Post by thread:[EE] Integrator circuit

Hi Guys I am building an integrator circuit using an OPA627 and I got the integrator to generate the ramp I require, but what I need also is to when the integrator reaches a certain point on the ramp, which will be variable, I need it to stop integrating. I don't mean reset as I do that after, but I need it to stop at that value for something like 300nS. I have -10V DC on the input. Values for C is 2.2n and for R 470ohms. In theory if I remove the input signal the integrator should just stop integrating but in practice do to switching delays, decay of signal and OpAmp offset voltage it will not work. Any ideas how to achieve this? Thanks Luis

Hi Luis. Store the resulted voltage in storage cell (next stage after integrator) As soon as integrator will reach required level switch will open shortly and voltage will be "copied" into storing capacitor. Also Hill & Horowitz book is good reference to start with this. WBR Dmitriy. Luis Moreira wrote: {Quote hidden}> > Hi Guys > I am building an integrator circuit using an OPA627 and I got the > integrator to generate the ramp I require, but what I need also is to > when the integrator reaches a certain point on the ramp, which will be > variable, I need it to stop integrating. I don't mean reset as I do that > after, but I need it to stop at that value for something like 300nS. I > have -10V DC on the input. Values for C is 2.2n and for R 470ohms. In > theory if I remove the input signal the integrator should just stop > integrating but in practice do to switching delays, decay of signal and > OpAmp offset voltage it will not work. > Any ideas how to achieve this? > Thanks > Luis > > -

Even easier when I read your message one more time. Switch between integrator output and storage cell is always open except those periods when you need to "keep" integrated voltage. Dmitriy Kiryashov wrote: {Quote hidden}> > Hi Luis. > > Store the resulted voltage in storage cell (next stage after integrator) > As soon as integrator will reach required level switch will open shortly > and voltage will be "copied" into storing capacitor. > > Also Hill & Horowitz book is good reference to start with this. > > WBR Dmitriy. > > Luis Moreira wrote: > > > > Hi Guys > > I am building an integrator circuit using an OPA627 and I got the > > integrator to generate the ramp I require, but what I need also is to > > when the integrator reaches a certain point on the ramp, which will be > > variable, I need it to stop integrating. I don't mean reset as I do that > > after, but I need it to stop at that value for something like 300nS. I > > have -10V DC on the input. Values for C is 2.2n and for R 470ohms. In > > theory if I remove the input signal the integrator should just stop > > integrating but in practice do to switching delays, decay of signal and > > OpAmp offset voltage it will not work. > > Any ideas how to achieve this? > > Thanks > > Luis > > > > --

Hi Luis. You don't have to sample quick ( that's the trick :) The switch is always open ( except moments when you need to "keep" the voltage ). Which means the value on the storage cell output will be following integrator output except when it is "frozen" ( there is no delay and other things related to quick sampling ) If you need to remove influence of storage cell input to integrator output place a unity buffer in between them. WBR Dmitriy. PS. What type of TV system you are trying to built onto ? :) Luis Moreira wrote: > > Hi Dmitriy, > I can see what you mean, but the problem is the storage cell, you will > need some time to sample and my ramp has a dt of 1uS, by the time you > sampled, the value on the cell will not be the one I wanted to stop at. > Best regards > Luis > > > {Original Message removed}

Hi Dmitriy, I am getting confused... The switch should be normally closed not normally open, right? You open to freeze? The storage cell we are talking about is a cap or it's there a chip available with some special function? Best regards. Luis P.S- no TV system, I am not that interested in that field, I am extracting a voltage signal from a time delay between to pulses. Luis Moreira spam_OUTluis.moreiraTakeThisOuTjet.uk tel. 01235464615 JET PSU Department UKAEA Culham Division J20/1/55, Culham Science Centre Abingdon Oxfordshire OX14 3DB {Original Message removed}

At 09:05 AM 12/2/2005 +0000, you wrote: >Hi Guys >I am building an integrator circuit using an OPA627 and I got the >integrator to generate the ramp I require, but what I need also is to >when the integrator reaches a certain point on the ramp, which will be >variable, I need it to stop integrating. I don't mean reset as I do that >after, but I need it to stop at that value for something like 300nS. I >have -10V DC on the input. Values for C is 2.2n and for R 470ohms. In >theory if I remove the input signal the integrator should just stop >integrating but in practice do to switching delays, decay of signal and >OpAmp offset voltage it will not work. >Any ideas how to achieve this? >Thanks > Luis 300ns is a pretty short time, even for a fast op-amp like the OPA627, and you'd also have to worry about charge injection in the analog switch(s) (though not so much with such a high integrator current). How are you 'removing the signal'? If you ground the input, the drift due to offset should be no more than 0.5mV/(470 * 2.2nF) = 484V/sec ~= 0.15mV in 300ns. If you open the resistor up, it should be less again. What spec are you looking for? Is this to get an ADC measurement? Could you simply follow the integrator with a S&H circuit? You can also consider using a current source and demoting the op-amp to buffer status. Are you building something like a LIDAR that you need sub-nanosecond resolution? Best regards, Spehro Pefhany --"it's the network..." "The Journey is the reward" .....speffKILLspam@spam@interlog.com Info for manufacturers: http://www.trexon.com Embedded software/hardware/analog Info for designers: http://www.speff.com ->> Inexpensive test equipment & parts http://search.ebay.com/_W0QQsassZspeff

Hi Spehro, The application is similar to an ADC and the idea was to use an S&H but the problem is the sampling time I need less than 100nS acquisition. Best regards Luis P.S-thanks for introducing me to LIDAR, never heard of it but after a bit of googling, it seems very interesting. Can you build one? Luis Moreira luis.moreiraKILLspamjet.uk tel. 01235464615 JET PSU Department UKAEA Culham Division J20/1/55, Culham Science Centre Abingdon Oxfordshire OX14 3DB {Original Message removed}

> I am building an integrator circuit using an OPA627 and I got the > integrator to generate the ramp I require, but what I need also is to > when the integrator reaches a certain point on the ramp, which will be > variable, I need it to stop integrating. I don't mean reset as I do that > after, but I need it to stop at that value for something like 300nS. I > have -10V DC on the input. Values for C is 2.2n and for R 470ohms. In > theory if I remove the input signal the integrator should just stop > integrating but in practice do to switching delays, decay of signal and > OpAmp offset voltage it will not work. > Any ideas how to achieve this? Hi Luis, Maybe you're doing it right already but just need a higher quality capacitor? Cheerful regards, Bob

At 03:54 PM 12/2/2005 +0000, you wrote: >Hi Spehro, >The application is similar to an ADC and the idea was to use an S&H but >the problem is the sampling time I need less than 100nS acquisition. >Best regards > Luis It's not really a trivial problem >P.S-thanks for introducing me to LIDAR, never heard of it but after a >bit of googling, it seems very interesting. Can you build one? Probably, with some effort. I did a preliminary time measurement design that worked quite well in simulation. It used multiple GHz ft discrete bipolar transistors. Best regards, Spehro Pefhany --"it's the network..." "The Journey is the reward" .....speffKILLspam.....interlog.com Info for manufacturers: http://www.trexon.com Embedded software/hardware/analog Info for designers: http://www.speff.com ->> Inexpensive test equipment & parts http://search.ebay.com/_W0QQsassZspeff

On Fri, 2005-12-02 at 09:05 +0000, Luis Moreira wrote: > I am building an integrator circuit using an OPA627 and I got the > integrator to generate the ramp I require, but what I need also is to > when the integrator reaches a certain point on the ramp, which will be > variable, I need it to stop integrating. Luis, Another option to make the integrator stop integrating is by feeding it a function whose integral is zero. This is hard to achieve in practice, but one possibility is a high frequency square wave. Many years ago I needed a DAC, but only had a spare opamp and an I/O pin. So, I created an integrator from the opamp, and applied a step function to the input to achieve a ramp and then a high frequency square wave to achieve a small amplitude triangle wave. The small amplitude triangle wave approximated the DAC output I needed. I had the luxury of milliseconds though, so I doubt this is possible with your circuit. The output looked something like: /\/\/\/\/\ / / / / / Scott