No announcement yet.

Bedini Comparator Cap Dump

  • Filter
  • Time
  • Show
Clear All
new posts

  • Originally posted by RS_ View Post

    Here is a 12V version of Bedini's Comparator Trigger Cap Pulser......

    hope this helps every one build them at 12V
    Hello RS
    Thanks for all that information and the 12V Circuit.
    I have 2 Questions please:
    I am not sure how the Opto is connected.
    Pin5 -> + charging
    Pin4 -> Base Q1
    Pin6 -> "nothing"
    Pin1 ??
    Pin2 ??

    Is there a different if i use my SSG with or without the cap dump on the charging battery? Charges faster or what does it do?

    Thanks so much.


    • Cap Dump Circuit for 36V system

      Hi RS,

      I'm late to the party but have been developing a solid state generator (attached) and am at the stage where I wish to compare the battery charging when feeding the HV pulses (attached) directly to the two 36V battery stacks to when some cap dump system is used. The HV scope traces were taken with a 45:1 voltage divider so the pulses are around 700V, although will be around 8,000V when I have switched to a different set of coils with high permeability cores.

      I was sent your circuit by another member but, as relatively new to some of these things, I wondered if you could ask some questions re the circuit (attached).

      1. I would like to be able to run the cap dump from the drive batteries and not some external power source so can I ignore the FWBR on the left and connect directly to my 12.7v supply rail?

      2. This circuit uses an optocoupler that I have heard said can impair the fast switching of the FET. Is this something you recognise or is it insignificant?

      3. Which components allow one to set the discharge voltage threshold (I assume the discharge rate will be a function of that and the cap value)

      4. Can one build in a simple means of indicating that the main cap is charged (LED?) and also of safely discharging it after the main generator is turned off?

      Thanking you

      Attached Files
      Last edited by JulesP; 01-30-2019, 02:24 AM.
      'Consciousness came First'


      • Hi Jules,

        I am sorry that i have not seen your post any sooner....

        got a new puter before x-mas and just now logged back into this site

        The Cap Dump compairtor circuit is powered from the energy that is coming into the dump cap, no need to power it with the drive battery's

        I don't know why you would want to ignore the FWBR, and connect directly to the 12.7v supply rail, this would just fill the dump cap to 12.7V and never do any thing else.... The FWBR should be on the out put of your oscillator, to feed the spikes into the dump cap, to fill up the cap till the comparitor trips and dumps the cap, and then repeats....

        The H11D1 is what JB has always used. There are newer faster Opto's out there, and they can be used in place of the H11D1.

        R10 pot adjusts the voltage that the circuit discharges at....

        the discharge rate is a function of the amount of energy coming into the Dump cap, and the discharge rate will vary as the input energy vary's.
        Such as the primary discharges, etc....

        The comparitor's output could be fed to another Opto that can kill the trigger of your oscillator during the cap dump if needed


        • Hi RS,

          Thank you for your thoughts.

          Since writing that post I have had a chance to get to understand the circuit better and have drawn my own version of it (attached) that I find helpful as a preparation for a build and which incorporates my battery swapper and an LED off the LM741 to indicate activity. I now recognise that the timing control power supply is from the energy delivered to the main cap but even so, from the SG Intermediate handbook*, I read that the FWBR is not required as the HV pulses are unidirectional anyway and to save on voltage drop. That is why my circuit brings the HV feed straight to the 15,000 (or greater) cap.

          *"p30-31: The full-wave bridge requires the output pulse to travel through two diodes that each have a voltage drop of about .6 volts. . . . . .So, it seems clear that collecting the output pulse from the main coil winding (MC) and using a single diode (D2) to direct that impulse to the capacitor (C) is more efficient."

          With regard to the charge and discharge voltage levels that a cap dump system working with 36-38V batteries should be adjusted to, I understand that the upper threshold should be about 48V and the lower around 40V? I have displayed this in the attached graphic and wondered if you think those are reasonable.

          My battery charging experiments with the solid state generator indicate that as the frequency increases the charging benefit declines and is certainly not a liner function of frequency. I'm thinking that as the interval between succesive pulses gets shorter and shorter the battery is unable to fully 'process' them. I'm hoping that a cap dump circuit in between will be better able to absorb the hot and cold aspects of the HV pulses and relay them more effectively to the batteries and that increasing my cap to 60,000uF will be useful too.


          Attached Files
          'Consciousness came First'


          • Hi Gary,

            Small introduction:
            As discussed in the below mentioned thread*, I could not get steady results with consecutive runs of my SG in C.G. (common ground) mode and decided to hookup my comparator (Nityesh Design), hoping that my results would stabilize (COP would go down in consecutive runs). And maybe also improving, since I’ve noticed that ‘pre-‘charging my output battery with the TeslaGenX chargers seemed to improved COP.

            Dusting of the comparator and firing it up again:
            I hooked up two of my ‘lab rat’ batteries for input and output. Just to see where the settings of my comparator were when I put it in a box long time ago. I call the batteries I use ‘lab rats’ since they are batteries that perform the least of all the batteries I have: 12V - 24Ah Flooded -> VMF - 52805 12N24, 24Ah 293A(EN)
            First thing that I noticed is that my input amp meter would ‘bounce’ rather than having a steady value like it would in C.G. mode without the comparator. And also the bouncing seemed erratic, not sure if I can upload videos here, but I’ll try otherwise I’ll make a link to my YT channel.

            I had my digital scope hooked up to the capacitor bank and also there the behavior looked erratic: after a couple of fairly regular pulses (cap dumps) it showed some really irregular ones… see attachment 1.

            The next issues I ran into were the following:
            To determine the voltage to where the cap should charge before dumping and from where it started to charge again, I looked on page 46 of the intermediate handbook, where it read +/-2x battery voltage to dump, and +/-5V above battery voltage to start again. The first one I assumed to be then 24V, and the latter 5V above the fully charged voltage, so say 15V+5 = 20V.
            However with the pot-meter controlling the peak for the dump voltage dialed in its extreme position, this value was at +/-26-27V (so I could not bring it further down). The other pot-meter setting the value where the cap would start charging again was at +/-19V, but if I would turn that dial up (voltage where to start charging would go higher), also the peaks to where it would charge would move up.

            For this initial test I used the following capacitor:
            Mallory CGS253U075X4C, 75V, 25,000uF - ESR 0.014-ohm
            I have three of them which I can put in parallel (75,000uF), but started with 1 for now.

            The ‘bouncing’ input amperage reading makes it a bit hard for me to get an ‘average’…. And also I don’t know what is the cause of the erratic cap dumping. Gary, do you have any idea?

            Youtube link bouncing Amp meter:

            *I just switched to this thread from a long conversation with Gary on another thread:

            Best regards,

            Post nr 140 - Photo of setup.pdf
            Post nr 140 - Attachment 1.pdf


            • Hi Rodolphe,

              The ‘bouncing’ input amperage reading makes it a bit hard for me to get an ‘average’…. And also I don’t know what is the cause of the erratic cap dumping. Gary, do you have any idea?
              Sorry, I don't have any ideas either.

              Gary Hammond,


              • Hi Rodolphe,

                I will have a think about my own experience of building and testing a cap dump circuit and put some thoughts together in the next day or so.

                'Consciousness came First'


                • Hi Gary, Jules,

                  Maybe the erratic behavior is because I have the wires to and from the mosfets too close to each other and too much crossed? (see photo)

                  Although I don’t remember seeing the capacitor signal being so irregular when I first built it. But then I was running in Radiant mode, not in C.G. (common ground) mode…

                  Whatever the cause, I’m not going to worry about it too much for now: It seems that for now the comparator will not make things easier, but rather more complex…
                  For a while I’ve thinking about leaving the Bedini SG for what it is, and going to a simpler setup; solid state (just coil pulsing) and from there add complexity/parameters. Not sure if ‘simpeler’ is the correct term, but something that I can possibly have more control over…
                  Being made aware of this thread you started Jules by Gary, reading through your document, makes me want to go into that direction even more:


                  Best regards,

                  Click image for larger version

Name:	142 - Comparator photo.jpeg
Views:	40
Size:	551.8 KB
ID:	73829


                  • Hi Rodolphe,

                    As you may know from Gary, I built a cap dump circuit for my energy harvesting project as part of a test rig to confirm or refute prior claims of CoP>1. As it turns out I’m getting CoPs >10 but that isn’t due to the cap dump circuit but rather direct HV pulses of about 1,000V directly to the battery.

                    Nevertheless, if you are having trouble with your setup I can offer a few thoughts based on my build.

                    Mine was a high-sided version of a low-sided design by Niytesh etc as I think you said you also used. As I already had my generator built I was unable to change the whole configuration to make a low-sided version work hence the adaptations as in the attached circuits.

                    I first built a breadboard version of the main components so that I could check the function and iron out any problems and that’s where I found one that I couldn’t have predicted and which is worth checking on yours.

                    I had assumed that the comparator I used was a good and effective item but I was not getting the output voltage as expected and it was not switching at the expected threshold. I came to suspect that the item was ‘dodgy, as it is quite common now. People make poor copies of devices that may still work but not to the spec that the original was designed for. So my main suggestion to you is, if you haven’t already done this, swap the comparator for one of ‘better provenance’, perhaps from a trusted stockist like Farnell, Mouser etc.

                    Your wobbling current might be to an unstable comparator for which the switching threshold is not as stable as it should be.

                    Apart from that there could be a dry junction or two cables too close to each other somewhere so thoroughly check continuity at key points and where you think there might be a poor joint, reheat it and reflow the solder.

                    Let me know how you get on.

                    If you need higher quality versions visit the files on the link in the 'Suggestions' doc in the next post under the 'An energy harvesting project'

                    Click image for larger version  Name:	image_6604.png Views:	6 Size:	54.5 KB ID:	73834

                    Click image for larger version  Name:	image_6605.png Views:	4 Size:	70.3 KB ID:	73835

                    Click image for larger version  Name:	image_6606.jpg Views:	5 Size:	512.1 KB ID:	73836
                    Last edited by JulesP; 11-06-2022, 12:04 PM.
                    'Consciousness came First'


                    • Jules,
                      Thanks a lot for you input. I think for now I’m not going to worry about the cap dump circuit; I think I’ll have my hands full already in making an attempt in trying to make a solid state version similar to what you have. But thanks a lot for the help so far!

                      However, I did had a wick look at my OpApm, seems to be from a decent brand; Texas Instruments:

                      Furthermore, in case I continue with it in the future; could you send me the schematics you posted via e-mail (will send you a private message), since they got very blurry here.

                      Best regards,