how much quiescent current is pretty safe to run in class AB at +/- 53V psu?
i tried this with 2sc5200 and found that anything beyond 300ma its getting pretty hot even a big heatsink is getting hot. Calculated as the power dissipation is as much as 18w per transistor so total 36w is being dissipated for push pull with Ipeak as 600ma and for being in class A its 2.8W
so calculating how much quiescent current is safe per transistor or IRFP240?
is it recommended to reduce the psu voltage to 45v or reduce it to 30v and use the amp in class A mode with higher bias as 2A Ipeak and get 24w of class A may be we require two transistor for n and two for p to be more safe each sharing 500ma of Quiescent current but If we can know what is the safe current value..
i tried this with 2sc5200 and found that anything beyond 300ma its getting pretty hot even a big heatsink is getting hot. Calculated as the power dissipation is as much as 18w per transistor so total 36w is being dissipated for push pull with Ipeak as 600ma and for being in class A its 2.8W
so calculating how much quiescent current is safe per transistor or IRFP240?
is it recommended to reduce the psu voltage to 45v or reduce it to 30v and use the amp in class A mode with higher bias as 2A Ipeak and get 24w of class A may be we require two transistor for n and two for p to be more safe each sharing 500ma of Quiescent current but If we can know what is the safe current value..
100 to 200 mA is usual for those MOSFETs for Class AB.
For BJT like 2SC5200 maybe 30-50 mA.
The bad thing people do is to have too much current.
They think more is good.
But this is in fact not at all true.
Keep bias current at a restricted level for Class AB.
(We are not talking Class A now)
For BJT like 2SC5200 maybe 30-50 mA.
The bad thing people do is to have too much current.
They think more is good.
But this is in fact not at all true.
Keep bias current at a restricted level for Class AB.
(We are not talking Class A now)
BJT compelments have a very small "optimum bias" current that minimizes gm doubling distortion
MOSFETs don't
power MOSFET like lots of bias, linearity improves with increasing bias - there is no "optimum" for complementary MOSFET output stage bias - just what your power supply and heatsinking will bear
MOSFETs don't
power MOSFET like lots of bias, linearity improves with increasing bias - there is no "optimum" for complementary MOSFET output stage bias - just what your power supply and heatsinking will bear
In the hundreds of existing threads about output transistor pair bias, the typical bias level nominated for Vertical Mosfets in class AB is ~100mA per pair. There is no critical value, just a trade-off of linearity against dissipation but above a level of about 80mA, the benefits start to drop off to marginal levels. Since you have quoted Self's work in another thread, you may be aware that BJTs are not the same as Mosfets and in class AB, have a critical bias level that offsets the base-emitter junction voltage to give the minimum crossover distortion. Above or below that, we are just trading dissipation against rising distortion level for particular subjective sound qualities.
Maximum current ratings, whether in the steady DC bias of class A or in class AB signal peaks, are always limited by the SOA ratings of the individual devices. The idea is to read, understand and keep within the specifications and the Safe Operating Area diagram limits, so that the parts always work within their ratings and survive rather than burn up. Obviously, each part has different specifications and limits and sufficient heatsinking is also a part of the rating.
Download datasheets for the required parts from the vendors' and various archive sites and de-rate the current from the maximum permissible according to your expected temperature conditions. There will be a graph showing the de-rating characteristics. This will indicate the maximum allowable bias current without further safety margins.
Maximum current ratings, whether in the steady DC bias of class A or in class AB signal peaks, are always limited by the SOA ratings of the individual devices. The idea is to read, understand and keep within the specifications and the Safe Operating Area diagram limits, so that the parts always work within their ratings and survive rather than burn up. Obviously, each part has different specifications and limits and sufficient heatsinking is also a part of the rating.
Download datasheets for the required parts from the vendors' and various archive sites and de-rate the current from the maximum permissible according to your expected temperature conditions. There will be a graph showing the de-rating characteristics. This will indicate the maximum allowable bias current without further safety margins.
Last edited:
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.