Use for the Tesla Switch

[Bit's-n-Bytes]

Sorry guys, I've been away from test bench due to "professional development" Waiting for 40A SS relays and slowly populating Pulsinator board (according to Bit's schem. ). I need to read back all pages I have missed and catch up since you've made some interesting discoveries when I was absent This is great place


Vtech

Vtech, welcome back my friend.

Bit's

[Bit's-n-Bytes]

Where did we all go....?

Anyway..
I have been running my switch with all the components planned except the generator. So far.
I have 6 115 amp hour deep cycle's. 3 on each side to make a 24 volt potential. They are just off the shelf. I have not even put a charge in them.
The motor I have been running with a mechanical load to force it to pull 15 amps at 24 volt.
I have made runs on the setup. First was about an hour I didn't record it but it was good. The second was an hour, I did not loose anything after a 3 hour rest of the batteries.

Today I ran 8 hours and rested 4. NO LOSS.

Left Side
Start
A. 12.71
B. 12.68
C. 12.72
Finish after rest
A. 12.71
B. 12.69
C. 12.72

Right Side
Start.
C. 12.78
D. 12.78
E. 12.64

Finish
C. 12.78
D. 12.79
E. 12.62

The batteries do not climb above about 12.90 at any given time. I was running 25 seconds each way.
I am going to do another through the night for a total of 24 hours and see if any consumption shows up.

At this rate I do not need to add charge to the batteries, but even if the totals were to start to fall off, after eight hours the amount of supplement power needed to maintain these batteries charge is easy to come by.
I am going to get the generator mounted up. I want to max the batteries out.
Then start pulling other loads in addition. An inverter or some other batteries for charging.

I have also started to purchase all the parts I need to build a version with the Solid State Relays.

If anybody is interested in helping, I need a PCB designed and few bugs worked out. PM me with reference. Payed or Bartered Job.

Cheers
Matt

Nice job Matt

Bit's

[nvisser]

Matt
I can do PCB artwork for you but it will be on pcb express. You just have to download the small program and then print 2 copies with a laser printer on transparencies and put the to on top of each other to get a nice dark transparency.
Mail me the diagram if you are interested. No charge involved.

[blackchisel97]

It's been awhile...

Sorry guys, I've been away from test bench due to "professional development" Waiting for 40A SS relays and slowly populating Pulsinator board (according to Bit's schem. ). I need to read back all pages I have missed and catch up since you've made some interesting discoveries when I was absent This is great place


Vtech

[Matthew Jones]

More Good News

Where did we all go....?

Anyway..
I have been running my switch with all the components planned except the generator. So far.
I have 6 115 amp hour deep cycle's. 3 on each side to make a 24 volt potential. They are just off the shelf. I have not even put a charge in them.
The motor I have been running with a mechanical load to force it to pull 15 amps at 24 volt.
I have made runs on the setup. First was about an hour I didn't record it but it was good. The second was an hour, I did not loose anything after a 3 hour rest of the batteries.

Today I ran 8 hours and rested 4. NO LOSS.

Left Side
Start
A. 12.71
B. 12.68
C. 12.72
Finish after rest
A. 12.71
B. 12.69
C. 12.72

Right Side
Start.
C. 12.78
D. 12.78
E. 12.64

Finish
C. 12.78
D. 12.79
E. 12.62

The batteries do not climb above about 12.90 at any given time. I was running 25 seconds each way.
I am going to do another through the night for a total of 24 hours and see if any consumption shows up.

At this rate I do not need to add charge to the batteries, but even if the totals were to start to fall off, after eight hours the amount of supplement power needed to maintain these batteries charge is easy to come by.
I am going to get the generator mounted up. I want to max the batteries out.
Then start pulling other loads in addition. An inverter or some other batteries for charging.

I have also started to purchase all the parts I need to build a version with the Solid State Relays.

If anybody is interested in helping, I need a PCB designed and few bugs worked out. PM me with reference. Payed or Bartered Job.

Cheers
Matt

[Matthew Jones]

Check the batteries in your meter. If it keeps happening switch them to a lithium ion battery.

More than likely the breadboard just screwed up some kinda way.

But this mechanical relay version I am running right now will suck the juice out of alkaline batteries through mid air, if I am switching any faster than about 1 seconds.

I don't know why. I just switched the batteries to lithiums.

Matt

[ausev]

Back again

Hi All,

Hope you are all well and life is treating you good.

I have been away for a while with my day job keeping me busy but it is good to see the work on this continues. Congrats Matthew on getting the relay setup working.

I had a really wierd thing happen today.

I was testing my circuit and my multimeters were sitting near it but the leads were NOT connected to it yet one of my meters maxed out on 1000 DCV and the other on 50 ACV went to 50.

The other wierd thing is I am using a breadboard circuit and the metal base of the breadboard gave me a nasty electric shock.

The air in my study also became electric.

I am thinking this is some form of induction or wireless transmission of energy going on.

Then all of a sudden everything just stopped working. Now trying to debug my circuit again.

What gives?

Regards,

AusEv

[nvisser]

Solar panel input

I connected my two solar panels in parallel. A 10W and a 14W.
Together they charge a 30 000uf cap up to about 20V.
The 'simple" switch runs on only the solar panels today and even with lots of cloud cover and the occasional sun it clicks away happily.
With a 16v zener it switches when the cap bank (90 000 uF) reach 16V and dumps it to the battery bank.
I replaced the parallel cap on the relay coil with a 47uF to keep the cap bank charge time to the minimum to fill them.
When the sun comes out it clicks twice a second. With lots of cloud cover it clicks about every 5 seconds.
The input current on the panel meter measures just over 1A with each pulse. The output on the Suzuki ammeter is about half of what it was with the 26v psu on the I/P.
As I said the Suzuki ammeter responds very quick . It goes to about 10A with the very short pulses.
The O/P pcb tracks does not run hot with this setup.
I think this is where I will stop experimenting with other configuration of the solar charger. This simple arrangement works best for me. Later it can be updated to solid state with the bipolar switches.
I have to say that there are no sparking whatsoever on the relay contacts.

[nvisser]

Input vs output

Another reason why I thing there is some gain on this switch is the temperatures on the input and output tracks of the PC board that contains this caps and relay.
The track are 2mm wide. The input tracks are cool to the touch while the output tracks burns my finger.
Makes you wonder.

[nvisser]

Back to basics

This afternoon I decided to go back to straightforward parallel caps. No switching from series to par. and no radiant inputs
I used a 24v power supply as input for my pulsar.
6 x 15 000uf caps in parallel. (90 000uF)
Switched by relay every 1 second.
This is as simple as it can get. Charging a 12v battery with a 23v pulses.
I sorted the panel meter problem out, But I think they respond a bit slow to really show what is happening.
On the input side it pulse at 7.5A. At the output it pulse 7A.
I have a cheap Suzuki motorbike ammeter that I got with a Wouter hydroxy cell. This thing can go plus, minus 30A and it responds very quick.
It obviously overshoots a lot but when I connect it on the input side it pulse up to 15A. When I connect it to the output side it pulse up to 30A. Double the input current!
I think this meter shows what really happens with this pulsar as you cannot see it on the normal panel ammeter.
The charging up of the caps takes a while , while the discharging into the battery is very quick.
Anyway the charging is impressive.
Tomorrow I will wire my 2 solar panels in parallel to a 30 000uf cap and use that as input. Together they should deliver about 1.4A @ 17V.
Then I will get a better idea of what this pulsar can do.

[nvisser]

Crash and burn

Leroy
I think the "1A at DC" was the spec for the safe operating area of the transistor at the specific voltage you are using it.

I have to report the the 7A pulses are not true. They supplied the wrong shunts with the ammeter. I think it is more like 1.5A
I will have to calibrate it with some constant loads and my digital meter. I do have 5 shunts now that I can use to get the correct combination.
Anyway i still think that the charging are good and I get more out of them after every charge

[ldissing]

High I think. I use 2 supplies.One to drive a SSG from a 24V supply and the other one is a huge 18V ac transformer and bridge driving a SS joule thief circuit that runs quite hot. At this stage I only try to get enough radiant to try and get the charger part to work correctly.
But I will measure it some time.
I finished a 9 wire coil yesterday. One trigger and 8 drive. Still busy with the pc board that must contain the 8 transistors and other components. That should give me enough radiant energy and a good indication off how much I need and how much power it draws. I will use it in a solid state mode as I do not have a large wheel yet. I did buy a lot of ceramic magnets though.
Need more time!

I measured the current drawn. You must remember the ssg and joule thief are adjusted to give max output and not for efficiency.
The ssg draws 16.4W. 16.6V x 617mA
The joule thief draws 18.12W. 21.45V x 845mA
Together they draw 34.5W continuously.
The pulsar discharge about 7A @ 19V every 1 sec. That is not continiously so I dont know how to compare it.
JB's solar charger draws current from the supply in pulses . The same time intervals as that the charger supplies.
So he probably don't even use radiant generators.
I don't know how else to charge series caps up to a high voltage.

Vissie,

I'm not going to say this is correct, but I would compare it in this way if I were doing it, and maybe this will make sense to you. Maybe it is crap.

I assume you are measuring the current with analog meters, but in essence the analog meter will measure the average current through the system when measuring the SSG, for instance. So, since that is a constant train of pulses, then the average makes sense and you can use that number off the meter. The problem comes in with the big pulse every one second which only lasts for an instant (I assume). It isn't giving you the instantaneous current on the meter, it is the average current, so it should be greater than 7A that you are sending into the battery. I'd think you need to put a 1 milli ohm or so resistor in there and measure the voltage across that very small resistance on a scope. You'll still have to approximate the current and power, but you'll get a better idea of the actual current and power passed into the battery and how long that takes. So, I'd measure the length of time for that pulse and then take several points along that scope trace and approximate the total "current" and the "power" over 1 second or however long it takes to get another "pulse". (The area under the curve is the total current (V/R)). At least you have an approximation of the "instantaneous" power out and can calculate as to what it would be over 1 second. Then you can compare apples to apples, at least this is what I'd do.

Of course, you know the Joules contained in the caps too (from starting V to battery V), so if you just knew how long it takes to discharge the cap into the batteries and the length of time from one pulse to the next, you could calculate average power delivered too in 1 second.

I think I know what you are trying to do, but I'm not sure it is the energy transfer that is important, but the way in which it is delivered, i.e. in impulses.

The following has nothing to do with what you wrote Vissie, but falls in line with the SSG and some things that JB has written.

The following are my thoughts on what JB has written in this forum and in other places, not what JB has actually written. So if I say something that sounds like JB wrote it, take it as my opinion (with a grain of salt) as to what he wrote:

I believe that JB has written, "You need 1 amp at DC." Well, in your case above, the SSG is only "using" .617ma, then I would have originally thought...I need more current going through there, I need at least 1 amp? But I don't think JB is talking about the average current, I believe he is talking about the current during those pulses. So, if the pulse is only on for 25% of the total time, then the actual current through the system during that on time is 0.617 / 0.25 (in the case you mention above). If the pulse is on for 50% and off for 50%, then it is 0.617 / 0.5 or 1.234A. You have made at least 1 amp at DC in either of the two cases above. So you are meeting JBs 1 amp at DC. I am not sure why I thought about this in regards to your post or even why I felt the need to post it, but there it is anyway.

Leroy

[nvisser]

Whats the Load on the primary power supply?

Matt

High I think. I use 2 supplies.One to drive a SSG from a 24V supply and the other one is a huge 18V ac transformer and bridge driving a SS joule thief circuit that runs quite hot. At this stage I only try to get enough radiant to try and get the charger part to work correctly.
But I will measure it some time.
I finished a 9 wire coil yesterday. One trigger and 8 drive. Still busy with the pc board that must contain the 8 transistors and other components. That should give me enough radiant energy and a good indication off how much I need and how much power it draws. I will use it in a solid state mode as I do not have a large wheel yet. I did buy a lot of ceramic magnets though.
Need more time!

I measured the current drawn. You must remember the ssg and joule thief are adjusted to give max output and not for efficiency.
The ssg draws 16.4W. 16.6V x 617mA
The joule thief draws 18.12W. 21.45V x 845mA
Together they draw 34.5W continuously.
The pulsar discharge about 7A @ 19V every 1 sec. That is not continiously so I dont know how to compare it.
JB's solar charger draws current from the supply in pulses . The same time intervals as that the charger supplies.
So he probably don't even use radiant generators.
I don't know how else to charge series caps up to a high voltage.

[Matthew Jones]

Whats the Load on the primary power supply?

Matt

[nvisser]

It looks like it is working as expected.
I only used 4 x 15 000uf caps. When in series they are only 3750uF and they charge up very quickly to about 75V when the relay switch and dump the then parallel caps (60 000uF) with about 18V to the batteries.
It dumps twice every second.
Thank you Hob for the diode circuit idea!!

I received a 15A panel meter today. The setup as above pulse at 8A every 0.5 to 1 second. When I install another zener to make it switch at a higher voltage like 120V, the amp reading goes up to around 10A, but of coarse it switch much slower.
Then I removed two caps and only left 2 x 15000uf in circuit. It charged so fast up to 120v in series that the relay started to switch very fast with about no energy passed to the batteries.
The next step will be to add two more 15 000uf caps. 6 in total and see what the results will be.
I am looking into the par/series coil setup that Hob posted to try and run 2 or more SS SSG coils like that to see if I get faster charging on the series caps.
This will take some time but I will be back with results.
I think that my batteries are conditioned now but will do it a bit longer.
I recieved my PICAXE today The second post from the UK only took 8 days.
So I think when I start testing on the full TS again I will try to see if my batteries last any longer than when in parallel when driving a small inverter and 11W cfl bulb.
I will leave the self charging bit till we get guidance again or until Leroy cracks it! Way to go Leroy!
To get the ser/par cap bank to deliver enough energy from a 17V, 3A solar array to deliver 10A pulses to a battery bank I need a lot more of clever circuits to deliver good strong radiant pulses ... I think.