Tuning Assistance

[maxgoenergy]

This problem of mine specifically refers to tuning the SG so to gain the sweet tune conditions

I will place a video with fotos on soon however for now the problem is that I cannot achieve a sweet tune or point where the current suddendly decreases for a sudden increase in wheelspin. I have tried it all throughout the 1kOhm 3W Potentiometer to no avail. My magnet to coil core distance is about 3-4mm which is what was recommended. My wheelspin time is about 8-10 minutes though I do think the wheel I have is of a average standard

So I have followed the instructions, the principle in setup WORKS(light flickers if output battery off) but just having problems trying to tune it

Some imminent thoughts? Because this problem has not changed even with increasing the wheelspin time and is quite frustrating given the process in principle functions...

[BobZilla]

Hi there.

You are looking for the "sweet spot" but I am guessing you have built your machine in attraction mode which does not really have one. The other way to run the machine is repulsion mode which pushes the wheel away instead of pulling it in to the coil. Repulsion mode had sweet spots because of the physical load of pushing the wheel. Most people use attraction mode these days because it runs much cooler and is generally easier to tune.

[maxgoenergy]

How do you come to the conclusion I have it in attraction mode? Because I have followed the Monopole Instructions and also applied the more recent instructions from the book.

And if you are right how would I get into repulsion mode? What circuitry adjustment is required?

[maxgoenergy]

Or are you talking about running a coil counter to how I currently have it to reverse the effect?

[BobZilla]

As I said most people now a days have their rigs in attraction, that is what has been tought for years now. If you have the book then I believe it is all explained in there anyway. If you want to try repulsion mode then you could reverse all of your coil leads or flip your coils over. The coil is an electromagnet after all which is going to either attract or repulse the ceramic magnets depending on polarity.

Caution if you play with repulsion mode however. It WILL cause heat on your transistors and you run a much higher chance of blowing them out. There are advantages to both ways but you are best off using attraction to start out with.


Another thing worth noting is that there are smaller "sweet spots" in attraction mode but it's not quite the same. From your first post I got the impression you had probably read some old info where people were talking about the sweet spots. Even in attraction mode if you pay attention to your spikes per magnet pass you will find to a lesser degree some sweet spots where you will hit optimal rpm for lowest input. Just as it switches from say 3 spikes to two or from two to one you will have highest rpm for that draw. If you go a little past there you will consume a bit more current and eventually lower the rpm. It's something you get a feel for after tinkering with it for awhile.

[maxgoenergy]

I did a simple test on my Coil and it turns out it is in Attraction mode. This simple test involves placing a small magnet facing North or South on the Core of the Coil then connecting the coil wires to a battery which either does nothing if the coil end is different to the magnet or forces the magnet away if different. From this one can determine if Attractive or Repulsive.

Now NOT having an Oscilloscope how would I be able to find accurately the tune spots on a Attraction mode machine?

[BobZilla]

Get yourself an LED and connect a 300 Ohm resistor to one leg of it. Add some leads, gator clips work good for this. Connect one side to your primary positive and the other side to the junction where the power wind connects to the board which is at the output diode but not the rectified side of it. The output diode which is connected to the collector right where the coil end connects.

Put a white mark in the center of one of your magnets that you can point the LED at and this will be a simple timing light. Mount the LED on your coil in line with the center core. You will be watching when the coil is firing in relation to the core of your coil. When connected one way it will show you the ON period and if you switch the leads around it will show you the OFF period or the spike side so to say. As you adjust you will see more and less firings of the coil and the position where the mark shows up will change. Play around with that a bit and see what you can observe.

*EDIT*
The resistor value is not critical, you just need somewhere around 300. This is only there to protect the LED so as long as your getting it to light up and not burning it out your fine.

[Barrie Castle]

Guys,

It's easy & inexpensive to find the Sweet Spot in either mode using the same method. This method was detailed by John Bedini's instruction set on the Yahoo Groups. You can do it with a multimeter & a laser rpm meter that I got from the old Farnell website for A$29.95 - I'll bet you can get a hold of a good laser one for much less. The old forum test reports used this method almost exclusively. It takes much longer, but saves you rushing out to buy the latest 'scope. What you want is the fastest wheel speed for the lowest primary current draw. Multiply your rpm reading by the number of magnets on Your wheel & divide this by the current reading. Allow about 5 minutes settling time for each potentiometer setting. Then you can post what you get here & compare with everyone else on an even footing. Then, once you've calculated the final "magnets per minute per milliamp" values, it's easy to see which is the highest, so sub in fixed values until you duplicate this "m/m/mA" value.

All The Best.

[BobZilla]

Hi Barrie,
I do remember people measuring up this way and have not heard anyone mention it in a long time. I prefer a light so I can see the pulses but it's good to have options when analyzing these things.

Thanks for your input.

[Barrie Castle]

I tried the pulsating LED once & between that & the spinning wheel, I think I must have been driven unto madness! So the rpm meter saved me from becoming a total nutter (or from becoming more of a nutter than I am today)! It takes me maybe a day flat out to finish running calculations to get the sweet spot, as once I get a good value with one resistance I home in on an ultimate sweet spot even to the point of going ohm-by-ohm. When I get organized & can retrieve my data successfully from my old computer & failed USB sticks (so much for data back-up!), I'll post on here the results for all my SSG's. When I get building again, I'll re-build the lot even better this time.

[maxgoenergy]

Here is the latest. I have tried using the LED technique and(assuming I have wired this all right) on another smaller SSG build I have already made(The newer vertical is still in need of fixing) and it is not helping whatsoever. From 40mA to 140mA there seems to be no point where I can notice any indications for sweet tuned conditions. As I bit by bit turn the Potentiometer the wheel picks up speed in some points and the LED gradually gets to a point where it is permenatley on red. It seems to hit this point at 80mA. There is no part where the current goes a little and then the LED goes from somewhat flickering to a more spontaneous action and the real corresponds in action... It seems to be consistent with only a few periods where from 120mA to 150mA if I turn the Potentiometer very slowly then stop and repeat does it seem to pick some momentum up(though very little) for very small increases on the Ammeter. I am now trying a 2nd Ammeter to see if its an error.

So at this moment I am considering the Laser RPM technique. Thoughts on this all so far??

[Barrie Castle]

That actually sounds about right for what you have to work with, believe it or not!

As noted on this & other forums, the potentiometer is probably one of the worst things you can use on your set-up. I don't think there is a consensus as to why, but it could be that the thin carbon track responds differently (& very strangely!) to radiant energy. Another better explanation is that pots don't like being pulsed & as JB notes recently, there is a lot of pulsing on the SSG input as well as the output.

Where you go from here depends on your end purpose for the SSG.
1. If you want to tune it to get the very best results & experience over-unity figures every time you recharge a battery, you will need to get building some seriously good (& expensive!) resistance substitution boxes as detailed (very poorly!) in my earlier threads on it. I will do a proper video & step-by-step build process, with suppliers & costs when I get back up again - could be months away.
2. If you just want it to restore old batteries or recharge dry cells, or other types, then you can buy a resistance wheel (a very, very basic one!) from the same place as the Laser Tacho.

Unfortunately, the days of just getting a $29.95 quality Laser Tacho from the old Farnell website are well & truly gone. I just checked both RS & au.element14 to find both have discontinued them maybe a couple of years after I bought mine from there. They have got some still, but they're way too expensive.

The quickest way out for a reasonable price is to go to your nearest Jaycar store & ask for their Laser Tacho for $79.95 & get a resistance wheel at the same time for $29.95...
http://www.jaycar.com.au/Test-%26-Me...meter/p/QM1448
http://www.jaycar.com.au/PRODUCTS/Pa...Wheel/p/RR0700

Go there to get everything you need in one transaction. Take the part numbers with you so the salesman doesn't have to spend any time with you, find the bits yourself & ask for a discount. If refused, ask the store manager. If you're not happy with the deal, go back there on another day & try again with different staff or try another store. Alternatively, give them a call first, ask to confirm stock levels & put them on hold behind the counter, making sure you get the guy's name & ask if you can get a discount (don't ask exact amounts over the phone, just an agreement is fine - so you can remind him later).

There's probably nothing wrong with the ammeter you're using, but maybe the way you are using it (more later). Are you using an analogue meter or a Digital Multimeter? Either is fine, just know how to use them correctly on this build, with the proper timed rests, fixed range, etc (as I said, more later!).

The effect you are witnessing with the LED being permanently on, has to do more with the response of the human eye & is somewhat of a limitation with the LED method I didn't think to tell you about before. At a flash rate of about 90Hz (90+ cycles or flashes per second), the eye's persistence of vision will interpret the LED as being continuously lit, when in fact it is just flashing too fast for you to pick up on & is a known limitation of this method. You can work with it & work around it, but I like to do things thoroughly well, using the best equipment within reason, for the best price I can get & take the time to get the results I deserve.

Let me know which way you want to go & I can walk you through things to save time & frustration.

[Yaro1776]

Here is the latest. I have tried using the LED technique and(assuming I have wired this all right) on another smaller SSG build I have already made(The newer vertical is still in need of fixing) and it is not helping whatsoever. From 40mA to 140mA there seems to be no point where I can notice any indications for sweet tuned conditions. As I bit by bit turn the Potentiometer the wheel picks up speed in some points and the LED gradually gets to a point where it is permenatley on red. It seems to hit this point at 80mA. There is no part where the current goes a little and then the LED goes from somewhat flickering to a more spontaneous action and the real corresponds in action... It seems to be consistent with only a few periods where from 120mA to 150mA if I turn the Potentiometer very slowly then stop and repeat does it seem to pick some momentum up(though very little) for very small increases on the Ammeter. I am now trying a 2nd Ammeter to see if its an error.

So at this moment I am considering the Laser RPM technique. Thoughts on this all so far??

Max - a couple of comments on the technique here.
The method Barrie mentions is very useful in determining the so called "sweet spot" bearing in mind that what you are really doing is calculating the firing frequency of the SSG transistors and then applying it to a calc with the primary amps. This does work, but you need to collect the data over a number of coil gaps. You will find that the best rpm numbers show up over a spread of coil gap with a similar calculated sweet spot value. As a side note some low rpm readings will be better than the high rpm reads using this method - just another SSG curveball.

The rpm aspect is an integral part of determining one of the best setups. I prefer low cost/high tech bike wheel speedometers ($10-$12) since they are frequency based and give one a continuous read of the rpm, whereas the laser has to be manually activated whenever you need an rpm read. Relatively easy to calibrate and IMO worth the effort for the real time display and updates.

BobZ's recommendation of using an LED is also favored, because again it gives you a visual in determining the sweet spot and fine tuning. Another tuning tool, if you will. Simple to setup the LED and use a thin strip of Reflective Tape attached to one spoke on the rim connection. Works well and is a great suggestion. Bear in mind that the normal frequency of the LED at high rpm may be at the limit of your visual perception but will show up on the reflective tape as a band whose width expands and contracts with resistance - note the band can expand into multiple spikes. Some play time is required to get the technique right.

Many experimenters prefer to charge their batteries at the lower RPMs for more kick

Back to the method Barrie mentioned. I do not believe that it can be used as a valid comparison across the board for the different builds out there due multiple factors as follows: 1) Each SSG build is different with various wheel bearings and SSG components, 2) All Ammeters are not created equal in accuracy, so the mix of various measurement devices and the ability of the experimenter will come into play with respect to the readings and subsequent calcs, 3) The size and condition of the batteries will come into play here. Undoubtedly there are other influences. So, a useful tool for sure, but there are inherent limitations.

Finally, don't obsess about COP>1 - learn the machine and have fun,
Yaro

[BobZilla]

Very well said Yaro, I agree.

The led is sort of supplemental to anything else you would like to use to monitor and it will only cost you a few bucks. I don't think some people are understanding the point of it but I know Yaro does because he just said it. You can see the width of the marker expand and contract and you can see how many fires per pass you are getting. Two spikes, three, one. The width tells you your on time, or current if you think on it that way. If you do a two led setup like I do with one in each direction you can understand even more about what is happening. For example you can see a certain point where the width of the reverse voltage and the width of the forward voltage is the same, go a bit more and your forward voltage is more than the reverse so it becomes wasteful. It is the equivalent of looking on an O-scope at the spike portion of the wave to see how high and then the length of the horizontal "H" portion.

Now no matter what your method is keep in mind when you take an RPM reading and amp reading that will be only for that static moment in time. As your back end charges up your front end is loosing voltage and not pushing the same current. Your RPM will also increase on a charged back end. Honestly to me most of that matters very little, I look at the end results, did the back get fully charged, where is the front at on the end of the run.

Use anything and everything you can if it helps you, no right or wrong way to observe.

[Barrie Castle]

Thanks for your input fellas,

The bottom line is that a bloke living about 1000 miles from me is having difficulty tuning his builds & is asking for a bit of help based on the experiences of the rest of us who can show we have succeeded. He should be able to ask a specific question & get a specific answer or be given a way to get the answer he seeks. And because, in the original instruction set at least, it shows us how to arrive at a sweet spot and there are detailed files showing how others have succeeded, provided he puts in the time, the money & the effort, he's entitled to get a similar level of success if he follows along step by step & can think about it all later. I've used the original method many times & while it does take a long time, it is very clear in where the best of a number of sweet spots are, then if he wants, he can adjust the coil gap later & make other changes that I can suggest. If he walks the same path as I did, I can help him do this for himself, which has got to be better than turning a pot & measuring an input current & wondering what it's all supposed to be doing or if something is wrong (like a not-so-cheap ammeter).

Magnets per minute per milliamp is a very useful method for comparison across builds, if the variables mentioned are kept as standard as possible & if he supplies me with enough other data I ask for as we go along, like unloaded wheel spin time (later), so I can make like comparisons on mine, after adjusting it to suit. Meter accuracy is pretty tight these days & if we get all the same bits, batteries, etc new from the same brand of shop here in Oz (where, let's face it, there's very little variety anyway). Once I know he has the right technique & comes up with a reasonable pattern of results, I can diagnose clearly & make suggestions for improvements until he's happy that he has done as good as he can or wants to. In the early stages, it's pretty important to get a good sweet spot, even if it takes a few days or a week to do so. Why? Because load testing for 20+ runs can take months! Even in my first 22 runs, I had COP>1 figures after each run in 53% of all runs at the end of each run. If others have achieved over-unity with this simple build & he puts in a similar effort as mentioned, taking into account the experiences of others along the way, he too should be able to achieve the satisfaction of success that his predecessors have enjoyed, if not better.

I reckon he's up to the challenge