Board,
I was wondering whether a high quality PTC thermistor with a switching temperature of 50-60c mounted to the heatsink and in series with a pot (along with the other bias resistor) could take the place of the usual small signal transistor in a SS power amp circuit? I'd gladly swap a resistor for a tranny if the functionality remains.
Has anyone tried this method, with or without success?
TIA,
M
I was wondering whether a high quality PTC thermistor with a switching temperature of 50-60c mounted to the heatsink and in series with a pot (along with the other bias resistor) could take the place of the usual small signal transistor in a SS power amp circuit? I'd gladly swap a resistor for a tranny if the functionality remains.
Has anyone tried this method, with or without success?
TIA,
M
Could. But the plot of the thermal coefficient curve may or may not be what you want. A (the ?) reason for using a transistor is the best approximation of a BJT's thermal response is another BJT. At least that's the theory.
Unless the thermistor makes for an equivalent or better approximation, it is probably an inferior approach.
Unless the thermistor makes for an equivalent or better approximation, it is probably an inferior approach.
It may be my memory playing tricks with me, but I think
thermistors were often used long ago, when transistors
were still considered expensive.
thermistors were often used long ago, when transistors
were still considered expensive.
To get the curve you want, consider a pot in series with the
Thermistor, and a pot in parallel. Also, be prepared to try
some different values of Thermistors.
Thermistor, and a pot in parallel. Also, be prepared to try
some different values of Thermistors.
"To get the curve you want, consider a pot in series with the
Thermistor, and a pot in parallel. Also, be prepared to try
some different values of Thermistors."
Which leaves one twiddling two pots. If this were the only way one knew about, a Vbe multipler sharing a heatsink with the output transistors would look like a stroke of genius.
Thermistor, and a pot in parallel. Also, be prepared to try
some different values of Thermistors."
Which leaves one twiddling two pots. If this were the only way one knew about, a Vbe multipler sharing a heatsink with the output transistors would look like a stroke of genius.
"I was wondering whether a high quality PTC thermistor with a switching temperature of 50-60c mounted to the heatsink and in series with a pot (along with the other bias resistor) could take the place of the usual small signal transistor in a SS power amp circuit? "
No, these cannot be used for a bias circuit.
The type Nelson Pass describe are linear NTC type.
The type you describe are for fan speed, over temp cut-out, etc.
"Which leaves one twiddling two pots. If this were the only way one knew about, a Vbe multipler sharing a heatsink with the output transistors would look like a stroke of genius"
Considering how many designers use a Vbe multiplier AND a thermistor too, I would say these designers know something you are not aware of.
See schematics of: Adcom, McIntosh, Nakamichi, QSC, Threshold, etc.
No, these cannot be used for a bias circuit.
The type Nelson Pass describe are linear NTC type.
The type you describe are for fan speed, over temp cut-out, etc.
"Which leaves one twiddling two pots. If this were the only way one knew about, a Vbe multipler sharing a heatsink with the output transistors would look like a stroke of genius"
Considering how many designers use a Vbe multiplier AND a thermistor too, I would say these designers know something you are not aware of.
See schematics of: Adcom, McIntosh, Nakamichi, QSC, Threshold, etc.
"Considering how many designers use a Vbe multiplier AND a thermistor too, I would say these designers know something you are not aware of."
I certainly hope so since it's their job, not just a hobby.
I certainly hope so since it's their job, not just a hobby.
Seems like the thermistor by itself wouldn't give the supply rejection that a Vbe circuit would. I'm planning to try a Vbe circuit for a relatively low Vgs complementary MOSFET output stage using a fairly high current medium voltage (5A, 100V) insulated tab TO220 NPN bipolar transistor mounted to the output stage heatsink but running with less than 10mA collector current - it looks like the Vbe curves vs temperature spread out quite a bit relative to the nominal threshold with this device at such a low current and preliminarily to actually trying it, it looks like it should provide plenty of thermal compensation although I'll know for sure when the electrons hit the silicon.
well i have seen an excellent design . for protection only.
The thermistor and series resistance is placed in parrallel with the collector to base main R2 resistor which is fixed value , and a fixed resistor is also placed in parallel with the R1+pot in series which are connected between base and emitter of Vbe.
Now if pot opens -----the parallel resistance keeps bias in check to avoid driver blow due to VBe saturation And if pot shorts the current increase through the thermistor raises its resistance to keep Ib low. its good idea for driver protection
The thermistor and series resistance is placed in parrallel with the collector to base main R2 resistor which is fixed value , and a fixed resistor is also placed in parallel with the R1+pot in series which are connected between base and emitter of Vbe.
Now if pot opens -----the parallel resistance keeps bias in check to avoid driver blow due to VBe saturation And if pot shorts the current increase through the thermistor raises its resistance to keep Ib low. its good idea for driver protection
I used an NTD thermistor to augment a conventional 2 transistor Vbe multiplier circuit in my triple output stage 'e-Amp' here http://hifisonix.com/wordpress/wp-content/uploads/2011/03/The_e-Amp_V2.03.pdf
Check out the details around about page 40 for how this works - I call it 2 point thermal compensation.
I have not tried using just a thermistor on its own, but can tell you that the two point compensation scheme works exceedingly well - very stable and repeatable and easy to set up.
Check out the details around about page 40 for how this works - I call it 2 point thermal compensation.
I have not tried using just a thermistor on its own, but can tell you that the two point compensation scheme works exceedingly well - very stable and repeatable and easy to set up.
- Status
- Not open for further replies.