How is a bias sense transistor chosen. I sometimes see TO-126 type devices with medium hfe (100-300) With vce 120v to 180v
What are the parameters that are relevant?
Q2: In a conventional EF design, how much of the rail is the CE of the biasing transistor exposed to ( eg the Vas, Drivers and OP may see the full rail to rail voltages, is this true for for the biasing bjt too?)
I ask because I am playing with a kit and at lower rails, (+/-60v) the 2sd669 biasing transistor has no issues, but as I raise the rail, it blows along with the OP and drivers. The rails are 85vdc each in this scenario, well within the vce of the drivers and the OP ( 180 & 230 respectively) but over the 160vce off the 2sd669.
Thoughts? Should I replace it? I do not have the schematics of this kit, and I know I am asking a lot
What are the parameters that are relevant?
Q2: In a conventional EF design, how much of the rail is the CE of the biasing transistor exposed to ( eg the Vas, Drivers and OP may see the full rail to rail voltages, is this true for for the biasing bjt too?)
I ask because I am playing with a kit and at lower rails, (+/-60v) the 2sd669 biasing transistor has no issues, but as I raise the rail, it blows along with the OP and drivers. The rails are 85vdc each in this scenario, well within the vce of the drivers and the OP ( 180 & 230 respectively) but over the 160vce off the 2sd669.
Thoughts? Should I replace it? I do not have the schematics of this kit, and I know I am asking a lot
Last edited:
A main reason to select a TO126 (TO220) device is because it can be mounted easely alongside the output bjt's: quick temperature response.
When things go wrong, this 2SD669 1W casing is fried with (1W / ( 2*85V) = less then ) 6mA.
Backwards engineering is always a nice puzzle, so post clear pictures.
Besides, why raising the rails???
When things go wrong, this 2SD669 1W casing is fried with (1W / ( 2*85V) = less then ) 6mA.
Backwards engineering is always a nice puzzle, so post clear pictures.
Besides, why raising the rails???
The bias transistor probably blew up because the OP/driver devices did, not the other way round. Early designs often only used BC107's as a bias device. Only needs to work at a few volts and (typically back then) 6mA. Insulating the device from the output heatsink was always a challenge. Plastic encapsulated devices with a single mounting hole (or fully insulated+ clip) makes mounting on the heatsink easier.
Modern designs usually bump the VAS current up a little, so a higher current device would be preferable.
Modern designs usually bump the VAS current up a little, so a higher current device would be preferable.
Higher gain is helpful but not essential. I've generally used lowish values of base bias resistors so that minimum gains of 100 work and that has been fine, in the main, even using non-optimised BD139's (the only reason for using BD139's was that I had several due to bulk ordering, not because 80V is needed in the bias reg).
My designs have used at least 10mA VAS often 15-25mA and with low value bias resistors shunting 1/10 the current work fine. Devices with low "quasi saturation" are preferable, and the BD139 is quite suitable at these currents. Before using BD139, I maybe would have used a BC337 glued on.
My designs have used at least 10mA VAS often 15-25mA and with low value bias resistors shunting 1/10 the current work fine. Devices with low "quasi saturation" are preferable, and the BD139 is quite suitable at these currents. Before using BD139, I maybe would have used a BC337 glued on.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.