Here is the circuit to change an sx-amp from class A to class AB.
To prevent confusion, I have called this one the 'ux-Amplifier'
1. It uses the same component numbering as the sx-amp
2. The change to convert to class AB is to move R21 top connection from the emitter of Q5 and connect it instead to the base of Q5
3. Further, I increased the value of R21 from 1k to 1.5k - this should allow a decent adjustment range - but if not, you may have to experiment with the value of R21 a bit.
4. The supply rails can be increased to 35V which will give 65W output in class AB.
5. I recommend you stick with two pairs of output transistors - there are no drivers in this design, so we need to exploit the high hFE at low to medium Ic of the NJW3281/1302 pairs - two pairs keeps the Ic operating range of the output pairs narrower than one pair . If you reduce the supply rails to +-20 to +-25, then you can remove one pair. The output power in this case will be 20~30W in class AB
6. Don't forget to link out J2
7. The fuses are changes to 3A slow blow types.
To prevent confusion, I have called this one the 'ux-Amplifier'
1. It uses the same component numbering as the sx-amp
2. The change to convert to class AB is to move R21 top connection from the emitter of Q5 and connect it instead to the base of Q5
3. Further, I increased the value of R21 from 1k to 1.5k - this should allow a decent adjustment range - but if not, you may have to experiment with the value of R21 a bit.
4. The supply rails can be increased to 35V which will give 65W output in class AB.
5. I recommend you stick with two pairs of output transistors - there are no drivers in this design, so we need to exploit the high hFE at low to medium Ic of the NJW3281/1302 pairs - two pairs keeps the Ic operating range of the output pairs narrower than one pair . If you reduce the supply rails to +-20 to +-25, then you can remove one pair. The output power in this case will be 20~30W in class AB
6. Don't forget to link out J2
7. The fuses are changes to 3A slow blow types.
Attachments
Last edited:
ux amplifier
Ok Bonsai thanks for all this effort.🙂
I wish good luck for this ux amplifier!
Best regards.
Thimios.
Ok Bonsai thanks for all this effort.🙂
I wish good luck for this ux amplifier!
Best regards.
Thimios.
Last edited:
mr bonsai.are there any different quality of sound betwen UX amp and NX amp? .because my next project after SX amp( ready for adjusment & listening test ,of course ill post pict / result next time) is NX AMP.
bias transistor position
Hi Bonsai ,i want to ask about bias transistor position for class AB .
Is it necessary putting this in close contact with radiator or no?
Hi Bonsai ,i want to ask about bias transistor position for class AB .
Is it necessary putting this in close contact with radiator or no?
This is probably a silly question but what is the reason for the ux when you already have the nx and it is a proven performer? Is there something about the ux that is an improvement?
2)I have 2pairs NJW0281G,0302G are these ok for output?
3)I haven't any more same hfe drivers.My collection is only ksa1381D AND KSC3503E.Can i use these?
4)I haven't 0.33R/5W but only 0.22R/5W non inductive.Are these ok ?
Regards.
Thimios.
Attachments
Last edited:
NJW are good for output duty.
2sa1381 are more suited to VAS and pre-driver duty. Did you compare the datasheets?
You could temporarily connect a 0r1 2W and the 0r22 5W to create a 0r32 6.4W as emitter resistor, until your order arrives.
2sa1381 are more suited to VAS and pre-driver duty. Did you compare the datasheets?
You could temporarily connect a 0r1 2W and the 0r22 5W to create a 0r32 6.4W as emitter resistor, until your order arrives.
Last edited:
Thanks for answer ,yes i mean ksa 1381,3503 for using in VAS .
I ask because my ksa is on D and ksc is on E hfe category
Very neat Thimios! I am looking forward to your finished amplifier!
Do you have good heatsinks?
Are those 0.5 W resistors for the feedback network?
Ideally you want the VAS transistor hFE grades to be the same. But, if they are different, it won't stop your amp from working.
Do you have good heatsinks?
Are those 0.5 W resistors for the feedback network?
Ideally you want the VAS transistor hFE grades to be the same. But, if they are different, it won't stop your amp from working.
Last edited:
Second channel working
2)0.5W resistors?Look i have 0.22R /5W non inductive resistors instead of 0.33R/5W (Bias resistors).
BTW This second channel is alive! Just for test with a ridiculous heatsink up to now.🙁
I love the bass of this amplifier (C.F.A with deep&strong bass???).
Best Regards.
Thimios.
1)unfortunately i haven't good heatsinks.🙁
2)0.5W resistors?Look i have 0.22R /5W non inductive resistors instead of 0.33R/5W (Bias resistors).
BTW This second channel is alive! Just for test with a ridiculous heatsink up to now.🙁
I love the bass of this amplifier (C.F.A with deep&strong bass???).
Best Regards.
Thimios.
Attachments
The resistors I am referring to are the feedback resistors (1.5k) - they should dB 0.5,W because at full power they dissipate about 2W.
You can probably use the 0.22 ohms for the emitter degen resistors - as long as you can adjust the bias to 700 mA per oak it should work ok. Because of the OPS bias arrangement, there is no thermal runaway on this amp.
Anyway, very glad you have the first channel running and you like the sound!
The heatsink temperature should not go above 65 Celsius.
You can probably use the 0.22 ohms for the emitter degen resistors - as long as you can adjust the bias to 700 mA per oak it should work ok. Because of the OPS bias arrangement, there is no thermal runaway on this amp.
Anyway, very glad you have the first channel running and you like the sound!
The heatsink temperature should not go above 65 Celsius.
Bonsai, first, your amplifier designs are magnificent from my point of view and you`re doing very good job, congratulations!
Second, one question: have you noticed any differences in sound (in some of your designs in class AB) with over-biased bjt output stages because of gm doubling? I hear a little darker sound when over-bias bjts from "optimal 26mV" for class AB.
Second, one question: have you noticed any differences in sound (in some of your designs in class AB) with over-biased bjt output stages because of gm doubling? I hear a little darker sound when over-bias bjts from "optimal 26mV" for class AB.
Thank you for your kind comments Bogdan - I've enjoyed looking at your designs as well.
Yes, I do over bias my output stages - I've done this on the nx-Amp, and also on the Ovation 250. The e-Amp is set at the optimum bias - but I can easily over bias it if I want.
I think the sound with an over biased OPS is warmer and I prefer it. I did comment somewhere that the Ocation 250 amp has a very 'dark' sound which I like very much.
Note however that the distortion specs I quote are done at optimum bias settings. The over biased distortion is a bit higher.
Yes, I do over bias my output stages - I've done this on the nx-Amp, and also on the Ovation 250. The e-Amp is set at the optimum bias - but I can easily over bias it if I want.
I think the sound with an over biased OPS is warmer and I prefer it. I did comment somewhere that the Ocation 250 amp has a very 'dark' sound which I like very much.
Note however that the distortion specs I quote are done at optimum bias settings. The over biased distortion is a bit higher.