the gate capacitor in order to regulate-it.
L.C, i am just asking myself a question. If we use a regulated SMPS, with HF switching frequency, could we have some benefits to use big film caps (lower µF value, less inductance) instead of lytics for the input stage and VAS DC filters of the VSSA ?
Last edited:
Now we have 10 Ohm and 1 mF/50 V filtering for the VAS/input stage DC rails, since T=R*C, the filtration is effective from 100 Hz up. If SMPS has 100 kHz we could use let say 1 uF/10 Ohm for the same HF ripple filtration job only. But there's also ripple caused by LF loading currents, so something in between 1 uF to 1 mF is probably correct to choose. My suggestion would be still to use quality 1 mF/50 V electrolytics like Nichicon FG or Panasonic FC or ELNA Silmic, etc. Smaller film capacitors would serve fine for HF decoupling but will not be really effective for filtration and current providing in LF audio spectrum, especially when amp's hardly loaded.
VSSA has already 1 uF multilayer SMDs parallel to all electrolytics at DC rails, as I know well that you don't like that much because of cap resonances, but until now they've proved to be done very good job in the circuit.
I think you have forgotten the 2xpi bit, surely its from 16Hz up?
Yes, 16 Hz would be cuttoff frequency where filter attenuates the signal to half its unfiltered power and that is only -3 dB, which is not relevant to be useful in practice at PSU-s. We usually set PSU passive filters to at least -20 dB attenuation to have useful effect in practice. Concretely at 100 Hz this low pass filter (10 Ohm/1 mF) has only -8,6 dB reduction in power.
Merci bien, cher Esperado,
I suppose you speak about secondaries of TX. I didn't think about that option because T1 disconnect C2 from C1 for a significant period of the cycle...can you explain?
I do not use T2 but many C2's or another reg, the PowerReg which is the only one that I know well how to build: https://picasaweb.google.com/lh/photo/rFQvssoVepboJSN6Uj5819MTjNZETYmyPJy0liipFm0?feat=directlink
Sorry for OT (but not too much OT I hope since we want the best PS for SSAs).
M.
I suppose you speak about secondaries of TX. I didn't think about that option because T1 disconnect C2 from C1 for a significant period of the cycle...can you explain?
I do not use T2 but many C2's or another reg, the PowerReg which is the only one that I know well how to build: https://picasaweb.google.com/lh/photo/rFQvssoVepboJSN6Uj5819MTjNZETYmyPJy0liipFm0?feat=directlink
Sorry for OT (but not too much OT I hope since we want the best PS for SSAs).
M.
As you know, the high energy parasitic oscillations generated by the switch off of the rectifiers find their way across grounds, primary of the trasfo down to AC lines, generates EMI/RFI. The best is to dump them at the source, whatever the precautions you can take after-it (like T1) to filter them before amp's rails.
The boring point is the optimal RC network depends of each particular trasfo and caps and even the way they are mounted, so it is kind a real life blind try, i don't know an easy way to calculate them.
Sory for another supid question, can LM334 be used as ccs instead of 15k resistor in vssa? Temperature correlation is only 0,33%/C so it shouldnt be a big factor (there is zero tempco aplication too).
Thanks
http://www.ti.com/lit/ds/symlink/lm334.pdf
Thanks
http://www.ti.com/lit/ds/symlink/lm334.pdf
Hi trunek
This thread's VSSA PCB sch has been locked to a SMD PCB group buy version so very much tested and fixed. PCBs are now in the production phase.
Scaled up VSSA version will be started some day in another thread.
BTW last Friday we had a test in one shop on KEF R700 and Reference 205/2 speakers and shop owner already ordered stereo VSSA in a flat case.
Attachments
Hello
After you set up the mentioned bias at first you have to use the two (one each side) trimmer to get read of the offset totally, after full warm up readjust it.
Then take them out and measure the the trimmers value, what you get that would be the precise resister value.
I hope these answer to your question
Greets Gabor