GB SMT Board Tested
I have verified that the SMT daughterboard works. I was able to get 0805 chip resistors to work with a lot of manual soldering to add enough solder to span the gap between the pads and the resistor. The 0805/1206 version will work a lot better. I used axial 1/4w resistors for the real big ones with bent leads. That was also handy to allow alligator clips on the 10R to verify bias current. On this particular unit, I am measuring 18mA bias at 45.4v Vcc. The mid-point output tie-point is exactly at 24.0v with a 9.0k resistor formed from a 6k8+2k2 chip resistor in a rooftop arrangement at R101. One thing to note is to use a very high quality 33pF NP0 capacitor. I used a lower grade NP0 and got higher distortion at high frequencies. I replaced it with a 2kV rated 33pF NP0 and distortion went back down to normal.
Closeup (note that R102 and R103 silk screen labels are reversed if using old schematic, updated schematic V2 is correct):
FFT shows THD is a tad higher than TH version, and this may be result of changes in actives (SMT equivalents are slightly different), plus using DZT5401 vs KSA1381 on the output. Also, the TH predominantly used carbon film throughout except for 47R 1% used to set bias current.
Here is FFT for 20Vpp, THD at 0.0012% and similar H2/H3 relative profile compared to TH version:
Listening to it now - sounds great.
I think we are ready to ship GB sets as all key functionality is verified. I have not verified the input selector switch or the CLC. But if using DC-DC step up, it has its own built in CLC used to block the switching frequency. In some countries (Germany for example), I have heard mains are very dirty and CLC is mandatory.
I have verified that the SMT daughterboard works. I was able to get 0805 chip resistors to work with a lot of manual soldering to add enough solder to span the gap between the pads and the resistor. The 0805/1206 version will work a lot better. I used axial 1/4w resistors for the real big ones with bent leads. That was also handy to allow alligator clips on the 10R to verify bias current. On this particular unit, I am measuring 18mA bias at 45.4v Vcc. The mid-point output tie-point is exactly at 24.0v with a 9.0k resistor formed from a 6k8+2k2 chip resistor in a rooftop arrangement at R101. One thing to note is to use a very high quality 33pF NP0 capacitor. I used a lower grade NP0 and got higher distortion at high frequencies. I replaced it with a 2kV rated 33pF NP0 and distortion went back down to normal.
Closeup (note that R102 and R103 silk screen labels are reversed if using old schematic, updated schematic V2 is correct):
FFT shows THD is a tad higher than TH version, and this may be result of changes in actives (SMT equivalents are slightly different), plus using DZT5401 vs KSA1381 on the output. Also, the TH predominantly used carbon film throughout except for 47R 1% used to set bias current.
Here is FFT for 20Vpp, THD at 0.0012% and similar H2/H3 relative profile compared to TH version:
Listening to it now - sounds great.
I think we are ready to ship GB sets as all key functionality is verified. I have not verified the input selector switch or the CLC. But if using DC-DC step up, it has its own built in CLC used to block the switching frequency. In some countries (Germany for example), I have heard mains are very dirty and CLC is mandatory.
Attachments
Last edited:
Playing in Stereo Now
Sounds very nice feeding a headphone amp. No noise, hiss, hum. Volume works butter smooth with no scratching noises - just even variation in volume of sound. I added a 6k8 3W CRC capacitor bleed-off resistor where the third cap of the CRCRC would go - 3rd cap not needed on my setup based on measurements with two 10mF CRC caps. Feel free to use third cap though with smaller 4700uF's etc.
Sounds very nice feeding a headphone amp. No noise, hiss, hum. Volume works butter smooth with no scratching noises - just even variation in volume of sound. I added a 6k8 3W CRC capacitor bleed-off resistor where the third cap of the CRCRC would go - 3rd cap not needed on my setup based on measurements with two 10mF CRC caps. Feel free to use third cap though with smaller 4700uF's etc.
Attachments
Last edited:
Also, to clarify, will this GB round be of the MELF version or the latest 0805/1206 version?
I just ordered 100 of the correct sized (6in x10in) kraft bubble padded envelopes to ship the rather large PCBs. So won’t be able to ship until this Saturday.
Spent some more time listening to the MoFo powered by this preamp tonight. Very nice.
Just to jog your memory, here are some measurements of my MoFo (1w rms into 8ohms):
Here is max power before clipping 26.8vpp (11Wrms into 8ohms):
Not too bad for a 1 transistor Class A power amp with a 24v smps.
Spent some more time listening to the MoFo powered by this preamp tonight. Very nice.
Just to jog your memory, here are some measurements of my MoFo (1w rms into 8ohms):
Here is max power before clipping 26.8vpp (11Wrms into 8ohms):
Not too bad for a 1 transistor Class A power amp with a 24v smps.
Last edited:
JPS64 - 1 set SMT, Paid
Redarc12 - 1 set, TH, Paid
CFT - 1 set, TH, Paid
Jacques Antoine - 1 set TH, Paid
Silasmellor - 1 set THT, Paid
Bvtrinh- 1 set TH, Paid
bvtrinh - 1 set TH (second set), Paid
ggetzoff - 1 set TH, Paid
tubesguy - 1 set TH, Paid
Aatto - 1 set SMT, Paid
Mordikai - 1 set TH, Paid
Johnny2Bad 2 sets total 1 TH & 1 SMT, Paid
vvs07 - 2 sets TH, Paid
PCGAB - 1 Set SMT, Paid
Harry3 - 1 set MB plus TH, Paid
Wayne1 - 1 set TH, Paid
obh - 1 Set TH, Paid
PJN - 1 set TH, Paid
Zman01 - 1 set SMT, Paid
AKSA - 1 set SMT, Paid
xrk971 - 1 set SMT, Paid
spind - 1 set TH, Paid
schlomoff- 1 set TH, Paid
Prasi - 1 set TH, Paid
muthumuthiah - 1 set TH, Paid
Algar Emi - 1 set, TH, Paid
Kokanee - 2 sets, 1 TH and 1 SMT, Paid (moved up from FB2)
—— List for GB #2 - please add name below ——
Thimios - 1 set SMT, Paid
Pinnocchio - 1 set SMT, Paid
Mituisho - 1 set SMT, Paid
Alurudin - 1 set TH, Paid
muthumuthiah - 3 sets all TH, ship all with order from GB #1, Paid
jwjarch - 1 set TH, Paid
jwjarch - 1 set of TH daughterboards only, Paid
briansz - 2 set TH, Paid
steefdebruijn - 2 sets TH + spare set SMT daughterboards, Paid
jmc207 - 2 sets: 1 TH and 1 SMT; +1 spare set TH daughterboards, Paid
marcus22ch - 2 sets SMT, Paid
vanofmonks - 1 set TH, Paid
pfisl - 1 set TH, Paid
neoinc - 1 set TH, Paid
redarc12 - 1 set of TH daughterboards only, Paid
gtose - 1 set SMT 0805/1206 version, Paid
Vunce - 1 set TH, Paid
*Malvin - 1 set SMT, Invoice sent
^^^^ I edited my name above to show that I've paid for two full sets of TH boards in GB #2.