op-amp info

Hi Ivan
I have wondered about the thermistors as well I think I will try some in another circuit. I think I have some that will be perfect.
I have looked on the net a lot for circuits like this and have never seen any that thermistors. There are also a lot using transistors basically as diodes as well. All of them seem to rise quickly in air flow but drop a little slowly so they would not work well if a person was interested in measuring wind gusts. 



I am always interested in different ways to control gain in op-amps. Thats an interesting circuit that I haven't seen but it uses an awful large amount of parts.  I use just one jfet and a resistor in series across another resistor.(one sets the low limit and one the high limit) There are cases where this doesn't work but most of the time it does. 

I have used the Jfet for switches and variable resistors many times but I don't like it for an amp.
I think where low base current exists the Darlington transistor pair is best. 

Lately I have been repairing organs from the 70's and 80's. I am not a musician but I can play well enough to at least test them out. I didn't know there were so many still around these days. I have 2 in my garage to repair now.

Hi Jalex,
I am fine, thanks. I check this forum almost daily, but when there are no new paragraphs, I do not log in.
I wonder why the anemometer employs two diodes instead of NTC thermistors. The NTC thermistors might be more sensitive and certain types more robust, too.
The common way of making an opamp with defined variable gain approximated by short straight lines is using a diode-resistor feedback network like this. Maybe it will bring some inspiration to you.

VBR from Ivan

Thanks Ivan
How have you been these days. We haven't talked for quite a while..
I think I figured it out. What it's doing is comparing the current of two diodes
(one cold and one hot) to get the DC offset and deriving the output
from the one in the air flow.
You know me I just screw around not building anything practical  LOL
That circuit confused me as I always though of instrument amps like you with 3 op-amps 
and not just two.
Here is the circuit.
http://www.mtmscientific.com/anem.html
What I am going to try is to watch the output curve with another op-amp and use a jfet
as a variable resistor to flatten the curve by varying the gain.
Thanks again for you help.



Just thought I would add I built it and it works pretty good but still running a few tests.

Hi Jalex,
it is an instrumentation amplifier. The more common model uses the third opamp as a differential amplifier between the outputs of opamps 1 and 2.
The aim of instrumentation amplifiers is to suppress the common voltage on the input to the maximum extent and provide a high impedance differential input. The gain of INA122 is controlled by a single external resistor.
Sorry I have a little experience with instrumentation amplifiers. Try someone else please.

BR from Ivan OK1SIP

Thanks Ivan
But could you explain it a little more.  Does the bottom amp vary the gain on the top one as the way it''s connected as both of the look like they are connected  non-inverted. to me it looked like the top one has a gain of 75 and the bottom has a gain of .75. So when you what happens when the resistors are connected like that. Does it make that amp into a current amp or what. I have never used an op-amp in this way so I didn't know what it would do.
  This IC was used in an anemometer circuit measuring diode current.  The output is very non.linear as expected and I was going to try to use a third op-amp to flatten it out if possible by varying the gain a little at certain points.



.

Hi Jalex,
the lower opamp is non-inverting, so its gain is never less than 1.

BR from Ivan OK1SIP

Hi  all
I want to copy the circuit of the INA122 and use it in LM324.  I do realize that it won't be a precision instrument circuit any more.
I could use a little explanation on how th INA122 works. Looking at the spec.  sheet  It looks like the lower op-amp is connected at less than unity gain,  
How exactly does that work?  Thanks