Use for the Tesla Switch

[Murlin]

load voltage very important to charging

I was using a 3.57 volt light to charge. If I don't go through the correct starting sequence when I turn on my commutator the light will blow.

Well I blew my last 3.57 and took a bulb out of a flashlight. probably 5 or 6 volt one.

My batteries wouldn't charge then.

I was surprised that even though I was getting 16 17 volts out of the positive side of the batteries,when I checked the test point before the diode on the positive and out of either neg 2 or 4, the highest voltage I could dial in was 12 .6 volts.

So its kind of tricky getting the potential to go in the direction you want it to.

I will go back to the Shack and get another assortment of bulbs at a lower voltage and see if I can get back to charging mode. I wish I knew how to figure that out without trial an error, but I guess there are always many ways to get from A to B.


On the subject of measuring voltage in the batteries with a volt meter.

I am not sure if this will give us the most accurate picture of what is going on in our batteries.

I think the SG test must be performed on each cell.

It's going to be a PITA but I am pretty sure I have been getting some false readings and might not be charging as much as I thought.

What makes me think that is that that bulb really pulled down the volts real quick in my batteries.

On the flip side though it could just be confirming that I was getting a charge previously.

6 of one, half a dozen of the other.....


regards,

Murlin

[Bit's-n-Bytes]

Thanks Bit,

I am sorry I don't have any CAD right now so I painted your circuit for my test circuit. So sorry.

I didn't change anything as possible asI can do and I tested.
The main difference is the signal, you using PIC and I using 3524(and 494).
Maybe there is a difference-ground, you using B1 and I using another B5
I tried a variety ways like serial and parallel to match Q1 and Q9.

Bit,
I don't complain you. I really wish to learn your method to make PNP closed on
with open bias.
Your method really makes a circuit simple further for another applications.
I will keep on.

JANG

JANG, Thanks. Here's what I can tell you, change the load to 1157 bulbs, place a DMM across batts 1 and 3 and see what the voltage rises to in the first half cycle, then repeat for the batts 2 and 4. My fear is the 3524 is not able to oscillate in the frequency range that we need to see. I am seeing the best results when at 1.6 (800 milliseconds per side) or just under 40 pulses per minute. I know Vtech was able to make this work but with great effort.

Hope this helps

Bit's

[nvisser]

Picaxe

Bits
Do you connect from your pc to the micro controller via the serial port or usb. What cable do you use?. I dont think the newer computers have serial ports anymore
Then all that is needed is the software?
Can it be as easy as that?
And of course the commands. But I`m sure you guys can help if I dont know how to change parameters
Vissie

[Bit's-n-Bytes]

Bits
Do you connect from your pc to the micro controller via the serial port or usb. What cable do you use?. I dont think the newer computers have serial ports anymore
Then all that is needed is the software?
Can it be as easy as that?
And of course the commands. But I`m sure you guys can help if I dont know how to change parameters
Vissie

Vissie, It's that easy. I use a straight thru RS232 cable. The program can be down loaded from here PICAXE. This has a chip simulator that is also nice. Manuals are included in the software under the help menu.

Hope the helps.

Bit's

[Inquorate]

Update

Ok, pwm is working. BUT my oscilloscope is on the blink, and when i can momentarily stabilise the waveform, i'm not getting the 'h', i have to figure out how to stabilise unwanted ascillations.

More to come

[JANGYD]

More Question

Thanks Bit,

My 494 runs under 2sec pulsing and ordered PIC will be delivered nextweek.
I think I can get rid of slow freq and timing problem.
My only concern is how to make PNP closed on when it opened bias.
I have some 1157 bulbs for load.

Do you mean the voltage rises on DMM between bat1 and bat3 is when
1)Q1 is pulsing on and Q2 is pulsing off
2)Q10 and Q9 is alway OPENED(in my case its Opened, yours Closed)
3)Closed path is only to load(1157bulg)

Is that right you want to check the voltage at the moment above?
Next morning I will check and post.

JANG

JANG, Thanks. Here's what I can tell you, change the load to 1157 bulbs, place a DMM across batts 1 and 3 and see what the voltage rises to in the first half cycle, then repeat for the batts 2 and 4. My fear is the 3524 is not able to oscillate in the frequency range that we need to see. I am seeing the best results when at 1.6 (800 milliseconds per side) or just under 40 pulses per minute. I know Vtech was able to make this work but with great effort.

Hope this helps

Bit's

[Bit's-n-Bytes]

Thanks Bit,

My 494 runs under 2sec pulsing and ordered PIC will be delivered nextweek.
I think I can get rid of slow freq and timing problem.
My only concern is how to make PNP closed on when it opened bias.
I have some 1157 bulbs for load.

Do you mean the voltage rises on DMM between bat1 and bat3 is when
1)Q1 is pulsing on and Q2 is pulsing off
2)Q10 and Q9 is alway OPENED(in my case its Opened, yours Closed)
3)Closed path is only to load(1157bulg)

Is that right you want to check the voltage at the moment above?
Next morning I will check and post.

JANG

JANG, Q9 and Q10 are the key to getting the voltage to rise over the battery voltage in respect to the side that is "recieving" the charge. For example when Q1 closes Q10 provides a charge path and therefore should be closed. Q9 should also be closed and we should see the voltage building across batts 1 and 3. when this voltage reaches 14 - 18v, this is where we want to "open" the path Q9 is in, "to relieve the pressure". This process is the same for the opposite side, except Q10 becomes the "pressure relief".

Hope this helps.

Bit's

[Bit's-n-Bytes]

Bits
Your explanation does not make sense to me at all.
If you scope over the two series batteries you will see that the voltage rise to 24v immediately in about 2 usec, as fast as your opto`s rising edge.
Now if Q10 is normally closed as you said the voltage will drop over the two parralel batteries but it will also feed back through Q9 (if it is normally closed) to batt1 positive (which should not be a problem) and also to Batt3 positive which will put a dead short over Batt 1 with only 2 transistors and 2 diodes in the path.
Was that not the reason for Q9 and 10 to immediately stop that short path the same moment when Q1 swithed?

Vissie, I'll try and explain;

Was that not the reason for Q9 and 10 to immediately stop that short path the same moment when Q1 switched. (That is the reason but it would not quite be immediate, but you are on the right track).

You probably could use diodes as we tried in the earlier versions now that we have total switching control. These would have to 15amp or better and be fast switching though.

Hope this helps

Bit's

[ldissing]

Interesting thing

So, I was trying out the Bit's simulation and my batteries continued to lose potential, so I thought, there is still one difference. I'm was using the original TS version with six transistors, but my top and bottom transistors are together on a different signal from the series transistor.

So, I thought...well you know, Bit's is not using any transistor on the bottom of the TS, so I'll just hook it up so those bottom transistors are not in the picture, but everything else is.

My digital starts by just setting each signal for 1 second then turns off, and sequences through each of the 4 signals (I have leds on them, so I can see them fire)...them bouncing back and forth as normal. So, when I fire it up this time and ONLY the series transistor is fired on the LEFT side....BOTH lights light up. When the left top transistor is turned on, nothing lights up, then when ONLY the series transistor on the RIGHT is fired, BOTH lights light up, and of course, nothing when only the top transistor on the right side is turned on.

So, I thought, Bits is incorrect in thinking that the PNPs are turned on with no signal applied. I have no signal on the top transistors and only a signal will turn on the NPNs on the top, because they need that collector voltage through the opto to turn on. Now, I'm not a goin' to say that Mr. Bits is incorrect, but I'm just telling you that with NO signal going to the top optos, both lights turn on when you do not have a bottom transistor.

Now, my switching is turning one side off and one side on with a single stoke of the processor, and I'm pulsing the top transistors too (but a little later) than when I flip the series side on, but I'm not sure it even needs to be pulsed...but that is what I'm doing. I am not flipping on the opposite side transistor on the top currently.

With my setup I can light a 50W lamp up, a 20W to full brightness, six volt bulbs without blowing them and I'm getting some charging with the six volt bulbs. I'd get more with 4 volt bulbs, I'm sure, and I'll check that out after I can get to RS. AND, the transistors do not get hot.

Now, the most interesting thing about this is....the voltages look exactly like they looked on the tesla switch with relays. When I look over those batteries with a volt meter, I can see about 17.4 - 18.9 volts over the batteries. And, depending on my time delay, I can get it to sequence from different low settings (17.4-18.1) up to the high side (18.2-18.9). It is very interesting to watch. It goes from 18.1 lets say in increments of .1V up to 18.9V, then back down in .1V increments to 18.1 again over about a 6-10 second time span. If I wait a little while between switching sides, I can get it to go to 0.2 volts just like Bits can, and when only one series side transistor is on and nothing else, I get 23-25 volts over the batteries. The transistors on the top, seem like they are doing something, but I don't know yet what that is since my scope probes are fried.

The transistor on the bottom is stopping the current from flowing both directions, not the one on the top, IMHO. Maybe this is the correct switching that JB was speaking of, but JB is out to lunch on us....he is busy working three jobs. He'll be back and put a rest to all of this I hope and soon.

Leroy

[JANGYD]

PNP closed on

Bit,

I undertstand what you mean by the switch Q9 and Q10.

What I want to ask you is "HOW Q9 and Q10 with PNP CAN BE CLOSED ON
WITHOUT BIAS OPEN like your circuit to make paths'
Timing and sequence are next problem.
All your explenations is under assumption that Q9 and Q10 is closed on
and that make two parallel paths.
and Then Q1 is puling and so on.

I think you know what my question is.
JANG

JANG, Q9 and Q10 are the key to getting the voltage to rise over the battery voltage in respect to the side that is "recieving" the charge. For example when Q1 closes Q10 provides a charge path and therefore should be closed. Q9 should also be closed and we should see the voltage building across batts 1 and 3. when this voltage reaches 14 - 18v, this is where we want to "open" the path Q9 is in, "to relieve the pressure". This process is the same for the opposite side, except Q10 becomes the "pressure relief".

Hope this helps.

Bit's

[Murlin]

So, I was trying out the Bit's simulation and my batteries continued to lose potential, so I thought, there is still one difference. I'm was using the original TS version with six transistors, but my top and bottom transistors are together on a different signal from the series transistor.

So, I thought...well you know, Bit's is not using any transistor on the bottom of the TS, so I'll just hook it up so those bottom transistors are not in the picture, but everything else is.

My digital starts by just setting each signal for 1 second then turns off, and sequences through each of the 4 signals (I have leds on them, so I can see them fire)...them bouncing back and forth as normal. So, when I fire it up this time and ONLY the series transistor is fired on the LEFT side....BOTH lights light up. When the left top transistor is turned on, nothing lights up, then when ONLY the series transistor on the RIGHT is fired, BOTH lights light up, and of course, nothing when only the top transistor on the right side is turned on.

So, I thought, Bits is incorrect in thinking that the PNPs are turned on with no signal applied. I have no signal on the top transistors and only a signal will turn on the NPNs on the top, because they need that collector voltage through the opto to turn on. Now, I'm not a goin' to say that Mr. Bits is incorrect, but I'm just telling you that with NO signal going to the top optos, both lights turn on when you do not have a bottom transistor.

Now, my switching is turning one side off and one side on with a single stoke of the processor, and I'm pulsing the top transistors too (but a little later) than when I flip the series side on, but I'm not sure it even needs to be pulsed...but that is what I'm doing. I am not flipping on the opposite side transistor on the top currently.

With my setup I can light a 50W lamp up, a 20W to full brightness, six volt bulbs without blowing them and I'm getting some charging with the six volt bulbs. I'd get more with 4 volt bulbs, I'm sure, and I'll check that out after I can get to RS. AND, the transistors do not get hot.

Now, the most interesting thing about this is....the voltages look exactly like they looked on the tesla switch with relays. When I look over those batteries with a volt meter, I can see about 17.4 - 18.9 volts over the batteries. And, depending on my time delay, I can get it to sequence from different low settings (17.4-18.1) up to the high side (18.2-18.9). It is very interesting to watch. It goes from 18.1 lets say in increments of .1V up to 18.9V, then back down in .1V increments to 18.1 again over about a 6-10 second time span. If I wait a little while between switching sides, I can get it to go to 0.2 volts just like Bits can, and when only one series side transistor is on and nothing else, I get 23-25 volts over the batteries. The transistors on the top, seem like they are doing something, but I don't know yet what that is since my scope probes are fried.

The transistor on the bottom is stopping the current from flowing both directions, not the one on the top, IMHO. Maybe this is the correct switching that JB was speaking of, but JB is out to lunch on us....he is busy working three jobs. He'll be back and put a rest to all of this I hope and soon.

Leroy

3 amps and change while charging....nice

regards,

Murlin

[Bit's-n-Bytes]

Bit,

I undertstand what you mean by the switch Q9 and Q10.

What I want to ask you is "HOW Q9 and Q10 with PNP CAN BE CLOSED ON
WITHOUT BIAS OPEN like your circuit to make paths'
Timing and sequence are next problem.
All your explenations is under assumption that Q9 and Q10 is closed on
and that make two parallel paths.
and Then Q1 is puling and so on.

I think you know what my question is.
JANG

Sorry JANG, I do not understand what your asking.

[Bit's-n-Bytes]

So, I was trying out the Bit's simulation and my batteries continued to lose potential
Leroy

Leroy, does "simulation" mean your are just running this with simulation software or are you actually running this on a real D-TS?

Thanks

Bit's

[ldissing]

Leroy, does "simulation" mean your are just running this with simulation software or are you actually running this on a real D-TS?

Thanks

Bit's

Simulation was the wrong word. I AM running the switch. A simulation wouldn't mean crap....you got me on that right? BUT, I do not think it was a "good" simulation, because I still do not believe your PNPs are normally on. So, I've got some code to write to "simulate" what you are doing. Doing that now. Upper transistors "off", and then turning "on" the other transistor on the top side for 1/5th of the time.

Leroy


Leroy

[citfta]

Questions for Leroy

Hi Leroy,

Can you post a schematic of how you have changed your circuit? I have a copy of a circuit you posted a while back that shows two bridges and six transistors. I am not sure if this is the same one you changed to try to simulate Bit's circuit or are you talking about another one. Also I am confused about which transistors you are actually turning on. And one last question if you don't mind, what size batteries are you using, because 3 amps or so is way more than I think I could safely pull from my 12volt 7ah batteries.

Thanks for any help you can give, Carroll