Large Coil SS SG

[Floris]

Hi everyone, I thought I post some details about the project I have been working on..

Been working in the shed building and testing a new large coil I made to run solid state (SS), or perhaps rotored later, and see if large is better. The coil, for now, is about 10cm wide x 20cm high, aircore. It has 25 x 130' #18 magnetwire twisted together.
It has been interesting to test it in various configurations and work towards a 24 branche system with 24 transistors all components matched . However, that will need to wait until money permits.

At first I first ran all 24 wires on one transistor, 100ohm at the base and with a diode to 2x50Ah batteries. From the base connected through a diode then two pots (1k), 5K) in series with a cap across, the triggerwire connected to plus. I use no B-E diode in SS mode. Very basic setup that appears to work best. The frequency range I have seen is somewhere around 50 Hz to 25KHz. I have seen good charging at 200ma with 5 or 16 Ah battery. But It also chargers well between 1 and 2A with the right batteries connected and all components stay cool. Surprising.

But OK, currently it runs with 8 MJL's thus 3 strands per transistor. With 200ma per transistor it charges a battery pretty much one to one. That is OK for starters so now I focused my attention to building a large cap dump setup. I will tell more about the setup later as there are more topics that I believe are worth sharing. Things like the "One wire change", batteries, SS tuning and cap dumping.

I'll post some photo's of the setup soon.

Floris

[Floris]

Here is a picture of the large coil setup.

[Floris]

Cap-Diode modification

As said, with this coil I use the cap-diode modification. The capacitor in the trigger circuit has the function of boosting the trigger signal to the base. It will run without a capacitor and charge batteries but choosing the right capacitor for the setup and batteries will really add to the amplitude of the spikes and can increase or decrease the running frequency. My tests with only one MLJ21194 showed me to use quite a small cap like .22uf. It gives the highest frequency and good charging. The drawback I found when further increasing current you start to see ringing on the base just before the cap discharges. Charging is still good but I need to explore that more. My exploration of the coil is far from finished but sometimes the .22uf is to small and 1uf works well. With 8 transistors anything bigger than 4.7uf seemed too big.

I must add here that I have been testing a fair bit with front and back in parallel not in series. That totally changes the behavior of the setup. With the all batteries connected in parallel charging is really strong, however, tuning a nightmare. It is very sensitive to the SOC of the batteries. Input current will cut in half after sometime. Both front and back impedance changes are at work here, I believe. Also duty cycle is much lower, in the area of 20%. The frequency will go up with increased input power, whereas in series config the frequency will go down with increased power (although in other SS coils I have also seen increases, in line with Patrick's builds?).

There is a lot to explore here, admittingly I may have focused too much on the back end of this. It appears that with batteries in parallel the primary is draining much faster than in series config. I need to do the runs and get the numbers but it was certainly different.

Please ask me any question you have. Hope I'm not boring people with sharing my endeavors.

More soon.

[Tom C]

Floris,

run branch circuits like JB did with his forced oscillator video, splitting the output with branch diodes.
Tom C

[BobZilla]

Hi Floris,
You are not boring me my friend so keep posting details. I am glad to see someone else who is playing with solid state, I know many others do but there is far too little discussion around it.

I am about to make a new build this weekend. I have been waiting for my buddy to have some free time to help me manage the coil winding process. My current setup is running as Tom suggested using branches and I agree that it is a great way to go. The problem I can see down the road is when we want to charge large banks. I made a different version of my machine that had four branches and it became a bit cumbersome to hook up all of the leads but I tell you at least if one pops off while under load the others take up the energy so you don't blow transistors.

For the benefit of comparison I will describe my setup. It is setup basically like a normal SG with B2E diodes and all. I have used Patrick's cap diode mod on the back end of the trigger just before the base. The trigger feed side of the coil is moved over with the power coils, just like Patrick describes.

The coil is made from three strands, trigger is 21awg, power are 2 strands of 18 AWG. All three are 200 feet. This comes out to about 1.2Ohm on power and 2.5Ohm on trigger for the natural resistance of the wires.

I am using 750Ohm base resistors and a adjustable mega Ohm pot. I think the mega Ohm is a bit too much because I usually have it adjusted almost at the far end but when turned all the way down the circuit does still oscillate very slowly. I have used a radio to listen to it and put a meter in line to measure the draw. On very low it starts working even before the meter reads any current but charging would seem to take weeks like that. Once I turn down the resistance I get more of a steady tone instead of pops. The cap on th trigger is a 1uf. I experimented with 2.5 and 4.5 but settled on the 1uf.

From my observation the machine does well at two settings, first pulling 120ma does charge but takes a long, long time. Second when I set the draw between 800ma to 1A or so I get faster charging and also the coil will make an audible ringing sound. My batteries are 75AH deep cell's.

I have not played with cap dumping but I would like too. Currently my charge side's go right to the charge batteries. In a branch mode you will see the loads jumping around as they share the energy. Mr. Bedini showed this in his video. They balance out towards the end.

I do not get two for one results with it but I do come close. Charging batteries come out to around 12.4 after they rest. I will run the machine until the primary is reading about 11 volts while under load and then stop.

Here is a picture of the machine I am describing.

[RS_]

Bob,


It is best to NEVER pull battery's down past 12.00V under load...! on these smaller battery's and car/truck battery's at 12V they are effectively dead, and pulling them lower is only hurting them.....

if you stop there, you will have faster charge times, and higher 1 hour to 4 hour standing Voltages like 12.7V and higher......

once a battery has been through many cycles and well conditioned, you can stop at 12.2 to 12.4V, and the charge times will get faster and the battery will start to last as long under load as when you were drawing it down to 12V
and the 1 hour to 4 hour standing Voltages will be in 12.9V to 13.10V+ range and still last under load the same or longer as when you were pulling them down to 12V

after many cycles like this, they will start to operate in this higher Voltage range, witch = smaller and smaller plate crystals, than when being pulled down to 12V or lower

[min2oly]

As said, with this coil I use the cap-diode modification. The capacitor in the trigger circuit has the function of boosting the trigger signal to the base. It will run without a capacitor and charge batteries but choosing the right capacitor for the setup and batteries will really add to the amplitude of the spikes and can increase or decrease the running frequency. My tests with only one MLJ21194 showed me to use quite a small cap like .22uf. It gives the highest frequency and good charging. The drawback I found when further increasing current you start to see ringing on the base just before the cap discharges. Charging is still good but I need to explore that more. My exploration of the coil is far from finished but sometimes the .22uf is to small and 1uf works well. With 8 transistors anything bigger than 4.7uf seemed too big.

I must add here that I have been testing a fair bit with front and back in parallel not in series. That totally changes the behavior of the setup. With the all batteries connected in parallel charging is really strong, however, tuning a nightmare. It is very sensitive to the SOC of the batteries. Input current will cut in half after sometime. Both front and back impedance changes are at work here, I believe. Also duty cycle is much lower, in the area of 20%. The frequency will go up with increased input power, whereas in series config the frequency will go down with increased power (although in other SS coils I have also seen increases, in line with Patrick's builds?).

There is a lot to explore here, admittingly I may have focused too much on the back end of this. It appears that with batteries in parallel the primary is draining much faster than in series config. I need to do the runs and get the numbers but it was certainly different.

Please ask me any question you have. Hope I'm not boring people with sharing my endeavors.

More soon.

Hi Floris,
nice work, glad to see you're back at it. I'm curious about a couple of things.

1. 3 strands/transistor - are the strands in series for 390' each - or do you have them in parallel?
only 800mA very small for so much wire, must be the 1uf :-)

2. Are you using only one #18 130' for the trigger coil?

1/1 w/ solid state is a piece of cake eh. - I'm currently having some very good results w/ RF's large window motor kit. Might start up a new thread if I can ever get it together - basketball season is coming to an end so we might have more time now.

solid state is fantastic for practical purposes - not so fun for show and tell
Kind Regards,
Patrick

[BobZilla]

Patrick I just wanted to say hello and let you know you have been very helpful to me with you video's and information you share. I really appreciate what you do for the community.



RS,
I think with these batteries it is ok. They are true deep cell. Heck I could be wrong but the chart in the manual says c20 rate is 3.75Amp to 10.5 volts. I never really hit the c20 discharge rate, the primary never pulls over 1.5A and I usually keep it below 1A.

I charge the primary conventionally but the charge batteries always charge from my SS. Once they charge I have other equipment that I run loads on and it pulls them down at about 1.25A and I do run them down to about 11 volts also. These are big batteries, I change them out once a day on a system that runs constantly. I have other batteries that I conventional charge for the system too because my experimenter batteries take longer to charge than is sustainable without extra batteries.

Here is the spec sheet for the 75AH I am talking about:
http://www.power-sonic.com/images/po..._12_Aug_27.pdf

These were in fairly new condition when I started using them on my Bedini SS tech. I had purchased them new and been running them from a conventional charger for a few months. I have those 75AH and also two 100AH of the same make.

I just hit a goldmine for batteries though. I was able to get UPS trays that were on their way to the recycle. I got seven trays which when I opened them up they were eight 5AH 12 volts wired up in series and parallel for a bank of 48 volt output. Yea really incredible, 56 of these little babies.

They will need conditioning as you described because these had actually reached their end of life, that's why I was able to get them. They came out of big rack mounted UPS systems for a data center. They look in very good condition, no swelling etc. The first one i checked had been sitting at around 7 volts. I put my small setup on it, not the one described above, and charged it. I have cycled it about 3 times so far and it is showing better capacity each time.

On the very first charge the meter readings were kind of strange. The volts started at a reading of around 27 volts when I started the charge and then started dropping progressively as I charged. I was thinking about why that could be and I think it may have been from the plate buildup holding back the spike a bit from the charger. As the gunk dissolved a bit more of the high voltage was absorbed. Anyway that is just a guess but after that the charge curve was normal looking as it climbed up from 11.8 to 13.5 when I stopped charging. I'm sure some of the others will do that so I will try and get a shot of the chart on that first charge. It almost looks more like a discharge curve but it is not.

[RS_]

Bob,

those are some very nice batterys..... But AGM battery's are not recommended by JB to be charged with a SSG, I don't know about cap pulsers chargers with them though.....

they will work good for the primary and recharged with a standard charger as you have been doing......

on the ssg charged battery's even though they are deep cycle, if you stop at 12V you will have higher standing voltages, and they will start to operate in a higher V range, and the batterys will last a lot longer in service.....

[BobZilla]

Bob,

those are some very nice batterys..... But AGM battery's are not recommended by JB to be charged with a SSG, I don't know about cap pulsers chargers with them though.....

they will work good for the primary and recharged with a standard charger as you have been doing......

on the ssg charged battery's even though they are deep cycle, if you stop at 12V you will have higher standing voltages, and they will start to operate in a higher V range, and the batterys will last a lot longer in service.....

RS,
I appreciate your advise. I will try going a little easier on them and see how it goes. I am aware that the AGM's are not ideal but as it worked out I had already purchased them before I knew anything about this technology. The external system which I am charging batteries for was already in use so I started experimenting with the batteries I had. I had discovered these charging methods in an effort to improve what I was already doing with conventional charging.

Whenever I buy anymore they will be flooded. I think I am stuck with AGM's for awhile now though, my big ones are AGM's and so are all of those salvage batteries I just added to my collection.

I have to say though that the SS systems that I am playing with do decent job of charging them. I'm not sure why the sealed AGM's are discouraged but I do understand that they are not recommended.

I did start a new build this weekend. This one is going to be quite different from anything I have done so far. I am winding coils one strand per coil ( like slave coils on a mechanical) and also the trigger will be on it's own coil. It may or may not work but I am just trying to experiment with different ideas. I'm curious if by changing different orientations of the coils it may help create a additive effect to the over all field. I will start a separate thread about that effort.

Thanks for the support everyone.

[Floris]

Hi Patrick,

Hi Floris,
nice work, glad to see you're back at it. I'm curious about a couple of things.

1. 3 strands/transistor - are the strands in series for 390' each - or do you have them in parallel?
only 800mA very small for so much wire, must be the 1uf :-)

2. Are you using only one #18 130' for the trigger coil?

1/1 w/ solid state is a piece of cake eh. - I'm currently having some very good results w/ RF's large window motor kit. Might start up a new thread if I can ever get it together - basketball season is coming to an end so we might have more time now.

solid state is fantastic for practical purposes - not so fun for show and tell
Kind Regards,
Patrick

1. Currently 3 strands in parallel per transistor.
2. Yes just one wire same as the others.


Yes SS is practical but I just might build a nice rotor and see how that works but that will be in a long time.
Patrick, have you ever been able to get an optocoupler working correctly. I unfortunately seem unable to get the optocoupler cap dump working, really frustrating... Problem is that the optocoupler seems to be latching, i.e. not properly swithing. I wish someone with more experience with this could help.

[min2oly]

Hi Patrick,



1. Currently 3 strands in parallel per transistor.
2. Yes just one wire same as the others.


Yes SS is practical but I just might build a nice rotor and see how that works but that will be in a long time.
Patrick, have you ever been able to get an optocoupler working correctly. I unfortunately seem unable to get the optocoupler cap dump working, really frustrating... Problem is that the optocoupler seems to be latching, i.e. not properly swithing. I wish someone with more experience with this could help.

I've not tried the opto. I did use the SCR/Zener combo and the SCR would lock up. It took me a while to figure out it would mostly happen with batteries that had high impedance. I've always leaned toward the most simple method possible that takes the least amount of energy from the system to run.
This is the best cap dump I've ever used for a SS energizer:

[Catrinisin]

On the very first charge the meter readings were kind of strange. The volts started at a reading of around 27 volts when I started the charge and then started dropping progressively as I charged. I was thinking about why that could be and I think it may have been from the plate buildup holding back the spike a bit from the charger. As the gunk dissolved a bit more of the high voltage was absorbed. Anyway that is just a guess but after that the charge curve was normal looking as it climbed up from 11.8 to 13.5 when I stopped charging. I'm sure some of the others will do that so I will try and get a shot of the chart on that first charge. It almost looks more like a discharge curve but it is not.

Hi Bob,

I also had this same experience charging "dead" UPS batteries from large trays. Using a small SSG with two transistors, I would charge them slowly one by one. At first the voltage was high in the 20's like you said. Then they would come down to normal and subsequent charges looked normal; 11V through 16V when I stop. Yes I said 16V. I've found that charging them all the way to 15.5 or 16 seems to do the best. I have read however that a lot of people stop sooner, like you stated. Is there a reason for this that has escaped me? I just seem to see better results, with "dead" batteries coming back to better than new condition when I charge them to 16V each time. After resting, they come down to about 13.4 or so. When I do C20 rate tests, they always last longer than 20 hours after being cycled to 16V a few times, being drained down to 11V. These are small 9AH 12V batteries from an APC 1500VA UPS in this example.

I've only been at this for a few months so I don't have a lot of experience to lean on here. If I am causing long term damage of some kind, will someone please set me straight?

Thanks!

Here is a pic of my small charger referenced above. I mainly use it to charge small AA's and the like, but it seems to do really well with larger batteries that are on the way to recycle.

[Floris]

Yes I have used the zener/SCR cap dump and still use that often. However, with the SCR you are limited to smaller size caps. With the new large coil I would really like to see and learn how large caps do with larger battery setups. I have some 80A SCR's but these didn't work well, they need too much gate current for the zener, 50ma IIRC. Anyway, they just latched, I would just like to have a versatile cap dump setup to experiment with.

[min2oly]

Yes I have used the zener/SCR cap dump and still use that often. However, with the SCR you are limited to smaller size caps. With the new large coil I would really like to see and learn how large caps do with larger battery setups. I have some 80A SCR's but these didn't work well, they need too much gate current for the zener, 50ma IIRC. Anyway, they just latched, I would just like to have a versatile cap dump setup to experiment with.

Hi Floris,
I meant to imply that maybe the problem is the battery you are charging and not the circuit?

perhaps someone with OPTO experience could ask better questions...
-KR
Patrick