PDA

View Full Version : Arduino pulsing Bedini SSG Coil (solid State)


tonytt98
05-09-2011, 01:15 AM
Newbie here,

I'm playing around with an Arduino switching a SSG circuit. Got it working. I ordered a oscilloscope to help see what's going on and to try to make things more efficient. Got a question:

If you take the back emf spike from the coil, and run it through a second Bedini SSG circuit, what back emf output would you get from the second coil?

I'm guessing higher voltage with less amps.

Would this second emf output be any more efficient for anything?

Thanks,
Tony

nilrehob
05-23-2011, 02:25 PM
Newbie here,

I'm playing around with an Arduino switching a SSG circuit. Got it working. I ordered a oscilloscope to help see what's going on and to try to make things more efficient. Got a question:

If you take the back emf spike from the coil, and run it through a second Bedini SSG circuit, what back emf output would you get from the second coil?

I'm guessing higher voltage with less amps.

Would this second emf output be any more efficient for anything?

Thanks,
Tony

Have you done any testing and measurements yet?
Please post a circuit diagram.

/Hob

erfinder
05-23-2011, 02:51 PM
Newbie here,

I'm playing around with an Arduino switching a SSG circuit. Got it working. I ordered a oscilloscope to help see what's going on and to try to make things more efficient. Got a question:

If you take the back emf spike from the coil, and run it through a second Bedini SSG circuit, what back emf output would you get from the second coil?

I'm guessing higher voltage with less amps.

Would this second emf output be any more efficient for anything?

Thanks,
Tony

Hello,

I don't recommend sending the kickback directly into a second SG circuit, your transistors in the second SG circuit will not like this... instead, use the kickback to charge a capcitor, this capcitor should then be used as the source to a second SG circuit....I have done this countless times, and recommend it highly, not just for running extra SG circuits for the cost of driving the primary, but also other load types....experiment with various loads.

Arduino.....I enjoy simplicity, the addition of this kind of control will ease some of the burdens in more advanced circuits, however with regard to the here and now, in simple soild state and simple rotor based SG circuits, the Ardiuno is useless....if for no other reason, one must carefully watch the SG, as its been cleverly designed to adjust itself to the loading condition! By incorportating Arduino and the likes, we tend to engineer that self regulating feature out. My opinion, Arduino should only be added to the output circuitry!!!

My 2cent...

Regards

barbosi
05-23-2011, 09:40 PM
Hello,

I strongly agree with erfinder on this:

I don't recommend sending the kickback directly into a second SG circuit, your transistors in the second SG circuit will not like this... instead, use the kickback to charge a capcitor, this capcitor should then be used as the source to a second SG circuit....I have done this countless times, and recommend it highly, not just for running extra SG circuits for the cost of driving the primary, but also other load types....experiment with various loads.


and from personal experience I can also recommend the use of ULN2803 IC to be driven by arduino. This is only for the first stage which will charge a capacitor (capacitor used later to drive the second stage as erfinder suggested). You have in this IC all needed 8 (identical) transistors, including the recovery diodes. Depending on the inductor used, you can drive those transistors way faster than a solid state SG circuit, at a lower current. I am all open if someone else would recommend some other parts.

I have no doubt you know already about this arduino project: Arduino playground - RegulatedPositiveVoltageBooster (http://www.arduino.cc/playground/Main/RegulatedPositiveVoltageBooster)
While I find a bit inefficient his method, I will point only to the possibility of regulating the pulses to achieve a certain output voltage.
In addition, with an analog input measuring the charge current using a shunt you may turn off the transistor efficiently when the current reached the max. (Too soon is premature and the kickback is not as big as desired; too late is a waste of energy drawn from battery). However, the analogRead takes around 1 millisecond so I found it not too practical on this configuration.

In conclusion, this is my personal preference as a primary source compared to solid state SG circuit because:
1. No moving parts
2. At the processing speed, one can achieve same results as a solid state SG but WAY faster with lower energetic cost.
3. Possibility of controlling the transistor's turning ON and OFF of according to the needs. That means too adjusting to load conditions but at a lower energetic budget.
4. The use of a control logic like in a "night rider" would improve the thermal condition of the transistors.
5. A lot of flexibility by change in software often not involving changing parts
6. The overall cost of circuitry.

The limitations of arduino appear basically when are needed "floating grounds". As an example, because the digital outputs have the same common ground, it will require a special design to fire a 3 phase full bridge thyristor but a more elaborate design if controlling igbt in the same bridge configuration. The same limitations are for the analog inputs.
Another limitation is the program itself which must be run in a continuous loop with very little room to treat events. In other words is polling style not real time event driven programming. And this is a serious limitation when arduino is used in controlling a complex output circuitry. It cannot handle competing events, it must be a simple crystal clear process with few timing constraints.

PS: As a final note, even I used the terminology "solid state SG" I must say there is no such thing and should not be called that way (but I wonder who would listen). The proper name used in electronic books is "Blocking Oscillator" and there is no doubt in my mind that even John Bedini had study them in school. His clever contribution is the addition of EMF into the coil when a magnet is approaching it, then turn off the transistor when the EMF (hence the current) is at max.