Design an adjustable voltage switch

[Roff]

I might be able to help you, although I have never laid hands on an LM2917, and I know of only one spice model, and it was made by a member of another forum.
I'll be unavailable until at least Sunday.

 

[Roff]

How many tach pulses do you get per revolution?

 

[stuhagen]

How many tach pulses do you get per revolution?

It is either 2 or 4. The circuit about has the correct components for my specifc RPM needs of 6800. I think he takes that into consideration.

Stu

 

[stuhagen]

it is either 2 or 4. The circuit about has the correct components for my specifc rpm needs of 6800. I think he takes that into consideration.

Stu

corrected: 3 pulses for my car

 

[Roff]

In terms of RPMs, where do you want the circuit to switch? Do you want hyteresis?
Don't make me drag the details out of you.
Tell me as much as you can think of.

 

[stuhagen]

In terms of RPMs, where do you want the circuit to switch? Do you want hyteresis?
Don't make me drag the details out of you.
Tell me as much as you can think of.

Ha..ha.. ok, well I just didnt want to burden you, but my intention is to use a range of around 3000-5600rpms. I need a 1000 rpm hysteresis if that is even possible. I plan on trying a rotary switch that has indents for 200 rpm increment, but a pot will do just as well. I need to trigger 2 of those VSVs (33 ohm each) at the exact same time. I think I can tie them together since there loads are around 250ma ea. Once triggered, they should not close until the rpms drop about 1000 rpms below that. That is how they function now. I have LEDs tied in parallel to them and I can watch to see when they trigger. Right now, they trigger together right at 4000 rpms, and the LED stays on until the rpms drop down to 3000rpms. So I assume that is the way the car's ECU works them. Now that I have changed other parameters in the car, I need to change these now to trigger at different rpms to better match the new configuration.

I hope I covered everything~!

Stu

 

[Roff]

Does the hysteresis need to be adjustable? I'm not sure how hard that would be, but I thought I should ask.

 

[stuhagen]

Does the hysteresis need to be adjustable? I'm not sure how hard that would be, but I thought I should ask.

IMO not really. As long as it drops at any level near the same. Not sure how hysteresis will work for this design, but assuming '0v' = 0 rpm and lets say 5v is redline, then 2.5v would be right in the middle, V wise. So in my case, my redline is 6800, and I need to be able to adjust in the range of 3000-5500rpms. Assuming this is a 5v cap, I am in the middle. This means (approx) for every 1000rpms is about .75v (6.8 x .75 = 5.1v) which means I need about a .75v hysteresis. This make sense? This can be static. Non varying.

If the overall voltage in the circuit is more like 0-12v then this of course raises the hysteresis proportionaly.

Stu

 

[stuhagen]

Ron..you still interested in looking at this RPM switch>?

Stu

 

[Roff]

Ron..you still interested in looking at this RPM switch>?

Stu

I don't feel real comfortable without a better spice model of the LM2917, so I'm working on that right now. I should be able to get to the RPM switch in a few days.
If you have a surplus of LM2917's, you could send me one, and I could do a hardware breadboard.

 

[stuhagen]

I don't feel real comfortable without a better spice model of the LM2917, so I'm working on that right now. I should be able to get to the RPM switch in a few days.
If you have a surplus of LM2917's, you could send me one, and I could do a hardware breadboard.

I was told radio shack carried them, but they are not on there website. I may see about getting some tomorrow.

Stu

 

[Roff]

First cut at the tach switch

Can you breadboard this? Do you need me to send the parts back first?

 

[nick007]

Thinking outside the box, you can use a Picaxe chip to sample the voltage, and have it either close or open a circuit (via MOSFET?) at a programmable value.

Andrew

 

[Roff]

Thinking outside the box, you can use a Picaxe chip to sample the voltage, and have it either close or open a circuit (via MOSFET?) at a programmable value.

Andrew

Stu and I are hopelessly stuck inside the box. I don't want to spend the time to learn to program a Picaxe, and I think I can safely say neither does he. Perhaps you would like to step in here?

 

[stuhagen]

Stu and I are hopelessly stuck inside the box. I don't want to spend the time to learn to program a Picaxe, and I think I can safely say neither does he. Perhaps you would like to step in here?

Dont think we can both fit in this box...ha...ha...

Thanks for the circuit....now to the etching and drilling~!

PS: digikey sent me an email stating that MOSFET has been discontinued. New version: **broken link removed**
Although I see very little difference.

PPS: Are we safe in not using a regulator, using 12v car is OK?

Stu

 

[Roff]

Dont think we can both fit in this box...ha...ha...

Thanks for the circuit....now to the etching and drilling~!

PS: digikey sent me an email stating that MOSFET has been discontinued. New version: **broken link removed**
Although I see very little difference.

PPS: Are we safe in not using a regulator, using 12v car is OK?

Stu

You're gonna go straight to PC board, without breadboarding and testing?

You shouldn't need a regulator. The LM2917 has a built-in zener regulator, and the rest of the circuit should be fine straight off the battery/alternator.
Let me know if you need help finding another MOSFET.

 

[stuhagen]

You're gonna go straight to PC board, without breadboarding and testing?

You shouldn't need a regulator. The LM2917 has a built-in zener regulator, and the rest of the circuit should be fine straight off the battery/alternator.
Let me know if you need help finding another MOSFET.

Why not~! You already did the testing! I am terrible with the bread board stuff. I can do a layout real quickly and drop in the solution. Although I may try the bread board thing. Last time I did a bread board, the wires shorted on me, so I need to do a better job. I think this MOSFET listed is pretty good isnt it? I tried for an hour searching and that 1 at digikey was the best oen I could find.

Stu

 

[Roff]

Why not~! You already did the testing! I am terrible with the bread board stuff. I can do a layout real quickly and drop in the solution. Although I may try the bread board thing. Last time I did a bread board, the wires shorted on me, so I need to do a better job. I think this MOSFET listed is pretty good isnt it? I tried for an hour searching and that 1 at digikey was the best oen I could find.

Stu

If a PCB works best for you, go for it!
I did breadboard the LM2917 part, and I simulated the 2 transistor driver. I didn't test it on a vehicle, of course.
The MOSFET is fine. Just don't specify it when you go into production.
BTW, if you make money on this, I want a cut!

 

[stuhagen]

If a PCB works best for you, go for it!
I did breadboard the LM2917 part, and I simulated the 2 transistor driver. I didn't test it on a vehicle, of course.
The MOSFET is fine. Just don't specify it when you go into production.
BTW, if you make money on this, I want a cut!

YoU keep the parts that I sent. May come in handy later. Why do you say "do not specify he MOSFET" ? You have something better?

Your "in" on the cut~!

Stu

 

[Roff]

YoU keep the parts that I sent. May come in handy later. Why do you say "do not specify he MOSFET" ? You have something better?

Your "in" on the cut~!

Stu

I just mean don't go into production with an obsolete part, unless you know you can get a substitute. In your case, though, that won't be a problem, since the MOSFET requirements are very loose.