Simple up/down 0-10V ramp generator

[lsuphysmn]

I would like to power on and off a high voltage (HV) power supply using up and down ramp generators. The HV power supply can be controlled with a 0-10Vdc input that linearly provides 0 to max output. Here is the idea:

When turning "ON" the HV power supply, a slow 0-10V ramp will produce a corresponding 0 - max HV output.

When turning "OFF" the HV power supply, the 10V input will be slowly ramped back down to 0, thus the HV output will correspondingly ramp back down to zero.

Off hand I am thinking a simple op-amp integrator circuit that charges/discharges a capacitor might provide a linear voltage ramp. Ideally this will be a simple push-button (on/off toggle) operation. Ramp rates (0-10V input) would be on the order of 1V/min. I would be interested in any other suggestions.

 

[barry]

You could just use an R-C if you don't need a linear ramp. That's about the cheapest way to go.

 

[FvM]

Of course this can be done with an integrator plus limiter circuit. Considering the large time constants, a small µP with a DAC or PWM analog output is probably a recent alternative.

 

[lsuphysmn]

Linearity is definitely not a requirement. The main issue is to adjust the HV up and down slowly. Currently the professor has a grad student who turns a pot manually every 6 seconds to change the voltage by 100 until it reaches maximum on power up and zero on power down.

 

[FvM]

I guess, linearity won't hurt. Using a simple RC low pass results in a rather long time to settling time to the final setpoint. You can however implement a strictly linear slew rate limiter according to the below scheme. The output follows the input, rate limited to about 1V/min (with +/- 15V supply). A more precise rate definition is possible using additional OPs.



Although I'm traditionally an analog guy, I would prefer a digital implementation these days, as already mentioned.

 

[goldsmith]

Dear FvM
Hi
Sorry , it is not my thread , but i have , a question about the circuit that you attached here . is its input an square wave ?
And how it will be generate ramp wave ?
And again sorry , Dear lsuphysmn , that i asked my question at your thread.
Thank you
Goldsmith

 

[FvM]

The input can be a square wave or a manual setpoint control, whatever you like. The rate is limited by the output voltage range of the first OP, which is translated to a ramp by the integrator. The first OP implements a DC feedback that ends the ramp at the setpoint voltage.

R2, R3, C2 are a frequency compensation means required for loop stability with universal compensated OPs, but not necessary to understand the basic operation.

 

[barry]

If a linear ramp is not required, then why get carried away with opamps and microprocessors? A SPDT switch, a 20K resistor and 100uF cap should do it (depending on what maximum allowable slew rate is). We're only trying to replace a (non-linear) grad student, we don't need a lot of components

 

[lsuphysmn]

A linear ramp would probably be better, but again, it isn't a priority. The whole idea is to have the on/off ramping functionality without the grad student having to sit there for ten minutes each time the HVPS is powered on and off. Goldsmith, no problem with your post; I welcome all comments and suggestions. Barry, you are right; the simpler the solution the better, especially since the prof is tight with his money and e-shop time.

Footnote: HVPS input impedance is 1.1 MOhm so I want to be careful that whatever circuit I use is properly matched. The HVPS is actually controlled from 0-5Vdc (not 10V) and has a built-in 5V source that can be used as a reference for programming HV.

Thanks for all of your responses.

 

[KMoffett]

With that high input impedance, barry has the simple, no...most elegant solution.
RC Time Constant

Ken

 

[lsuphysmn]

FvM, I tried your circuit and it works perfectly. Although linearity is not required I'm thinking that the prof would prefer it to a simple nonlinear RC circuit. Thanks again for the help.

 

[tsmith35]

I guess, linearity won't hurt. Using a simple RC low pass results in a rather long time to settling time to the final setpoint. You can however implement a strictly linear slew rate limiter according to the below scheme. The output follows the input, rate limited to about 1V/min (with +/- 15V supply). A more precise rate definition is possible using additional OPs.

FvM, how does this circuit work? How would I go about changing the resistances and cap values to change the slew rate, as well as C2, R2 and R3 for stability? Yes, I'm green when it comes to designing anything electronic.

 

[FvM]

Maximum slew rate is maximum output voltage of voltage divider divided by RC time constant. The voltage divider has been introduced to handle the low slew rate of V/min. For faster slew rates it can be omitted and saturated OP output voltage divided by time constant gives the slew rate.

R2 and C2 have to be adjusted for sufficient phase margin of the overall feedback loop.