Terry,just what i noticed.
Do not attempt to adjust the offset when amplifier start up.
Leave the amplifier to come hot and then adjust the 0 offset.
This topology without servo controller need stable thermo to keep offset near to zero.
A piece of polyourethane around the Q1 may be help
Hi Guys,
I have been listening for about a half hour. I hooked it up to my A/B setup so I could get a better feel for it. My impression so far is that is has its own sound. I compared it to a PeeCeeBee, an SKA GB150 and a DX Super A. It is milder than the PeeCeeBee and Super A but not as dark as the SKA. It kind of has a "fat" sound through the upper bass and mids. I kind of like it. Everything sounds good through it and without the A/B I may not have noticed the differences. I have them mounted on some smallish heatsinks, 3-1/4" X 7" X 1-1/4" but them seem to be fine. I'm listening at a descent level and they are comfortable to hold. I have th bias set with 40mV across R21. Q3 is pretty hot. I can just keep my finger on it. I'm not sure if that is because of the 44V rails or not. It's probably fine but if the layout is going to be adjusted at all I would leave room for a small heatsink.. As it is it is right up against the trimmer.
Thanks Christian for a cool little amp.
Blessings, Terry
I have been listening for about a half hour. I hooked it up to my A/B setup so I could get a better feel for it. My impression so far is that is has its own sound. I compared it to a PeeCeeBee, an SKA GB150 and a DX Super A. It is milder than the PeeCeeBee and Super A but not as dark as the SKA. It kind of has a "fat" sound through the upper bass and mids. I kind of like it. Everything sounds good through it and without the A/B I may not have noticed the differences. I have them mounted on some smallish heatsinks, 3-1/4" X 7" X 1-1/4" but them seem to be fine. I'm listening at a descent level and they are comfortable to hold. I have th bias set with 40mV across R21. Q3 is pretty hot. I can just keep my finger on it. I'm not sure if that is because of the 44V rails or not. It's probably fine but if the layout is going to be adjusted at all I would leave room for a small heatsink.. As it is it is right up against the trimmer.
Thanks Christian for a cool little amp.
Blessings, Terry
Attachments
Thank you Terry and Thimios for tests
You use 2200uF in C4 ?
My PCB fits a heat sink in Q3 .
Best Regards
You use 2200uF in C4 ?
My PCB fits a heat sink in Q3 .
Best Regards
Glad to see another tested amplifier soon
Yes feedback cap value is 2200uf
Never i have noticed overheated Q3
Last edited:
I try to make a habit of touching all the transistors of a new amp. That is how I noticed it. It is not hot enough to smell hot. Like I said, it might be because of the 44V rails. I don't have a lower voltage PSU right now to test it with.
Hi Terry
With the simple bootstrap load, the current through Q3 is a function of rail voltage, and at 44V I'm not surprised that its running a little hot! 35V is about the most I would be comfortable without some form of heatsink, and on my own layout I've left room around the VAS transistor for a small U shaped clip on type.
I also like to keep Q1 well away from it and other heat sources, otherwise I find that the offset tends to wander as the amp warms up. Of course one should trim once warmed up, but then the offset will be off on cold start up. Its no big deal but I mention it in case people become concerned with perhaps 100mV offset across the operating range.
Thanks for the listening report 🙂
With the simple bootstrap load, the current through Q3 is a function of rail voltage, and at 44V I'm not surprised that its running a little hot! 35V is about the most I would be comfortable without some form of heatsink, and on my own layout I've left room around the VAS transistor for a small U shaped clip on type.
I also like to keep Q1 well away from it and other heat sources, otherwise I find that the offset tends to wander as the amp warms up. Of course one should trim once warmed up, but then the offset will be off on cold start up. Its no big deal but I mention it in case people become concerned with perhaps 100mV offset across the operating range.
Thanks for the listening report 🙂
Hi Christian,
That makes sense. If I ever decide to build a case for this I will be sure to use a lower rail voltage.😉
That makes sense. If I ever decide to build a case for this I will be sure to use a lower rail voltage.😉
Thank you Terry, Thimios,
We are all getting a ringside view of this amp and its sound qualities!
With 44V rails, Q1 has dissipation of 28mW, and runs cool. Q2 runs at 60mW, again well under the benchmark 100mW for a TO92. But Q3 runs at (44-4)/(2.2+1.2)x(44-4)= 432mW and that is HOT for a TO126. A small 32C/W flag heatsink would be good there to keep its cool. Benchmark for a TO125 I find is 350mW, higher than that and you need a heatsink. Warming transistors make the offset wobble......
Terry, no-one has more experience with amp sound than you (or speed of build BTW) so if you like it, and its 'fat' sound appeals, I'd reckon this must be a very good amp!
Because of the blurring of the crossover issue that bedevils complementary SS AB amps, and because of the quasi topology and the tweaks that Ranchu has implemented, I would think this amp would be very good for listening at low level through voice. It would be a really good amp for listening to the radio, in fact, maybe through your iPod. It should be very good for MP3/DAB music too, fattening out to remove the digititis.
The 3.3V zener is well replaced by a blue led since their tempco is close to zero. Nominally they are 2.6V, but some are more with nearly 11mA through them! I see it brightens up the heatsink! I like a few leds on my amps, it adds a bit of colour, and you feel kindly about an amp with a shining led on it.
Great work, thank you.
Ciao,
Hugh
We are all getting a ringside view of this amp and its sound qualities!
With 44V rails, Q1 has dissipation of 28mW, and runs cool. Q2 runs at 60mW, again well under the benchmark 100mW for a TO92. But Q3 runs at (44-4)/(2.2+1.2)x(44-4)= 432mW and that is HOT for a TO126. A small 32C/W flag heatsink would be good there to keep its cool. Benchmark for a TO125 I find is 350mW, higher than that and you need a heatsink. Warming transistors make the offset wobble......
Terry, no-one has more experience with amp sound than you (or speed of build BTW) so if you like it, and its 'fat' sound appeals, I'd reckon this must be a very good amp!
Because of the blurring of the crossover issue that bedevils complementary SS AB amps, and because of the quasi topology and the tweaks that Ranchu has implemented, I would think this amp would be very good for listening at low level through voice. It would be a really good amp for listening to the radio, in fact, maybe through your iPod. It should be very good for MP3/DAB music too, fattening out to remove the digititis.
The 3.3V zener is well replaced by a blue led since their tempco is close to zero. Nominally they are 2.6V, but some are more with nearly 11mA through them! I see it brightens up the heatsink! I like a few leds on my amps, it adds a bit of colour, and you feel kindly about an amp with a shining led on it.
Great work, thank you.
Ciao,
Hugh
Last edited:
Hi Hugh,
Thanks for doing the math. It backs up what I saw. I agree about the low volume sound. It is very good. I also listened to some Ella and some Gerry Mulligan and it is very pleasing. Folks who find accurate amps sterile will like this amp.
Thanks for doing the math. It backs up what I saw. I agree about the low volume sound. It is very good. I also listened to some Ella and some Gerry Mulligan and it is very pleasing. Folks who find accurate amps sterile will like this amp.
Bax resistor is now 120R according to your suggestion.
Temperature on Q5=48 C(+/-32V ).
Attachments
Last edited:
Wow, Thimios, maybe it should be 150?
Then the resistor would pass 3.67 mA, and then the diode would pass 4.2+1.2-3.67 = 1.73mA and it would not turn on until the output voltage rises over 2.5V, well over the critical crossover area, since at this point an 8R speaker would be passing at least 312mA, where the switching would never be heard. Furthermore, with +20dB output, 14.14Vp into 8R, the mosfet NEVER turns off; it gets down to 3.6mA, staying on. At higher outputs, the mosfet does turn off, typically around 22Vp output (30W into 8R) so this situation prevails for quite a bit of higher output.
This is depending on 105mA quiescent according to LTSpice, so this is a simulation. I'd need to have your oscillograms to be absolutely sure of the reality of course. But it should be close......
Congratulations, very nice result!
Cheers,
Hugh
Then the resistor would pass 3.67 mA, and then the diode would pass 4.2+1.2-3.67 = 1.73mA and it would not turn on until the output voltage rises over 2.5V, well over the critical crossover area, since at this point an 8R speaker would be passing at least 312mA, where the switching would never be heard. Furthermore, with +20dB output, 14.14Vp into 8R, the mosfet NEVER turns off; it gets down to 3.6mA, staying on. At higher outputs, the mosfet does turn off, typically around 22Vp output (30W into 8R) so this situation prevails for quite a bit of higher output.
This is depending on 105mA quiescent according to LTSpice, so this is a simulation. I'd need to have your oscillograms to be absolutely sure of the reality of course. But it should be close......
Congratulations, very nice result!
Cheers,
Hugh
Thank you Thimios!
I'm about to birth a brand new SAKSA 85. It's been running two days on my bench, and I'm finishing the input/output wiring. I will take some piccies of the amp very soon for people interested. It is based on three of my commercial amps, and goes back to 2000 with the AKSA input stage.
I look forward to your 150R Bax resistor, Thimios..... who knows, I might build one of these too, they are wonderful at low output....
Ciao,
Hugh
I'm about to birth a brand new SAKSA 85. It's been running two days on my bench, and I'm finishing the input/output wiring. I will take some piccies of the amp very soon for people interested. It is based on three of my commercial amps, and goes back to 2000 with the AKSA input stage.
I look forward to your 150R Bax resistor, Thimios..... who knows, I might build one of these too, they are wonderful at low output....
Ciao,
Hugh
Unfortunately this 150R Bax resistor solve one problem but a new one created.Welcome to instability country!😱
Gee, this is very odd....... I will go back to my LTSpice, I suspect we need to put in 47p across C to B on the phase inverter..... Damn.........
One down, another up. I got my SAKSA 85 going today. Absolutely wonderful, I'm thrilled with the bass, PRAT, and engagement. This is a fabulous amp.
Hugh
One down, another up. I got my SAKSA 85 going today. Absolutely wonderful, I'm thrilled with the bass, PRAT, and engagement. This is a fabulous amp.
Hugh
Attachments
No problem, i will try to add this capacitor again in circuit soon.
This new amplifier looks nice!