Some measurements.
When i switch on the amp i measure ~20mV on 0,1R Re. But it's hard to measure because it changes quickly.
After 5 minutes 19mV.
When the temperature of the heatsink stay constant (37C) i get 18,20mV-18,23mV
Then i drive the amp into clipping with about 1kHz sinus. The heatsink is still at 37C.
After several minutes i rise the frequency up to 10kHz temp is the same.
Later freq up to 50kHz fuses are blowing.
Swaping the fuses for bigger ones i continue the stress test. My heatsink warms up to 50C in several minutes but i have to stop because my dummy load (6R) is getting hot too hot.
I switch off the drive signal and measure the Ibias. I get 18,40 mV as soon as i remove the drive. Not so bad 0,20mV difference between normal condition and the condition after the stress test.
After the heatsink cools down to normal temperature (37C) i get back my original Ibias 18,20mV.
Comming back later with more measurements.....
zsaudio
When i switch on the amp i measure ~20mV on 0,1R Re. But it's hard to measure because it changes quickly.
After 5 minutes 19mV.
When the temperature of the heatsink stay constant (37C) i get 18,20mV-18,23mV
Then i drive the amp into clipping with about 1kHz sinus. The heatsink is still at 37C.
After several minutes i rise the frequency up to 10kHz temp is the same.
Later freq up to 50kHz fuses are blowing.
Swaping the fuses for bigger ones i continue the stress test. My heatsink warms up to 50C in several minutes but i have to stop because my dummy load (6R) is getting hot too hot.
I switch off the drive signal and measure the Ibias. I get 18,40 mV as soon as i remove the drive. Not so bad 0,20mV difference between normal condition and the condition after the stress test.
After the heatsink cools down to normal temperature (37C) i get back my original Ibias 18,20mV.
Comming back later with more measurements.....
zsaudio
Attachments
according to Mihai's schematic:
q17, q13, q11 BC560C Onsemi. Vendor is Farnell.
q9, q7, q6 BC550C Onsemi. Vendor is Farnell.
J1-J2 2SK170BL Idss 9,31-9,31 mA. Vendor is a diyAudio forum friend.
2SC3423Y-2SA1360Y Toshiba from Ampslab.
For Q3-Q1 i use 2SA1837-2SC4793. Vendor is Mouser
Big Power BJT's are NJL0281D/NJL0302D Onsemi from Mouser.
Re resistors are 0,1R/3W VISHAY DALE - LVR03 1%.
More measurement later.
Regards,
zsaudio
q17, q13, q11 BC560C Onsemi. Vendor is Farnell.
q9, q7, q6 BC550C Onsemi. Vendor is Farnell.
J1-J2 2SK170BL Idss 9,31-9,31 mA. Vendor is a diyAudio forum friend.
2SC3423Y-2SA1360Y Toshiba from Ampslab.
For Q3-Q1 i use 2SA1837-2SC4793. Vendor is Mouser
Big Power BJT's are NJL0281D/NJL0302D Onsemi from Mouser.
Re resistors are 0,1R/3W VISHAY DALE - LVR03 1%.
More measurement later.
Regards,
zsaudio
It's very hard to pass the 0.1uF load test without some ringing. Did you saw any amplifier immune to this test?
Frankly speaking all of my amps shows the same behaviour and never took care about it but some people use this test for stability. I usually drive the amp with about 100kHz sinus and load resistive and capacitive 2n2-1u and look for oscillation. Doing this test i've not observed any funny things up to 200MHz what is the BW limit of my scope.
But i do not know. I can test the amp within my limitations. So if anyone has some request about test i try to do it.
Best Regards,
zsaudio
Hi Mihai,
output coil is removed now and input lowpass soldered (100pF).
http://www.diyaudio.com/forums/solid-state/51370-soundstage-class-ab-amps-2.html#post1394667
post 20 by Aksa interesting. I'll give it a try.
Best Regards,
zsaudio
output coil is removed now and input lowpass soldered (100pF).
http://www.diyaudio.com/forums/solid-state/51370-soundstage-class-ab-amps-2.html#post1394667
post 20 by Aksa interesting. I'll give it a try.
Best Regards,
zsaudio
ZSAUDIO,
thank you for all your measurements and efforts to prove the performance and stability of this AMP /PCB. Great job!
I will now start and correct the slight flaws of the layout that I found during my build of the prototype:
- emitter resistor PADs being very small
- clearance / isolation ring in the middle of the PCB being very small
- increase the space around C18
- rename all the components' names to confirm to Mihai's convention
- increase the space between emitter resistors and the 4 caps in the middle of the PCB
- reconsider the functionality of the embedded PSU
If you found additional issues during your prototype-build: please tell me (per EMail).
I will now connect the AMP to a "better-quality- source" (CDP), but have to build an attenuator (I will finish and use
my lightspeed passive attenuator) first.
If Mihai's FC-100 keeps being so promising, I will open a group-buy for it in April.
I am looking forward to your listening impressions.
Best regards - Rudi_Ratlos
P.S. And - of course - I will write a "FC-100 Builder's Manual".
thank you for all your measurements and efforts to prove the performance and stability of this AMP /PCB. Great job!
I will now start and correct the slight flaws of the layout that I found during my build of the prototype:
- emitter resistor PADs being very small
- clearance / isolation ring in the middle of the PCB being very small
- increase the space around C18
- rename all the components' names to confirm to Mihai's convention
- increase the space between emitter resistors and the 4 caps in the middle of the PCB
- reconsider the functionality of the embedded PSU
If you found additional issues during your prototype-build: please tell me (per EMail).
I will now connect the AMP to a "better-quality- source" (CDP), but have to build an attenuator (I will finish and use
my lightspeed passive attenuator) first.
If Mihai's FC-100 keeps being so promising, I will open a group-buy for it in April.
I am looking forward to your listening impressions.
Best regards - Rudi_Ratlos
P.S. And - of course - I will write a "FC-100 Builder's Manual".
Last edited:
I measured the THD1. I used my diyed 1kHz ultralow THD oscillator from Linear Tech LT1115 datasheet.
My spectrumanalyzer is an EMU 0404 USB a laptop win7 x64 software is Spectraplus 5.
Sorry for the 50Hz and sideband trashes but i did not take the time to tame them.
All measurements were done for 6R resistive load.
100mW
1W
10W
30W
My spectrumanalyzer is an EMU 0404 USB a laptop win7 x64 software is Spectraplus 5.
Sorry for the 50Hz and sideband trashes but i did not take the time to tame them.
All measurements were done for 6R resistive load.
100mW
1W
10W
30W
50W
Before clipping
and the sinus signal on the scope.
Sorry guys for the mess signals but i thinks it's enough to show the amp's THD1 performance.
I'd measure THD20 if i could but i do not have the right equipment.
I know THD20 would be more interesting.
The amp (only one channel) plays music on an average speaker.
So the judgement will be on my better speakers when the other channel will be finished.
zsaudio
Before clipping
and the sinus signal on the scope.
Sorry guys for the mess signals but i thinks it's enough to show the amp's THD1 performance.
I'd measure THD20 if i could but i do not have the right equipment.
I know THD20 would be more interesting.
The amp (only one channel) plays music on an average speaker.
So the judgement will be on my better speakers when the other channel will be finished.
zsaudio
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.