ITT726 was this a good idea?

Thread Starter

hp1729

Joined Nov 23, 2015
2,304
ITT726 a.PNG ITT726 b.PNG A temperature controlled pair of NPN transistors. Another old chip.
Was this a good idea to build in a heater? It seems any environment that would add heat to the circuit would still add heat to the circuit, more than the heater added? I have seen this built-in heater on voltage references (LM399, LM3999) and thought it was a good idea, but now ... ?

Attached is a page or two of info and an example from the data sheet (to big to include here).

so what do y'all think? Good idea or bad idea?

Some data sheets might just show the two transistors. Added to this is a circuit that acts as a heater. Temperature is controlled by an external resistor. No specs are provided that I could see for how much resistance is suggested for what heat..
The example provided is straight from the data sheet and uses a 709 op amp for a gain of 1,000. Certainly any other op amp could probably be used and frequency limit isn't mentioned. You can't get much out of a 709. It is a forerunner of the 741.
 

Attachments

OBW0549

Joined Mar 2, 2015
3,566
Was this a good idea to build in a heater? It seems any environment that would add heat to the circuit would still add heat to the circuit, more than the heater added? I have seen this built-in heater on voltage references (LM399, LM3999) and thought it was a good idea, but now ... ?
I think in both cases, the "heater" is actually a regulator which stabilizes the die temperature against ambient variations.

A temperature-stabilized transistor pair was probably a good idea at the time (early 1970's, judging from the uA709-based application circuit), as "supermatched" duals such as the LM394 and related matched pairs hadn't yet been developed. Nowadays even the LM394 is obsolete, since many op amps are available with extremely low Vos drift.

Good idea at the time, but no longer needed.
 

Thread Starter

hp1729

Joined Nov 23, 2015
2,304
I think in both cases, the "heater" is actually a regulator which stabilizes the die temperature against ambient variations.

A temperature-stabilized transistor pair was probably a good idea at the time (early 1970's, judging from the uA709-based application circuit), as "supermatched" duals such as the LM394 and related matched pairs hadn't yet been developed. Nowadays even the LM394 is obsolete, since many op amps are available with extremely low Vos drift.

Good idea at the time, but no longer needed.
Thanks.
I see. consistent temperature was the objective. As the environment heated and cooled the built in heater sensed that and compensated.
 
Last edited:
Top