I tweaked a few resistors in the tails of the input stage, can't remember what values I used, but it's all Ohms law to keep the currents through the diodes similar to the 35-40V version.
I removed the IV limiting in my version because of space constraints, but it depends on your usage as to if you use it or not.
I removed the IV limiting in my version because of space constraints, but it depends on your usage as to if you use it or not.
My KSA boards run on different voltages.
For the time being both rails are not regulated yet untill i have the voltage regulator boards from Bob Ellis's Groupy B worked out.
No V/I limiters as they have only 35Vdc on the output rails and the output devices have a tremendous SOA.(for a BJT)
Depending on the type of output device used and the number a V/I limiter is not necessary.
Unless you intend to parallel a huge number of output devices alla Threshold Stasis amps i'd use a V/I limiter on 60Vdc rails.
Which i do on the Leach amp boards i'm working on that has 10 output devices, supposed to run on 75 Vdc for the output stage rails.
The few extra volts on the front end rails are to enable to drive the output stage almost to the output rail voltage, without having Vbe losses.
(some will disagree and talk clipping chit-chat but plenty designers who have done this, including the master himself)
For the time being both rails are not regulated yet untill i have the voltage regulator boards from Bob Ellis's Groupy B worked out.
No V/I limiters as they have only 35Vdc on the output rails and the output devices have a tremendous SOA.(for a BJT)
Depending on the type of output device used and the number a V/I limiter is not necessary.
Unless you intend to parallel a huge number of output devices alla Threshold Stasis amps i'd use a V/I limiter on 60Vdc rails.
Which i do on the Leach amp boards i'm working on that has 10 output devices, supposed to run on 75 Vdc for the output stage rails.
The few extra volts on the front end rails are to enable to drive the output stage almost to the output rail voltage, without having Vbe losses.
(some will disagree and talk clipping chit-chat but plenty designers who have done this, including the master himself)
Hi,
I am planning 6pair output stage using twin drivers (a bit like KSA100) on +-42Vrails.
Not started the build.
You need to scale:-
r103/107
r108/109
Why change r112/116/113/117? They are the load for the LTP.
Keep the zeners near 24V to 30V. Arrange for a zener current of slightly more than 10% of it's maximum dissipation. Consider adding a ceramic cap across the Zener to hold constant voltage and attenuate noise.
If using 500mW zener @ 27V then Iz>=2mA.
LTP current ~=3mA
Total current through r108>=5mA. But you need to check this at both minimum mains supply voltage and at maximum mains supply voltage. Check maximum dissipation on r108/109.
I103/107 ~=0.8mA
You may want to reduce r125 slightly to ensure a good range on setting output bias. Particularly if you want a high Iq. r124~=2k0 and r125~=510 may be a good starting point.
R144 (wrongly labelled low bias) could be brought in/out by relay. Summer setting (r126 on it's own) = standby or sink overheat.
I am planning 6pair output stage using twin drivers (a bit like KSA100) on +-42Vrails.
Not started the build.
You need to scale:-
r103/107
r108/109
Why change r112/116/113/117? They are the load for the LTP.
Keep the zeners near 24V to 30V. Arrange for a zener current of slightly more than 10% of it's maximum dissipation. Consider adding a ceramic cap across the Zener to hold constant voltage and attenuate noise.
If using 500mW zener @ 27V then Iz>=2mA.
LTP current ~=3mA
Total current through r108>=5mA. But you need to check this at both minimum mains supply voltage and at maximum mains supply voltage. Check maximum dissipation on r108/109.
I103/107 ~=0.8mA
You may want to reduce r125 slightly to ensure a good range on setting output bias. Particularly if you want a high Iq. r124~=2k0 and r125~=510 may be a good starting point.
R144 (wrongly labelled low bias) could be brought in/out by relay. Summer setting (r126 on it's own) = standby or sink overheat.
Right, changing r112/116/113/117 was just brain dysfunction. After considering everything I think I'll use a lower voltage, around 40 - 45 volts. I wanted to use 60 simply because I already have it built from a previous project. Another option is to wire the 2 primaries in series and use 30 volts.I could get wild and bridge 2 amplifiers at this voltage for more power. might be an option. Thanks for the info on the other resistors, etc.
Hi,
bridging will defeat the ClassA function.
A bridged amp thinks it is driving half the load impedance.
KSA50 can only manage 1.9A bias and 3.8Apk. The 8ohm speakers will appear as 40hm loading. 4ohm loading will stay in ClassA upto 15Vpk (about 29W).
That equates to 58W into 8ohms. What a waste of two channels of amplification.
bridging will defeat the ClassA function.
A bridged amp thinks it is driving half the load impedance.
KSA50 can only manage 1.9A bias and 3.8Apk. The 8ohm speakers will appear as 40hm loading. 4ohm loading will stay in ClassA upto 15Vpk (about 29W).
That equates to 58W into 8ohms. What a waste of two channels of amplification.
A ClassA spec PSU is a different animal from a ClassAB PSU. They do not swap over very well.,
Hi Jacco,
I do not understand.
Are you saying that Accuphase were daft enough to offer a bridging option in a full ClassA amplifier
or
That Accuphase changed the bias to keep the bridged amp in full ClassA
or
That Accuphase conned the populace into believing that spending good money on a ClassA amplifier could increase the ClassA power by simply bridging?
Which option do you think is closest to the twisted truth.
Or
Is there something I have misunderstood about ClassA current?
I do not understand.
Are you saying that Accuphase were daft enough to offer a bridging option in a full ClassA amplifier
or
That Accuphase changed the bias to keep the bridged amp in full ClassA
or
That Accuphase conned the populace into believing that spending good money on a ClassA amplifier could increase the ClassA power by simply bridging?
Which option do you think is closest to the twisted truth.
Or
Is there something I have misunderstood about ClassA current?
Among several other manufacturers Accuphase produced fully symmetrical class A amplifiers.
For the output voltage an amplifier may see only half the impedance in bridged mode. The output current still only views the same loudspeaker.
An amplifier that does 50 watts class A in 8 ohms still will deliver 50 watts in class A in 8 Ohms when the channels are bridged. Bridging doesn't increase class A output level, nor does it decrease it.
A major difference besides different slew and damping numbers is the highly increased AB output power.
All in all, a bridged class A amplifier has several advantages.
For really high power applications it is often cheaper, there's no linearity between component voltage rating and component price.
The disadvantage is a somewhat lower efficiency for a bridged amp that has the same output rating as a non-bridged version with higher rail voltages.
A lot of the famous high power class A amplifiers only did a minor portion in class A.
Check the Luxman M-05 thread for instance. Very expensive amplifier, advertised as 105 watts in 8 ohms amplifier. 105 watts continuous in AB maybe, the bias was only good for 36 !
Why someone would bridge a class A amplifier that has 60Vdc rails illudes me, but that's his choice.
For the output voltage an amplifier may see only half the impedance in bridged mode. The output current still only views the same loudspeaker.
An amplifier that does 50 watts class A in 8 ohms still will deliver 50 watts in class A in 8 Ohms when the channels are bridged. Bridging doesn't increase class A output level, nor does it decrease it.
A major difference besides different slew and damping numbers is the highly increased AB output power.
All in all, a bridged class A amplifier has several advantages.
For really high power applications it is often cheaper, there's no linearity between component voltage rating and component price.
The disadvantage is a somewhat lower efficiency for a bridged amp that has the same output rating as a non-bridged version with higher rail voltages.
A lot of the famous high power class A amplifiers only did a minor portion in class A.
Check the Luxman M-05 thread for instance. Very expensive amplifier, advertised as 105 watts in 8 ohms amplifier. 105 watts continuous in AB maybe, the bias was only good for 36 !
Why someone would bridge a class A amplifier that has 60Vdc rails illudes me, but that's his choice.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.