Announcement

Collapse
No announcement yet.

IGSL

Collapse
X
 
  • Filter
  • Time
  • Show
Clear All
new posts

  • Originally posted by porkluvr View Post
    Whether to use 10k pot, or parallel 10k fixed res with 100k pot? There is a very slight difference in overall response - oh about 2 degrees at the CW end but either way
    looks like very good linearity.

    You might say my simulation is invalid because as you can see I added a 3pF capacitor in the circuit. I don't think that having the inherent instability that is otherwise encountered
    is a good idea. I think that using 3pF adds about one~two degrees of range at the CW end. In my experience, the disc ZCD waveform can be very erratic without that capacitor.

    What I did is to use a linear potentiometer model and step the wiper position in eleven 10% increments, from 0% to 100%. These sims ran at ~15kHz.
    Nice sim!

    I would guess with an ideal op amp, both pot configurations should be the same (assuming op amp output independent of load). Also, both pot configurations should have same relationship of wiper voltage to pot position, as long as no significant current is drawn from the wiper, the way I see it. I'm open to other theories...

    The 3 pF shouldn't affect the comparison if the op am is ideal I would think -- but often simulations defy my intuition!

    -SB

    Comment


    • Originally posted by Comp View Post
      {unless the wiper/slider connection was having to supply any appreciable level of current, which would be limited by the higher resistance of the track to wiper section,.....)

      Cheers, Fred
      Apologies to everyone, especially to: Comp, simonbaker, porkluvr.
      I thought that SB is thinking about replacing the pot. in general , not just IGSL in which will work just fine. Just now i read whole post of Comp, and it's exactly what did I want to say in a first.
      Porkluvr great simulation, and the 3pF cap is excellent idea.
      I attached txt file for switching pins on ic's, if someone likes to make my change.
      Cheers.
      Attached Files

      Comment


      • Nice sim indeed.
        @porkluvr, could you be bothered to zip and post your .asc-s here? I am kinda convincing myself to squeeze a whole IGSL into LTspice, but my lazy bottom refuses the effort. Pretty please :shy kiss:

        Comment


        • Originally posted by wallker View Post
          Apologies to everyone, especially to: Comp, simonbaker, porkluvr.
          I thought that SB is thinking about replacing the pot. in general , not just IGSL in which will work just fine. Just now i read whole post of Comp, and it's exactly what did I want to say in a first.
          Porkluvr great simulation, and the 3pF cap is excellent idea.
          I attached txt file for switching pins on ic's, if someone likes to make my change.
          Cheers.
          Yes, you were right in general, they're not equivalent (I think), so that was a good point.

          Cheers,

          -SB

          Comment


          • Originally posted by detecto View Post
            Last night i look a little in IGSL lowpass and highpass filters, and get some unexpected answers. This is just for one chanel for second is the same. I used this RC filter calculator http://www.ekswai.com/en_lowpass.htm
            Opinion? Answer? Conclusion?
            Regards
            Detecto
            WHOOPS!! I think you are reading FAR too much into that, THOSE are NOT the filters I think are being talked about. Sure those are just to remove HF switching noise in some circumstances, but I think the "filters" in question here are the active ones with the Op-Amps (Integrators I believe).

            SO, the dv/dt of the recieved signal MUST be above a certain rate or you get no ouput (THIS is how the ground signal is filtered out) as the target signal has a much higher rate of change than the prevailing slow ground signal.. See not rocket science after all.

            Tesoso use TWO POLE "Filters". One thing I've never understood though, and this varies from mfg ot mfg, is that Whites use FOUR op amps and call this a TWO POLE deisgn (I think) - two poles over TWO channels (I & Q) whereas Tesoro trying to sound like they are doing a better job, call this a FOUR pole filter (which is technically correct).

            Try sweeping aa moders VLF VERY slowly, no output right? There is your answer.

            SUGGESTION. SB can you modify the IGSL PCB to use jumpers to select the tones? Might be an idea.

            Comment


            • Originally posted by Davor View Post
              Nice sim indeed.
              @porkluvr, could you be bothered to zip and post your .asc-s here? I am kinda convincing myself to squeeze a whole IGSL into LTspice, but my lazy bottom refuses the effort. Pretty please :shy kiss:
              OK, Davor. Give me a day and I'll post a zip (just please don't kiss me ). It will take some time because there are some support files that need to be included and I also need to make repairs.

              I have noticed some problem areas (mistakes) in my sim. For one thing the oscillator that I used was only outputting 5Vp-p so you really could not see problems that would otherwise be apparent from inputting a 16Vp-p signal to U101B, with a +8/-6.5 power supply.

              What I had done with the oscillator was to kick it with 1nA initial condition into the coil. That was enough for startup but it would not sustain oscillation with r6=250. The oscillator amplitude was steadily decaying at 2mS when waveform recording started. Reducing R6 to 100 is key.

              I am always making modifications from the 'standard' circuit but sometimes I forget the extent of those mods, and they will slip through un-noticed. I can be comparing apples and oranges and don't even realize it. It is not wise to forget what has been changed from the original circuit, and then be complacent about what appears to be happening.

              Comment


              • Part of me is glad I never started the IGSL got all the bits incl circuit board months ago, just never got round to it, although built a second TGSL for further experiments, maybe nows the time to go for it, how is it rated upgainst the TGSL anyone?

                Regards

                Comment


                • TIA
                  I'll not kiss you though

                  Comment


                  • Originally posted by satdaveuk View Post
                    Part of me is glad I never started the IGSL got all the bits incl circuit board months ago, just never got round to it, although built a second TGSL for further experiments, maybe nows the time to go for it, how is it rated upgainst the TGSL anyone?

                    Regards
                    I also have the parts gathered but have not built the IGSL. Therefore am not qualified to compare the two. However since the main difference between IGSL/TGSL is multiple tones, I suspect their performance would be very similar.

                    Right now the various VID projects have got my interest. I have bought the Arduino Uno controller board and am going to take a swing at writing some software for it using dfbowers project that he posted recently as a start point.

                    My goal is to have a TGSL with many tone ID plus visual indicator.

                    This is a great hobby.

                    Jerry

                    Comment


                    • Here's the simulation file, Davor. A few points deserve mentioning.

                      The LF353 does not have a particularly wide output voltage range and that could cause problems when using a +8/-6.5 power supply if the transmitter reference signal strength is too high. I selected the values of R5/U3 (referring to the simulation designators) to give slightly less than 16Vp-p. That may have been unnecessary - but I definitely found it necessary to change the value of R8 to 75kΩ - reducing the chance of U2's output breaking up. Using a rail-to rail output amplifier would mitigate that potential problem and allow having a bigger transmit signal. Using a true -8V supply would do the same but that is a different problem in itself.

                      MC33178 is NOT a advertised as a rail-to-rail amplifier but it does come close and it might be my preference over the LF353. MC33178 is used in the Tesoro Bandito and from looking at its specs it should be adequate for the TGSL/IGSL although a little higher GBW would be better.

                      It seems like I stay in a perpetual design/PCB layout phase, so don't take my word for it, you experiment.

                      ---------------------------
                      Let me know if you have problem. It seems like I always forget to include some essential model or symbol file no matter how hard I try not to.
                      Attached Files

                      Comment


                      • Thanks. There are missing models of 2N222A and 2N2907A but I'll manage without them. Everything else works nice.

                        I might have a thing or two to suggest soon.

                        In case you have some other IGSL modules in LTspice as well, I'd be happy to have them as well. No rush, It is my lazy bottom speaking.

                        Comment


                        • Originally posted by Sean_Goddard View Post
                          WHOOPS!! I think you are reading FAR too much into that, THOSE are NOT the filters I think are being talked about. Sure those are just to remove HF switching noise in some circumstances, but I think the "filters" in question here are the active ones with the Op-Amps (Integrators I believe).

                          SO, the dv/dt of the recieved signal MUST be above a certain rate or you get no ouput (THIS is how the ground signal is filtered out) as the target signal has a much higher rate of change than the prevailing slow ground signal.. See not rocket science after all.

                          Tesoso use TWO POLE "Filters". One thing I've never understood though, and this varies from mfg ot mfg, is that Whites use FOUR op amps and call this a TWO POLE deisgn (I think) - two poles over TWO channels (I & Q) whereas Tesoro trying to sound like they are doing a better job, call this a FOUR pole filter (which is technically correct).

                          Try sweeping aa moders VLF VERY slowly, no output right? There is your answer.

                          SUGGESTION. SB can you modify the IGSL PCB to use jumpers to select the tones? Might be an idea.
                          Good explanation and points there. I think someone did put jumpers for playing with tones -- is it in this thread? (I'm not the right guy to modify the IGSL PCB)

                          -SB

                          Comment


                          • Originally posted by Davor View Post
                            Thanks. There are missing models of 2N222A and 2N2907A but I'll manage without them. Everything else works nice.

                            I might have a thing or two to suggest soon.

                            In case you have some other IGSL modules in LTspice as well, I'd be happy to have them as well. No rush, It is my lazy bottom speaking.
                            Here ya goes.

                            Add the lines to the standard.bjt file in the LTspice\lib\cmp folder.

                            If you have not already done so you should join the LTspice Yahoo group. There you will find additional library parts including CD4000 and 74HC digital devices.

                            Don't you be lazy.
                            Attached Files

                            Comment


                            • Don't worry, I am very friendly with LTspice, and I have models already. I notified you just to be aware that the model is missing in a set. Nothing else.

                              I myself tend to avoid adding models into LTspice library and instead put them in a working folder. This approach yields more portable projects, and as I play on several computers it makes sense.

                              Comment


                              • Finished my igsl board

                                Well I finished my igsl here is a picture if anyone sees anything wrong please let me know I am an amateur.
                                Attached Files

                                Comment

                                Working...
                                X