LM2917 Frequency Switch Circuit (Shift Light)

[EZ Tutty]

Hey, this is my first post on here so go easy on me.

I'm a bit of an electronics amature, but i'm looking to learn.

I've had a go at building a 'Shift Light' circuit for my car using a LM2917 IC (i have attached the circuit diagram to this post), to do this i made an ETCH'ed PCB

Overall the circuit functions as it should, however it seems to be very temperature sensative (the shift setpoint creeps up with ambiant temperature as much as 1500rpm to 5*C)

I've tried replacing the diode that goes to Pin 11 (or the inverting input of the comparitor op-amp) for a zener diode that when fitted in reverse gives a voltage of 0.7v as i assumed that a small signal diode i was using was fluctuating with temperature, however this has made no difference.

Please can someone try and point me in the right direction?


Thanks

Paul

 

[ericgibbs]

hi,
The 15VZ and transistor on the +12V input will give very poor regulation.

If the 'regulated' voltage changes, so will the forward voltage drop on D1 0.6V.
This will change the operating point.

EDIT:
I would split the 12K into two resistors, say about 1K and 4.7K
at the junction of the two resistors add a 6V2 zener diode, the 1K goes to the 'regulated' and the 4.7K to the diode D1

This will fix the junction at 6V2.

 

[EZ Tutty]

Opps, thats actually a 12v Zener regulating the voltage (voltage input varies between 12.6 and 14.4v)

Using a 4.7v Zener in reverse at D1 seems to give me a nice stable 0.7v at pin1, (this is only the reference voltage so i'd assume that 0.7v is perfectly acceptable.


However i have just made a bit of a breakthrough, i set the Pot so that the LED output was on at idle RPM (only just a few revs less would put the light out), then i applied a small amount of heat to all of the components none made any difference untill i reached capacitor C2, this almost instantly put the LED out, and i found that i could heat and cool C2 to turn the LED on and off.

I'm using small round orange capacitors, are there any less susceptable to heat?

Thanks for the help

Paul

 

[ccurtis]

If the 1uF capacitor is an electrolytic, it will have poor tempco.

I don't understand how the circuit trips at 1500 RPM to begin with, however, unless the tach input freq is 100 times less than the engine RPM.

Your trip-point frequency at the TACH input to the circuit can be no greater than 1/RC=10Hz.

 

[EZ Tutty]

the trip point is adjustable by the 50k Pot.

the LM2917 IC is a frequency to voltage converter.

the 1uF capacitor is electrolytic, however heating this component doesn't seem to make the trip point any different.

however the temperature of C2 (0.022uf) seems to move the setpoint up as it get warmer, C2 is a ceramic disc type capacitor, which i thought was very stable though

 

[ccurtis]

The value of C2 should have little effect on the trip frequency if the circuit is operating normally and the capacitor is not faulty.

How did you calculate the trip frequency of the circuit? No matter what value the pot is set at, the trip frequency can be no greater than 10Hz (1/RC). The Pot and 15K resistor will only reduce the trip frequency to below 10Hz.

What is the relationship between your engine RPM and the TACH input frequency?

 

[EZ Tutty]

According to the datasheet for the LM2917 IC, C2 is the timing capacitor, and also provides internal compensation for the charge pump.

on circuits requiring a higher rpm range C2 is reduced and vice versa.


As the tach input runs from one of the ignition coils i'd assume 1pulse = 1rpm, thats as far as i've looked into it.

other than the temperature of C2 altering the setpoint the circuit functions perfectly, with the pot adjusting the set point nicely.

 

[ccurtis]

According to the datasheet for the LM2917 IC, C2 is the timing capacitor, and also provides internal compensation for the charge pump.

Egads, you are right. I have C3 and C2 confused. My apologies! Lots of help I've been so far.

 

[ccurtis]

Not all ceramic caps are temperature stable. Try one with an "NPO" dielectric for best temperature stability.