This Circuit Puzzles me?

[Mikebits]

Below is a circuit for a voltage reference using an LTZ1000, what puzzles me is the transistor 2N3904, the heater, and op amp A1.
I assume the 3904 is a current source for the heater, but I am puzzled to what is the feedback path for Op amp A1? The .1uF cap blocks DC feedback and since this is a DC circuit I am at a lost how A1 and its feedback path work. What I am saying is, I do not understand this circuit and its operation. Where the heck is the feedback path for A1?
Thanks

 

[ericgibbs]

Morning Mike,
If you consider that A1 is being used as comparator to switch the load On/Off, the feed back required for that is derived fromA2 and Vbz & R4
Eric

 

[Mikebits]

I thought A1 was being used in conjunction with the 2N3904 as a current source for the heater. This circuit really seems awkward to me. I have studied to length yet I find its operation eludes me.
I will have to revisit A1 as a comparator, This might make more sense.
Thanks Master Eric

 

[Tony Stewart]

This explains it better with the whole circuit better than I can at the moment. The NPN pair act as a Diff Amp to the collector current with zener bias and forcing voltage. Note star grounds.

https://www.eevblog.com/forum/begin...n-voltage-source-best-way-to-trim-a-ltz1000a/

https://www.eevblog.com/forum/metrology/ultra-precision-reference-ltz1000/

It is a long read, so pack a lunch.

 

[Mikebits]

Morning Mike,
If you consider that A1 is being used as comparator to switch the load On/Off, the feed back required for that is derived fromA2 and Vbz & R4
Eric

Okay doak, I think I got it. When LTZ1000 first powers up, the NPN internal transistor is off due to the bias voltage of .5vdc. With NPN off A1 output goes high thus turning on 2N3904 allowing the heater to operate.
As the internal temp rises, Q2 turns on which causes A1 output to go low, this in turn, turns off heater.

Does that sound about right?
Thanks

 

[AnalogKid]

Note that A1 is an integrator with a time constant of 0.1 s. Audio power amplifiers that use a "DC servo" to eliminate coupling capacitors have a similar circuit. As a comparator with a very slow slew rate, it is the slowest circuit in the loop. This prevents oscillations.

ak

 

[Mikebits]

I would still like to know, is my understanding of the above circuit correct? I would just like to know if I have not misunderstood the circuit mentioned above. Like I have said, the circuit had me puzzled so just wondering if I had figured it out or not.
Thanks

 

[Mikebits]

Note that A1 is an integrator with a time constant of 0.1 s. Audio power amplifiers that use a "DC servo" to eliminate coupling capacitors have a similar circuit. As a comparator with a very slow slew rate, it is the slowest circuit in the loop. This prevents oscillations.

ak

I am slightly unclear about your post. The circuit posted above is not an audio circuit, I guess your talking about the cap feedback in the op-amp. I must admit this had me puzzled as well. I just assumed it was to prevent oscillations.

 

[AnalogKid]

I cited "high-end" audio amplifiers as an example of a place where this type circuit is used. You are correct, the cap is there to prevent oscillations, but not oscillations in that opamp stage; oscillations in the overall control loop.

ak