I tried to run a simulation on this circuit that I copied form Analog Circuit Design by Jim Williams. I believe I have gotten everything correct except for the op amp which was an LM30A. The output from the op amp is supposed to switch between 13 and -13 volts. The circuit is not working that I can tell. Anyone have any ideas? I would appreciate any help. Thanks, Dan |
by dfitzgerald
September 08, 2013 |
by dfitzgerald
September 08, 2013 |
|
A quick look suggests you may have some component values out by a few orders of magnitude: C7 is set to 1F C5 is set to 1000uF C4 is set to 1F These values look way too high. Having played with the cap values and the control voltage, although it does appear to oscillate, it is not clear that the circuit is working as the designer intended. You may have made some drawng errors too. Trouble is, without the original Jim Williams circuit and no spec on what you want this circuit to do, it is difficult to know if that is the case. I've had a look through LT's appnotes: http://www./designtools/app_notes.php but haven't found one with JW's circuit in it. Can you post the original schematic somehwere? |
by signality
September 09, 2013 |
BTW: You said the opamp was an LM30A. Did you mean LM301A? |
by signality
September 09, 2013 |
Some clues to the original circuit here: Figure 23-7. A version of Peaseās elegant voltage-to- frequency converter circuit spotted in: http://www.introni.it/pdf/Williams%2007%20-%20Book%20Chapters.pdf Need to look through stuff by Bob Pease ... :) |
by signality
September 09, 2013 |
Your right. In the circuit from Bob Pease C4 and C7 are 0.01. He does have C5 at .001. Also the OP amp is an LM301A. I'm not sure what transistors he used and I did not include his optional VOS adjustment or his optional gain adjustment because they were listed as optional. I'll look through the pages by Jim Williams relating to this circuit. Thanks for your help. I appreciate it. |
by dfitzgerald
September 09, 2013 |
Please note that if you modify the schematic since you first posted your question then the circuit in the original posting updates too so can cause chaos in trying to understand the original question. To avoid such confusion, it's a good idea to leave the posted circuit alone, save a copy and then continue working on that copy. :) |
by signality
September 09, 2013 |
I will see if I can copy the circuit from the book and post the picture. |
by dfitzgerald
September 09, 2013 |
I wasn't aware of that. Thanks. |
by dfitzgerald
September 09, 2013 |
"In the circuit from Bob Pease C4 and C7 are 0.01. He does have C5 at .001." Common cause of confusion in many, many datasheets, apps notes and design notes. They very often don't clearly, explicitly state all units and scale factors. Makes a nonsense of many of the formulas used in app notes. In your case all the caps would be in uF so: C4 and C7 = 0.01uF (10nF) C5 = 0.001uF (1nF) Also note that in CL 1000u = 1m i.e. m is not short for microfarrads it is short for milli! OTOH: in CL (unlike spice) the scale factors are case sensitive so M = Mega https://www.circuitlab.com/docs/the-basics/#human_friendy_inputs |
by signality
September 09, 2013 |
Thanks. Do you know how I could post a copy of a scanned image of the circuit? Or I could e-mail to you. |
by dfitzgerald
September 09, 2013 |
As mentioned in the section on Bug Reports in: https://www.circuitlab.com/forums/support/topic/8s9n9hav/how-to-use-the-circuitlab-support-forum/ you could try imgur.com Or you can email me via my website ... :) |
by signality
September 09, 2013 |
Here's a fixed version with all the caps set to the values given in Robert Peases orignal sketch ... :) |
by signality
September 10, 2013 |
Thanks signality. I would have never figured that out. I really appreciate your help! |
by dfitzgerald
September 10, 2013 |
The startup expressions are really just a refinement to reduce the time until - and guarantees - start of oscillation. The circuit works without that, it just takes a bit longer to start. You could just use a non-zero start time. Your circuit was fine as posted: all it needed was the correct cap values. :) |
by signality
September 10, 2013 |
Not sure if RAP's patent is still in force ... :) |
by signality
September 10, 2013 |
Did you mean to write the acronym GAP for George A Philbrick? In the book "Analog Circuit Design" Bob Pease mentioned that because the patent on the 4701 expired in 1990 he didn't feel bad about discussing the design. I was interested to see how that charge dispenser was working. Sure takes some time to simulate. A lot going on in the circuit I guess. |
by dfitzgerald
September 10, 2013 |
Please sign in or create an account to comment.
CircuitLab is an in-browser schematic capture and circuit simulation software tool to help you rapidly design and analyze analog and digital electronics systems.