Announcement

Collapse
No announcement yet.

Gary's "Complete Advanced" SSG Build

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

  • #46
    Hi Gary,

    I have some new measurements with the new 20 magnet rotor (see attachment) which I hope you want to have a look at with me:

    Cycles 1/2/5/6 are all done with the same input/output batteries as you can see, and done as previous: charge up to 15.3V, discharge 1Ah. Like with the previous rotor, the results are very inconsistent… after cycle 1 I had high hopes, but then with the results of cycle 2/5/6 that hope was through the window completely

    So then decided the replace the output battery B (AGM/12AH) with one of me new bought batteries, LA1 (flooded/24Ah), to see if then the outcome would get more consistent. As you mentioned before, Batter B (and A) may be damaged because of being discharge at high amperes early in this SG development.
    Before I started cycle 7 with LA1 at the output, I did 2 discharge/charge cycle of 2Ah, and charged it with a regular charger (Noco G3500) at the 0.9A setting. Then I discharged 1Ah out of it and started cycle 7. As you can see in the remarks, I quit the test after 62 minutes since the voltage was not raising above 14.75V and the battery was producing a bad smell, probably because it was producing gas instead of charging. I drained 1Ah out of it again and then started cycle 8 with the same result but now at 14.86V. I assume that initial charge/discharge cycles I did with the regular charger were not enough to get the battery up to it’s full performance yet. So my question would be: what is the quickest way to get it there?


    When it says that no more of 20% should be discharged out of it, than I could do discharge cycles where I discharge 0.2*24Ah = 4.8Ah out of it (instead of 2Ah or 1Ah). But to stay on the safe side I could go for e.g. 4Ah. And then recharge with the normal charger.
    Or do you recon that for this “break in” process, it doesn’t matter whether I discharge 1Ah, 2Ah or 4Ah out of it, that it is more a matter of the amount of cycles, and not so much the depth of discharge. I also have a 3.5A charging setting which I could use*.


    Is there any indicator telling me the battery is ready to be charged up to 15.3V? (Other than trying to charge it with my SG up to that level).

    Many thanks in advance,
    Regards
    Rodolphe


    *12V Norm Mode
    For charging 12-volt Wet Cell, Gel Cell, Enhanced Flooded, Maintenance-Free and Calcium batteries with the G3500 Smart Charger. When selected, a white LED will illuminate. 14.5V | 3.5A | Up To 120Ah Batteries
    Small 12V Norm Mode:
    For charging 12-volt Wet Cell, Gel Cell, Enhanced Flooded, Maintenance-Free and Calcium batteries. When selected, a white LED will illuminate. 14.5V | 900mA | Up To 35Ah Batteries
    https://no.co/support/g3500-charging-modes

    2020-08-10 - Measurements V1.pdf

    Comment


    • #47
      Hi Rodolphe,

      So my question would be: what is the quickest way to get it there?
      Don't try for the quickest way in this research. John Bedini always said "it's about time". And "nature always takes it's own sweet time". Lead acid batteries like slow charge/discharge rates.

      New batteries are usually not fully formed and can be partially sulfated as well from sitting on the shelf for a month or two. Using the C20 rate or less, discharge them to about 12.2 volts under load. Then charge back with the SSG, again staying under the C20 rate, until the voltage plateau is reached. It may take several discharge/charge cycles to get up to 15.3 volts. The batteries will start off-gassing at about 14.6 to 14.8 volts. Going above this with a conventional charger creates heat and causes excessive off-gassing. But the SSG, especially in radiant mode, can safely push the voltage to 15.3 volts or more after the battery is fully formed and conditioned.

      The CG (generator) mode charges with current and somewhat resembles the way a conventional "hot" charger works. The cap discharge mode is probably the best way to condition a new battery, and the radiant mode works good as well. Refer to EFTV series videos #33 and #34. John covered this pretty well in these 4 video discs.

      After the batteries are conditioned, you can then resume with normal testing.

      Gary Hammond,

      Comment


      • #48
        Hi Gary,

        I wrote a reply to your post above here #47, but need to update it to see what’s still relevant, since the forum has been offline so long. I will update what I wrote one of these days and post here, but what is most important now I think is my latest test results. In the attachment you find 3 sets of tests:
        Rotor with 20 magnets, no flywheel
        Rotor with 20 magnets, 5kg flywheel
        Rotor with 24 magnets, no flywheel


        I managed to get the bandwidth of the outcome within a more narrow range as before, mainly by fixing some parameters with all the tests (e.g. 2.4A input current, no other battery configurations used), by doing a set of test in a row without doing anything else with the batteries in between (e.g. no Radiant charging in between...) and by not letting days go by between tests.

        Analyzing the results:
        As you can see the “20magnets, no flywheel’ set has the biggest spread in outcome, 1.28Ah-1.42Ah. The Flywheel seemed to stabilize this somewhat.


        But what is more important is that 24magnets or 20magnets (or a flywheel) didn’t seem to have a significant impact… So my guess is that I should make a 3rd rotor. Just not sure whether I should go above 24, below 20 (and by how much) or in the middle (e.g. 22)…

        I might forward this message to Erik from TeslaGenX too, since he was curious about my 20magnet rotor results as well.

        Looking forward to your input,

        Regards,
        Rodolphe

        2020-08-22 attachment.pdf

        Comment


        • #49
          Note:
          While the forum was offline, Gary and I continued our talk offline, this was Gary’s response to my post #48:


          Hi Rodolphe ,

          The only thing I would point out to you would be to watch EFTV part 33 and part 34 again. All four discs. I think it’s probably from the conference I attended in 2013 (Part 34) where John recommends an odd number of magnets, and specifically 21 of them. He also recommends 7 power windings on the coil as being the maximum number that the trigger coil can efficiently switch. He also added two fans to control the speed with a slight load.

          That is exactly the way I built mine and got the results I previously posted.

          Best regards,
          Gary Hammond,

          Comment


          • #50
            Hi Gary,

            Rotor magnets
            I’ll try to watch the DVD set 33 and 34 again these days. I’ll pay extra attention if I hear him mention something about an odd number of magnets or a fixed nr (21), but to be honest, I do not see a logical basis for either (yet): The coil/electrical circuit doesn’t “count magnets passing”. It eithers sees a magnet or it doesn’t. So in that sense I do understand the logical basis for duty time what we talked about earlier, and hence that there should be a difference in performance in more or less magnets/narrow or wider magnets/rotor diameter.

            Power windings
            I’ll also pay attention in the DVDs regarding the max 7 power windings too. Although here my preliminary comments are, that I would be surprised that TeslaGenX would make an 8 coiler if the trigger circuit would not be able to handle it. Furthermore if the trigger circuit could not efficiently switch more than 7 coils, I assume this would be reflected in the fact that my amp draw would not go beyond a certain value but I can easily turn my variable resistor so that the amps draw is 3+Amps. However, this is all under the assumption that an efficient/inefficient switching of the trigger circuit would show in the above mentioned way. But rather than thinking this through too far, I’m testing with 7 power coils while writing this (disconnected 1 side of a coil and the corresponding base resistor). Testing now with 20magnets rotor, will do a 24magnets rotor test too and report back here.


            Drag/Fan
            In your post #29 you tested without any fans or generator coils. This yielded a COP of 0.99 / costed 1Ah at the input. So I assume that I should be able to get to that figure too in theory without fans/generator coil. Leaving them out, narrows down the parameters that influence the performance, and that is what I try now: narrow it down as far as possible and isolate the impact of the parameters I do work with.


            Batteries
            Will respond to thins in a separate/next post in reply to your post #47.


            Best regards,
            Rodolphe

            Comment


            • #51
              Hi Gary,

              It is not my intention to hasten it, just to do it at the most efficient way. Often I do not know if a certain parameter (like the charging mode of my regular charger) has an impact or not. If it does not have an impact, I rather charge at a higher amperage; at the moment I have some time available and then try to make as much use of it as possible to run tests/progress with the SG. Each test takes a long time with the charge/discharge and letting the batteries sit at least for an hour in between to rest.

              Coming back to your comment of the C20 rate
              -The discharge via the CBA I do at C20 rates: 0.6A for 12Ah AGM, and 1.2A for 24Ah flooded.
              -Input battery: So far I’ve been using Battery X1, 12Ah AGM. In CG mode I use 2.4A as an input current, which is way above C20 for that battery. However, at the end of the CG mode cycles, I test the voltage of the input battery while running and it is +/-12.30V… (so well above 12.20V).
              -For radiant charging I’m using +/-1.45A, which is also above C20, and there the voltage of the input battery running (at the end of the cycle) is even higher.

              Condition of my AGM batteries (see attachment)

              The reason why I bought the new 24h batteries was because initially I was under the impression that my AGM batteries where damaged based in the distorted CBA pattern as you know. But since that is fixed again, I was wondering if there was another way of determining if any of my AGM the batteries are damaged. Starting with battery B (with which I tried to get consistent results in CG mode, of which I posted the results in my previous post); I compared the discharge graphs of this battery B after being charged with a normal charger, SG radiant and SG CG. Then I compared these patterns to the CBA pattern from your website. If the battery from the graph on your website was still in good condition during that test, I would assume my battery B certainly is: it holds a higher voltage level and also more consistent over a longer period of time (assuming that this is an indication for the health of the battery).
              My input battery X1; I did a CBA discharge with that one too, but that graph also looks pretty good to me.
              Now there is still my battery A, which sits at a lower resting voltage while being exactly the same battery as B, but that one I’m not using at the moment for the test result to determine my SGs performance. So I leave it out of the scope for now.

              So since you in your post #29 got a COP of 0.99 with your AGM batteries (total of 12Ah), and the AGM batteries I’m using seem to be OK, I was planning to stick with them for now to keep the amount of parameters to influence performance as low as possible.

              Best regards,
              Rodolphe

              2020-08-28 - attachment.pdf
              Last edited by pearldragon; 08-28-2020, 05:05 AM. Reason: uploaded attachment

              Comment


              • #52
                Hi Gary,

                Power Cores
                I ran some test with 7 power cores instead of 8: No difference.


                Magnets
                Next parameter I’m going to change are the magnets. At the time of building my SG, I had no knowledge of ‘duty time’ etc. So ended up magnets that I could order locally and easily: 60mm x 20mm x 15mm (LxWxH). While in the Handbook specified: 50x25x12.
                I’m going to make 2 new rotors:
                The first one with magnets matching more or less your magnet size: I ordered 55x22.5x9 Y30
                And one with magnets matching more or less the handbook magnet size: Will order 50x25x20 Y30 or 50x25x10 Y30, not sure yet, awaiting some availability/stock answers.


                Hope that this (the magnet) is the bottle neck and that I’m starting to see a change in output performance. So far the amount of magnets, flywheel, power cores did not yield a significant change in performance.

                Meanwhile continuing to watch the DVDs again that you mentioned.

                Best regards,
                Rodolphe

                Comment


                • #53
                  Hi Rodolphe,

                  The only thing I wanted to bring to your attention was differences between your machine and the machine John Bedini demonstrated to us at the 2013 conference. I tried to copy that machine as closely as I could, which is the machine I used (and still use) to get the data I've posted in this thread, other threads, and on my homepage. I have several other machines as well, that are built differently and give different results.

                  Every different machine, by every different person, will give it's own data results. Some may be very similar and some may be very different. And again, the batteries and multimeters used for measuring are very big variables in all this.


                  Rotor magnets
                  I’ll try to watch the DVD set 33 and 34 again these days. I’ll pay extra attention if I hear him mention something about an odd number of magnets or a fixed nr (21), but to be honest, I do not see a logical basis for either (yet): The coil/electrical circuit doesn’t “count magnets passing”. It eithers sees a magnet or it doesn’t. So in that sense I do understand the logical basis for duty time what we talked about earlier, and hence that there should be a difference in performance in more or less magnets/narrow or wider magnets/rotor diameter.
                  In EFTV video part 33 disc #1 from 00:46 thru 01:34 John shows the machine with 21 magnets and 7 power coils. At one point he misspoke and said 9 windings on the coil when he should have said 8 windings as shown all throughout all 4 discs of the videos.

                  At 53:30 thru 01:01:15 of disc #1 he talks about multiple power windings, balancing the circuit, and optimal speed.

                  At 19:10 thru 20:20 of disc #2 he talks about two fans, a certain rpm, and 7 strand coil.

                  At 01:07:30 thru 01:10:24 of disc #2 he recaps the machine to Pat and mentions two fans, rpm, frequency, and why he used 21 magnets. His explanation didn't make a lot of sense to me, but I went by it anyway in my build.

                  Power windings
                  I’ll also pay attention in the DVDs regarding the max 7 power windings too. Although here my preliminary comments are, that I would be surprised that TeslaGenX would make an 8 coiler if the trigger circuit would not be able to handle it. Furthermore if the trigger circuit could not efficiently switch more than 7 coils, I assume this would be reflected in the fact that my amp draw would not go beyond a certain value but I can easily turn my variable resistor so that the amps draw is 3+Amps. However, this is all under the assumption that an efficient/inefficient switching of the trigger circuit would show in the above mentioned way. But rather than thinking this through too far, I’m testing with 7 power coils while writing this (disconnected 1 side of a coil and the corresponding base resistor). Testing now with 20magnets rotor, will do a 24magnets rotor test too and report back here.
                  In part 33 disc #4 from 01:40 thru 05:46 he talks about triggering 7 devices max and using 21 magnets. He also mentions the importance of balancing the circuit and making a very exact build.

                  Drag/Fan
                  In your post #29 you tested without any fans or generator coils. This yielded a COP of 0.99 / costed 1Ah at the input. So I assume that I should be able to get to that figure too in theory without fans/generator coil. Leaving them out, narrows down the parameters that influence the performance, and that is what I try now: narrow it down as far as possible and isolate the impact of the parameters I do work with.
                  That sounds reasonable, but as I mentioned earlier in this post every machine you build and every machine everyone else builds will give it's own data results. And the batteries must be considered as part of the build. The machine is only part of the total picture.

                  Paul Babcock, and some others, have suggested that even variations in the aether, the weather, the time of day, the phase of the moon, the altitude, and the geographic location can all affect the actual results as well. This is not a very well understood or exact science at this point. We are all still learning.

                  Gary Hammond,


                  Last edited by Gary Hammond; 08-30-2020, 12:29 PM.

                  Comment


                  • #54
                    Hi Rodolphe,


                    Originally posted by pearldragon View Post
                    Hi Gary,


                    [FONT=Calibri]Coming back to your comment of the C20 rate
                    -The discharge via the CBA I do at C20 rates: 0.6A for 12Ah AGM, and 1.2A for 24Ah flooded.
                    -Input battery: So far I’ve been using Battery X1, 12Ah AGM. In CG mode I use 2.4A as an input current, which is way above C20 for that battery. However, at the end of the CG mode cycles, I test the voltage of the input battery while running and it is +/-12.30V… (so well above 12.20V).
                    -For radiant charging I’m using +/-1.45A, which is also above C20, and there the voltage of the input battery running (at the end of the cycle) is even higher.
                    One big thing I see here is that your primary battery voltage is too low at 12.3 volts while it is powering the SSG. Compare that to the run battery voltage I posted on my home page, which is
                    The run voltage on the primaries started at 12.54 volts under load and only dropped to 12.52 volts during the run.
                    That tells me you have way too much voltage drop in your primary battery. This could be due to too much current, too much battery internal resistance, or too small of a battery capacity. Try paralleling two batteries for the primary run source and then dialing back the input current to a little less than 2 amps. What we are comparing are Amp Hours in to replace 1 Amp Hour previously discharged from a fully charged battery. The watt hours required will be less than a cop of 100, while the amp hours can go well above a cop of 100.

                    Gary Hammond,
                    Last edited by Gary Hammond; 08-30-2020, 02:36 PM.

                    Comment


                    • #55
                      Hi Rodolphe,

                      I just finished re-reading all our posts back and forth in this thread and just realized another big difference between our machines. I mentioned it before and then forgot about it til now. Your trigger circuit is different than either mine and/or the one used by John for his machine. You're using only 100 ohm branch resistors with a larger pot than either John or myself. We used well matched 470 ohm resistors for each individual transistor and then either a 12 ohm fixed resistor or 100 ohm pot to fine tune the trigger signal. It may be that a better balance between the transistors turning on and off together can be achieved with the larger fixed resistors for each transistor and a smaller tuning resistor.

                      It would be interesting to check the balance of your machine with your scope. Using the run battery ground for the negative scope lead you could place the two positive probes on the collectors of two different transistors and line up the traces to check for any timing or amplitude differences. Then by always using the same transistor for trace #1 you can rotate probe #2 to each of the remaining transistors to check the balance and timing of each to the one used as a reference in trace #1. All the traces should align perfectly.

                      Gary Hammond,

                      Comment


                      • #56
                        Hi Gary,

                        Thanks a lot for all your input, especially the specifics regarding the DVDs, by being so specific I get the impression that you watched them all again yourself…

                        DVDs
                        I watched the 34.1 (completely) and the parts from 33.2 you mentioned. I interpreted what he mentioned about the 21 magnets, that that is just what worked best for him (based the total set of his parameters, wheel, fans, batteries etc) but that might be a different number for you or me. So far I’m inclined to stick with what I mentioned in my previous post, that it is not the magic number 21 that is the key, but the specific setup of magnet in relation to the rest. But we’ll see, I might be wrong about this. Based on my tests with a 20 and 24 magnets rotor, at this point the bottle neck seems to be somewhere else. I will watch the other parts you mentioned too.


                        others, have suggested that even variations in the aether, the weather, the time of day, the phase of the moon, the altitude, and the geographic location can all affect the actual results as well.
                        This might be true, at this point in time there is too much “distortion” still going on to allocate certain performances to any of those. And for that reason I’m only judging performance of my machine when I have at least 2 or 3 cycles with a fairly steady set of parameters.


                        Batteries
                        I certainly do consider the batteries as a potential big influencer on the performance. But I also try to make an assessment with the likeliness of that they are my biggest bottle neck at this stage. Based on the discharged graphs of my post #51, I thought that they didn’t look too bad. And in order to stay close to your parameters of your post #29, I decided that for now I’d stick with the AGMs. However, now that I read that the input running voltage should be around +/-12.5XV… that made me reconsider this. So I’m starting a new series of tests, with one of my new Flooded Acid 24Ah as input battery, dialing my variable resistor in so the input current at start is 1.8A, and having my battery B, AGM 12Ah, as an output battery.
                        Just finished my first cycle: input voltage at the start during running… 12.48V, at the end 12.32V…
                        I’ll run some more tests with these parameters tomorrow, but I’m pretty sure this input voltage will be in the range of this first test, but we’ll see…(which then might indicate something else is still not working as it should).

                        The watt hours required will be less than a cop of 100, while the amp hours can go well above a cop of 100.
                        I’m not so concerned with Wh or Joules yet. First I need to get those Ah at the input down…


                        Resistors
                        Good one… never even remotely considered this… will do the scope measurements you proposed and will report back. Based on the outcome we can then decide whether I should try swapping them for 100-Ohms.


                        My current plan
                        So I plan the following test plan:
                        1) Do some more test with the F.A. 24Ah at the input, AGM 12Ah at the output
                        2a) Do the scope tests you proposed (might do it in parallel / at the same time as 1)
                        2b) possible swap out the resistors and redo 1) and 2a)
                        X) test with new rotor/new magnets 22mm wide
                        X) test with new rotor/new magnets 25mm wide
                        X) test with different battery setup (e.g. 2x 24Ah in parallel at the input, or 24Ah at in put and 24Ah at the output)
                        X) = not sure about the order yet, will be based on what 1) 2a) and 2b) yield.

                        Gary, I really appreciate your input… I remember last week that I was thinking “what possible else can it be that the performance is still not up there…”. Then I was already quite glad with my new rotors/wider magnets plan… but with your input I have a lot still to test/verify with the current rotor(s). You’re really a huge contributor to the development of my SG!!

                        Best regards
                        Rodolphe

                        Comment


                        • #57
                          Gary,
                          What is the wire Gage / strand length on the coil in the video and your coil...? 22Ga?

                          Rodolphe,
                          What is the wire Gage size on the coil you got from Teslagenx...? 18Ga?


                          A 18Ga 100ft Trigger Strand can power up to 48 Transistors. JB did this, I have done this....

                          A22Ga with a lot longer strand length can not power as many Transistors as the 18Ga....

                          Just checking, This may be where some of the differences are....

                          Comment


                          • #58
                            Hi Rodolphe,

                            by being so specific I get the impression that you watched them all again yourself…
                            Yes. I did spend quite a bit of time watching them. And I just now went back and watched some of part 34 disc 4 where John was demonstrating the CG charging mode. He answered Yaro Stanchak's question about amperage draw as "about three amps" and the voltmeter on the run battery showed 12.26 volts. Then as the charge battery was going above 15 volts, the run battery climbed to 12.28 volts. John made a big deal about the run battery increasing in voltage rather than being depleted.

                            In my experience with a longer run time than what John showed, the run battery voltage only goes up at the end of the charge as the current draw drops some. So the voltage and current draw in John's demonstration were closer to your results than to mine. So maybe I misled you a little on the run battery voltage under load. Your voltage looks about right for the amount of current it was delivering to the machine.

                            Gary Hammond,

                            Comment


                            • #59
                              Hi RS,

                              Originally posted by RS_ View Post
                              Gary,
                              What is the wire Gage / strand length on the coil in the video and your coil...? 22Ga?

                              Rodolphe,
                              What is the wire Gage size on the coil you got from Teslagenx...? 18Ga?


                              A 18Ga 100ft Trigger Strand can power up to 48 Transistors. JB did this, I have done this....

                              A22Ga with a lot longer strand length can not power as many Transistors as the 18Ga....

                              Just checking, This may be where some of the differences are....

                              My coil is 7 x 130' 20Ga power windings with 1 x 130' 23Ga trigger winding. And I'm pretty sure the coil in the video was the same as mine.

                              Gary Hammond,
                              Last edited by Gary Hammond; 08-31-2020, 04:47 PM.

                              Comment


                              • #60
                                Hi RS,

                                Rodolphe, What is the wire Gage size on the coil you got from Teslagenx...? 18Ga?
                                8x20AWG+1x23AWG 130' (TeslaGenX).

                                Regards,
                                Rodolphe

                                Comment

                                Working...
                                X