pooge said:
Hi Pooge
I have just been looking through the Thermal Dynamics chapter of Audio Power Amplifier Design Handbook (Fifth edition under way, that's why) but I don't think there' s anything about optimal, or any, delay. Are you thinking of the bit about insulating the sensor from the ambient air to minimise cooling? If you can give me a page number I'll see what I can tell you.
Bob Cordell said:
Hi Bob, good to meet you, so to speak.
Thanks for the prompt reply. I have just been looking at the Locanthi circuit, kindly linked by Andy; I don't follow Locanthi's argument that a triple Darlington is more stable. I've always thought that the fewer Vbe drops you have to worry about, the better. I've had no problems using the standard EF (double-Darlington) output stage up to 200W/8R but that may be about as far as you want to go as the VAS gets rather hot.
The feedback control bias spreader is a very interesting idea, and as soon as I have finished a couple of projects I'll dust off PSpice and have a look at it.
Cheers,
Douglas
DouglasSelf said:
Hi Doug,
I agree with you; in the context of thermal bias stability, the T circuit does not appear to be more stable than a Double, and potentially less so. The fewer the number of Vbe drops, the better.
I think that for some reason he was thinking in terms of dc stability as a result of the Triple placing a much lighter load on the VAS in comparison to the output impedance of his VAS. Nevertheless, a bit of a murky argument in light of modern amplifier architectures.
It is a nice aspect of the T Triple that the predriver emitter followers can but run at a relatively smaller current, and in turn the VAS can often be run at a smaller current than would be feasible with only a Double as the output stage.
Cheers,
Bob
Bob Cordell said:
Is it not correct to say that the impedance seen looking in the emitter will be much lower with a triple. Therefore, at optimal bias, the non linear impedance variation caused by crossover will be less in magnitude with respect to the impedance of the load with a triple.
It is like a voltage source with a weakly non linear source impedance. If the source impedance variation is smaller with respect to the load, the influence is lower by voltage dividing action.
Because the VAS output impedance is large, the triple does a better work than a double in this respect.
Is this correct?
JPV
JPV said:
While we have to be careful about generalizations, I think you are largely correct.
If the VAS, with Miller effect taken into account, had an output impedance of 5000 ohms at a given frequency, then the open-loop output impedance of a Double with total current gain of 5000 would be on the order of 5 ohms plus the value of one emitter ballast resistor.
If the same VAS is employed with a Triple with current gain of 500,000, then the open-loop output impedance will be on the order of 0.05 ohms plus the value of one emitter ballast resistor.
In other words, with a Triple, the output transistor feels more like it is being driven by a voltage source than with a Double.
YMMV. A well-designed VAS with local Miller compensation can have a quite low output impedance due to the shunt feedback effect of the Miller compensation, especially at higher frequencies.
Cheers,
Bob
DouglasSelf said:
In the fourth edition of your book, on pages 356-357 discussing the simulation results of Figure 13.11, you showed how placing a sensor on top of a TO3 overcompensates, and stated that some sort of insulating material rather than a thermal pad is required for near-zero long-term error.
pooge said:
Ah, I'm with you now. The purpose of the semi-insulating material is to reduce the sensor temperature rise to prevent over-compensation, not introduce a time delay. It may introduce a small delay because of the thermal inertia of the sensor, but that is just an unwanted side effect.
I have to say I never pursued that idea any further.
DouglasSelf said:
This discussion may get into arguing over semantics since I look at the insulator as reducing the speed of the temperature rise on the temperature sensor, or integrating the peak to spread the rise over a greater time. But I haven't studied it in a while.
In any event, you used in your simulation model a TO3 can as an example. I don't have the value of the insulator at my finger tips right now, but I found it interesting that the value of heat conductivity you used for the insulator in your model is very very close to that of the top side of an TO 264 transistor, so that bolting a temperature sensor to the top of the TO264 would appear to mimic your "ideal" model very closely.
DouglasSelf said:
You could also do it by reducing the temperature coefficient of the Vbe multiplier by inserting a diode (at non-heatsink temperature) or similar component in the collector side of the base voltage divider if you want. This lowers the multiplication needed for a certain bias voltage which also lowers the temperature coefficient.
ThermalTrak again
Hi Bob,
This tread was very interesting but it is dead now.
Your ThermalTrak bias circuit, in the begining of the tread, is really good one and I am going to construct an amp as I have got some pairs of njl1302/3281.
Do you have any news about TT bias stability and how good was circuit diagram you proposed. Do I have to change something to get better thermal stability?
Any help will be appreciated, and thank you in advance.
dado
Hi Bob,
This tread was very interesting but it is dead now.
Your ThermalTrak bias circuit, in the begining of the tread, is really good one and I am going to construct an amp as I have got some pairs of njl1302/3281.
Do you have any news about TT bias stability and how good was circuit diagram you proposed. Do I have to change something to get better thermal stability?
Any help will be appreciated, and thank you in advance.
dado
Hi dadod,
Yes, this was a good thread, and my participation in it helped me formulate some of my thinking for the ThermalTrak section of the thermal stability chapter in my book. I have learned quite a bit more about ThermalTrak biasing while writing my book, so I recommend you rely more heavily on what I wrote there.
One approach that I have liked, and which is not too complex, is to have one Vbe multiplier that uses the ThermalTrack diodes to manage the bias of the output transistors, and to have a second Vbe multiplier in series to manage the temperature compensation of all of the other transistors in the Locanthi output triple (i.e., the pre-drivers and drivers). I then mount the second Vbe multiplier on a largely isothermal bar to which the pre-drivers and drivers are also mounted. By keeping all of these transitors at essentially the same temperature, and recognizing that their power dissipation is not much of a function of program material, good thermal stability of the Triple can be achieved. Independent adjustment of the thermal multiplication coefficients of the two Vbe multipliers provides freedom in achieving good overall thermal tracking. After initial design centering, only one of the Vbe multipliers need have a pot for adjusting the bias of the amplifier.
Cheers,
Bob
Hi Bob,
Sorry to bother you, just some morre questions here.
<to have one Vbe multiplier that uses Tthe RhermalTrak diodes to manage the bias of the output transistors>
How to connect two TTdiodes (I suppose two are enough), in series with the Vbe multiplier or into base - collector connection of the Vbe multiplier?
Does this multiplier has to be mounted on the same heathsink as the output TT transistors?
dado
Sorry to bother you, just some morre questions here.
<to have one Vbe multiplier that uses Tthe RhermalTrak diodes to manage the bias of the output transistors>
How to connect two TTdiodes (I suppose two are enough), in series with the Vbe multiplier or into base - collector connection of the Vbe multiplier?
Does this multiplier has to be mounted on the same heathsink as the output TT transistors?
dado
Thermal attenuation
Hi Bob,
Regarding thermal attenuation, would you please have a look at this post: http://www.diyaudio.com/forums/solid-state/174218-rebirth-phoenix-5.html#post2353939
Am I right that we have to take this matter into account and make some provisions to compensate for thermal resistances by means of some (say 20%) electrical overcompensation?
Cheers,
E.
PS: Great book. Congratulations!
Hi Bob,
Regarding thermal attenuation, would you please have a look at this post: http://www.diyaudio.com/forums/solid-state/174218-rebirth-phoenix-5.html#post2353939
Am I right that we have to take this matter into account and make some provisions to compensate for thermal resistances by means of some (say 20%) electrical overcompensation?
Cheers,
E.
PS: Great book. Congratulations!
I have read your suggestion of using the left over TT diodes to detect the temp of the junction and protect the transistor before reaching let say 150°C. If we have such a detection coupled with a maximum current detector we are save with respect to short circuit protetcion.
The question remaining is secondary breakdown. Do you believe that monitoring the power max temp is enough. Secondary breakdown will happen with some reactive loads but it has been shown elsewhere that derating for temp during secondary breakdown should less stringeant than the derating of power dissipation because the effect of increase of emitter resistance with temperature. In other words, with increase in temperature, the normal power limit takes over the secondary breakdown limit for higher Vce than at lower temperature. If you couple this with the 10 ms SOA limit, is it not reasonnable to say that monitoring the 150°C crossing is totally safe for normall ( less than 45° phase) loads.
Then a totally non intrusive SOA protection device can be designed.
What do you think. The secondary breakdown is the problem.
JPV
The question remaining is secondary breakdown. Do you believe that monitoring the power max temp is enough. Secondary breakdown will happen with some reactive loads but it has been shown elsewhere that derating for temp during secondary breakdown should less stringeant than the derating of power dissipation because the effect of increase of emitter resistance with temperature. In other words, with increase in temperature, the normal power limit takes over the secondary breakdown limit for higher Vce than at lower temperature. If you couple this with the 10 ms SOA limit, is it not reasonnable to say that monitoring the 150°C crossing is totally safe for normall ( less than 45° phase) loads.
Then a totally non intrusive SOA protection device can be designed.
What do you think. The secondary breakdown is the problem.
JPV
Hi Edmond,
I'm sorry I'm late getting back to this post - I just lost track of this thread.
You are right about thermal attenuation within the ThermalTrak devices between the transistor junction and the tracking diode. This is certainly non-ideal, but the ThermalTrak arrangement is still far superior to the conventional arrangement, which suffers even more thermal attenuationa and far more thermal delay.
In general, some experimentation is needed in setting up the optimum compensation for a ThermalTrak amplifier. Some form of electrical multiplication usually has to take place in the Vbe multiplier associated with the ThermalTrak devices. Indeed, some multiplication is often needed anyway because the temperature coefficient of the ThermalTrak diodes is not quite the same as that of the BJT. So the needed amount of multiplication is all wrapped together the best we can.
In many cases I'll use two Vbe multipliers, one working with and incorporating the ThermalTrak tracking diodes and the other providing bias spreading for the remaining devices of the output Triple. That way, the amount of multiplication allocated to each process can be set somewhat independently. The transistor of the second Vbe multiplier usually wants to track the temperature of the pre-driver and driver transistors. Indeed, if the driver transistors happen to be mounted on the heat sink, then the transistor of the second Vbe multiplier may also be mounted on the heatsink.
It was a two Vbe multiplier arrangement like this that I used for the test amplifier whose bias behavior is shown in Figure 14.23 in the book. In that case, the second Vbe multiplier along with the pre-drivers and drivers were mounted on an isothermal bar/heatsink on the circuit board.
Cheers,
Bob
I obtained some samples of the NJL4281 and NJL4302 a few years ago and used them to make a stereo amp, and it's been working fine ever since.- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.