MikeML's tach circuit

Hi,

Rather than post to an old thread, I thought I'd start a new one here.

I was reviewing MikeML's engine tach circuit (which is great), but was just curious about the power dissipation needed in the resistors:

The original thread is here: thread



@MikeML,
Thanks for posting your tach signal condition circuit and simulation!

Just one question for you. In your design, wouldn't R3 have to be able to dissipate quite-a-bit of power? Using your simulation as a guide, if you had 630V as an input, would you not have (630V-0.65V)/(100k+10k)ohms=5.72mA going across R3? With a voltage drop of 100kohm*5.72mA = 572V for a total power dissipation of 572V*5.72mA = 3.27W? Granted, it would be a short duration and I presume the RC filter would help attenuate the spike, but just want to make sure I understand your circuit correctly...

Thanks!!!
Justin

You can measure the actual wattage dissipating in the simulation, you can also determine the RMS wattage over time, carbon resistors tolerate pulses quiet well as long as their total power and voltage limitations aren't violated.

I found this datasheet which lists peak voltage and joule ratings for those pulses and should be pretty typical for carbon resistors. Plot the graph for the power through the resistors and work the joule equivalent out, I'm going to hazard a guess that you'll be within tollerances for a quarter watt resistors, but definitely double check.

https://www.electro-tech-online.com/custompdfs/2011/05/CC-FA001.pdf

I'm going to hazard a guess that thin film resistors may not fair as well but you can Google that yourself =>

Hi Sceadwian,

Thanks for your help!

I re-created MikeML's circuit in LTspice and modeled the power across the resistor. The total duration of any power across the resistor is about 1ms. The peak voltage is less than 1kV.

You were right - at 1ms, a 100k 1/4 ohm resistor can handle up to 6kV. So, should be well within even 75% derating.

Thanks for your assistance!

coolj said:

Hi Sceadwian,

Thanks for your help!

I re-created MikeML's circuit in LTspice and modeled the power across the resistor. The total duration of any power across the resistor is about 1ms. The peak voltage is less than 1kV.

You were right - at 1ms, a 100k 1/4 ohm resistor can handle up to 6kV. So, should be well within even 75% derating.

Thanks for your assistance!

Click to expand...

hi,
I would suggest that you check the resistors datasheet for its maximum voltage rating, many are limited to approx 600V.
For EHT circuits its common practice to use a number of lower value resistors in series.

Thanks, Eric. I was actually planning on doing exactly that - putting several resistors in series. And I will double check their max voltage ratings!

Two resistors in series is good sense, since you're pretty close to the listed ratings; definitely reading the datasheet for your resistors that's again just good sense. The one I linked is obviously from a manufacture that is willing to state their specs, so probably a bit more expensive than cheap carbon resistor that will not rate their resistors for peak voltage or energy levels because of manufacturing variance.

While we're at it, would anyone have a guess as to the point of R2 and R5?

My thinking is R2 / R5 could be viewed as a voltage divider of the output of R3 / C1.

Or, R2 could be in place to allow R3/C1 filter to respond (versus Q1 sinking the voltage to 0.7V) with R5 as a weak pull-down for Q1.

Any thoughts?