# Help with LM2917

Discussion in 'General Electronics Chat' started by rrvroja, Feb 25, 2014.

1. ### rrvroja Thread Starter New Member

Feb 20, 2014
5
0
Hi

I'm trying to use the LM2917 IC as a frequency to voltage converter circuit. However I am unable to get a change in the output voltage which is proportional to the frequency. I have used the circuit given in the TI data sheet for LM 2917. The Fin is a 1kHz square wave with an amplitude of 5Vpp. The capacitance connected to pin 2 is 0.02 microF and the resistor connected to pin 3 is 560 kilo ohm. There is no change in my output voltage if I vary the frequency. I have tried this circuit with both the 8pin and 14 pin versions of LM 2917. Please help.

Thank You

2. ### pwdixon Member

Oct 11, 2012
488
56
Perhaps if you posted your circuit diagram someone could help.

3. ### pwdixon Member

Oct 11, 2012
488
56
Is your input bi-polar? ie. does your input go positive and negative?

4. ### rrvroja Thread Starter New Member

Feb 20, 2014
5
0
Yeah it is. Its a square wave that has the range between +2.5V to -2.5V

File size:
28.2 KB
Views:
82
5. ### pwdixon Member

Oct 11, 2012
488
56
If the diagram is right with 66Hz/V then 1kHz will be 15V, with a 12V rail I would guess nothing would change until your frequency was at least below 792Hz. Even then that assumes the device can swing rail to rail and it probably can't so I doubt that you would see any change before something less than 792Hz. What frequencies have you tried?

6. ### pwdixon Member

Oct 11, 2012
488
56
You also said that the resistor value was not as per the schematic and that probably makes a difference to the Hz/V ratio. Did you calculate the resistor from the datasheet? I haven't looked at the datasheet for that calc yet.

7. ### pwdixon Member

Oct 11, 2012
488
56
Quick look at the datasheet and there's a calculation for output voltage that if you put in 12V for the supply, 1kHz, 0.02uF and 560kOhm looks like more than 120V to me, obviously not possible and means fin might be limited to 100Hz.

8. ### rrvroja Thread Starter New Member

Feb 20, 2014
5
0
Thanks for help. When we tried the lower values of frequency, we were able to get a proportional output voltage.
Our main aim for this circuit was to create a capacitance to voltage converter. Instead of varying the frequency, we kept a fixed frequency and varied the capacitance. The paper we referred to mentioned that if we used Fin as 1KHz, Vcc =10 V, and the resistor at pin 3 as 1MΩ, we would be able to measure the capacitance in the range from 0.1nF to 10mF. However, we need to design a circuit to measure capacitance in the range from 10pF to 100nF. Can you help to design the circuit for these specification?

9. ### pwdixon Member

Oct 11, 2012
488
56
The problem is the number of decades of capacitance. If you have a processor you could try either switching in different resistors or you could change the frequency to get a reasonable output swing over the range of capacitance. Without a processor you will need something like a manually switched set of resistances. Obviously you will also need to read the output and refer back to the frequency of switched resistance to interpret the end capacitance value. All fairly simple stuff if you have a processor involved.

Feb 20, 2014
5
0
11. ### pwdixon Member

Oct 11, 2012
488
56
So you have completely changed what you are doing and all your previous questions need no further answer.

12. ### rrvroja Thread Starter New Member

Feb 20, 2014
5
0
It is a processor like what you mentioned and it uses a switched resistor mechanism. Thanks for the help