I rewound those 2 coils with 64 feet of 30 gauge wire each and when I turned on the power supply it didn't go into the fail safe mode, it produced a pretty strong magnet too, however I only had it on for roughly 5 seconds before I turned it off, I had the voltmeter held on the terminals and it held at 13 volts. (I'm about to go off memory here so bare with me ) the total resistance for the 2 coils (in parallel) is 3.7 ohms which by doing the math is still 3.5 amps ( I tested it with the ammeter and verified it was 3.1~), still too much, I really want to get it down to less than 1.5 (like you suggested) So with my new found information I'm going to calculate the resistance and wattage needed to get the proper current flow.

Question: Can I turn the power supply down to 11 V and still have the logic portion work correctly? or is the circuit pretty much set for 13 V? The reason why I ask is by turning down the voltage, I wouldn't need as big of resistors.

Other news: I am almost done with soldering the board, I have all the components on there and I have about half of the jumper wires done on the backside. Let me first say, I should’ve used a bigger board but if this works and there’s a next time, I will opt for a much larger circuit board.

Once everything is wired, I will look it over with a magnifying glass to spot any possible shorts and then I will do a battery of continuity tests to confirm proper connection.

After that I will be asking for your help to hook it up to power and then start testing the logic side with the multimeter (before the magnetic coils are hooked up) so that I can dial in the trim pots to the right settings. I haven’t given up on acquiring an oscilloscope, all my leads have ran dry but I have people looking still.
J.

P.S. Should I request this thread to be moved into the project forum?

 

Yes, bring down the current. It's still too much.

You can adjust the power supply to 11V, Logic and FETs will work with that voltage.

If you need further assistance just ask, we're here.

 

awesome, thank you. I think I have an oscilloscope. I go pick it up tomorrow...
hopefully!!

 

awesome, thank you. I think I have an oscilloscope. I go pick it up tomorrow...
hopefully!!

That's good news!

 

the oscilloscope that we have is for automotive use. before I go talk to the guy, what exactly do I need it for? to test analog signal? how about digital?
we also have the snap-on modis with a lab scope component as well.

im going to read through the literature to see what I can find out.

 

the oscilloscope that we have is for automotive use. before I go talk to the guy, what exactly do I need it for? to test analog signal? how about digital?
we also have the snap-on modis with a lab scope component as well.

im going to read through the literature to see what I can find out.

Almost any scope will do it I guess. If you tell us what could be available to you (model #) we can tell you if it will be suitable.

 

it's a modis eems300. I was plying around with it and it can display info from 20 seconds up to 50 us. it has 4 channels, I can see voltage from 100 mV to 400V, amperage up to 40 amps etc.

if this will work I can borrow this. we have 9 of them in the lab. I never use this portion of it. interesting! !

 

it's a modis eems300. I was plying around with it and it can display info from 20 seconds up to 50 us. it has 4 channels, I can see voltage from 100 mV to 400V, amperage up to 40 amps etc.

if this will work I can borrow this. we have 9 of them in the lab. I never use this portion of it. interesting! !

Nice. It's probably usable. Do you have by any chance a user manual in pdf format? I can't find one.

 

To JCox- didn't you say earlier that you would use proper bearings on the rotor?

In that case the load will be tiny and you won't need much current at all to get that rotor spinning. A couple hundred mA should be plenty.

For testing things like that I use a variable bench DC supply, and a resistor in series with the coil. Then I turn up the supply voltage and watch the ammeter, to see coil current. And obviously test how strong the magnet interaction is at each current value.

If you juts have a fixed 12v DC supply, then get a few 10 ohm 5W resistors and connect them in series, if you like you can put a little croc clip on the lead so you can adjust current with the clip by slecting how many resistors you use.

 

Nice. It's probably usable. Do you have by any chance a user manual in pdf format? I can't find one.

Darn copyrights!! I can't find it anywhere. We may have a digital cop on a disk somewhere, but then there's the challange of uploading it without infringement.

I also found out that I have access to a fluke 123??? I guess it's an oscilloscope. Maybe that one is more widely known and possibly more familiar to folks on this forum??

 

To JCox- didn't you say earlier that you would use proper bearings on the rotor?

In that case the load will be tiny and you won't need much current at all to get that rotor spinning. A couple hundred mA should be plenty.

For testing things like that I use a variable bench DC supply, and a resistor in series with the coil. Then I turn up the supply voltage and watch the ammeter, to see coil current. And obviously test how strong the magnet interaction is at each current value.

If you juts have a fixed 12v DC supply, then get a few 10 ohm 5W resistors and connect them in series, if you like you can put a little croc clip on the lead so you can adjust current with the clip by slecting how many resistors you use.

yeah, there will be a bearing however I am still thinking about the rest of the design, all because of the suggestion of using different colored diodes to show polarity, I am still designing it in my head as to how it'll look and function.

Hopefully tomorrow at lunch I can go and pick up some resistors and get the current down to the right level. Thanks for the advice.

 

Well the circuit board is complete,

Proandevou, can you give me a run down on what I need to do to set everything up? what I'm looking for with the scope, what junctions do I attach the scope to etc etc.

thanks
JC

 

Well the circuit board is complete,

Proandevou, can you give me a run down on what I need to do to set everything up? what I'm looking for with the scope, what junctions do I attach the scope to etc etc.
JC

Do not connect any load yet. Do not connect the MOSFETs to 12V yet.

Check for correct connections on your board , I guess you already did that.

Normally we use an adjustable power supply where we can limit current to the current we expect the circuit to draw. You don't have that.

So first I guess you should check that the power supply voltages arrive at the right pins of the ICs. Remove the ICs from the sockets and power up the circuit with 12V.
measure with a voltmeter the voltage at the power supply pins of each IC, 555, 4049, 4017 etc.
Make sure that the power supply voltage doesn't go to any other pins where it shouldn't (outputs of the ICs).

Then remove the power supply, plug in the ICs into the sockets and power up the circuit.

Connect the ground lead of the oscilloscope probe to 0V. Measure in the following sequence the pulses with the oscilloscope:

- The output of the 555, the frequency should be adjustable with the potentiometer
- each output of the 4017, adjacent outputs you can measure with two probes, only one output will be ON at any given moment
- the output of the 4025 and 4049, their outputs should follow the logic function, 4049 --» inverts the input signal, 4025 --» the output is a NOR function

Now you have confirmed the logic function of all logic circuitry.
(I assume the random part is not yet implemented)

I will write later more, unless someone else does, gotta go to work.

 

Ok, next step.

1. Connect the MOSFETs as shown in the circuit.

2. If you want some extra safety do as follows, if you are confident jump right to step 3. I'm referring to post #22.

Connect a resistor in series with the drain of T7. Something like 1k, it doesn't really matter. Don't connect anything as a load. If one of the MOSFETs is turned on while the other is still on (didn't turn off fast enough) then you will be able to measure a voltage spike on the resistor (12V). If it exists it's not good and we have to do something about it. It will be very short, probably a few 10s or 100s of ns, so you have to trigger your oscilloscope accordingly (no HF or noise suppression).
This is the only MOSFET pair where both MOSFETs are switched with the same signal, "deadtime" is solely determined by the resistor/diode combination at the gates.
If everything is ok remove the series resistor.

3. Connect resistors of 1k to 100k instead of the coils. This will be your initial load.

4. Put an ampmeter in series with the power supply, 20A range. If everything is ok, the ampmeter should be showing near 0A for the next steps.

5. Measure with your oscilloscope between the points where the load would be connected. The signal should go 12V, 0V, -12V, 0V etc.

The power supply you will be using absolutely needs to have a current limiter. When turning on the power keep your eye on the ampmeter.

6. If everything runs fine you can put the coils in.

 

thanks
I didn't have too much timewith last night to mess with it. I checked all the power ports for proper power and now I'm at the point of putting all the ic chips in. I was going over everything one last time and just realized that I didn't ground the unused inputs of the 258. I remember that I was supposed to ask you for clarification because one was marked + and the other was -, I wanted to make sure that both pin 5 and 6 both went to ground???

 

120 ft. of 30ga Wire will give you 12Ω for around 1A of resistance w/DC. (not sure how many turns that is, but put 60 feet worth on each coil if you are running them in series, and they will self-limit the current to just over 1A) 30Gauge can handle 1A for a fairly long period, though it will get warm after the first 20 minutes if run continually.

Don't try AC through it, the cores will saturate and it will get hot.

 

thanks
I didn't have too much timewith last night to mess with it. I checked all the power ports for proper power and now I'm at the point of putting all the ic chips in. I was going over everything one last time and just realized that I didn't ground the unused inputs of the 258. I remember that I was supposed to ask you for clarification because one was marked + and the other was -, I wanted to make sure that both pin 5 and 6 both went to ground???

The correct way to connect unused Opamp inputs is described here.
http://www.maxim-ic.com/app-notes/index.mvp/id/1957

Take any two equal resistors (a few kilo to a few hundred kΩ, whatever you have)

I only now realize that you are also building the random part. If you have a switch between pin 5 and the output of the Opamp, don't leave the pin 5 just open when the switch is opened. Connect a 10nF cap to it.

 

awesome, thank you,

I see that I have jumped the gun with building this, I should have done it on 3 separate boards, one with the random portion, second with the rest of the logic components and the third with the MOSFET's.

Before I plug this one in, I may build a second one giving myself more room and that way I can keep everything separate.

Although, this type of setback may take the wind out of some people's sails, in my case it hasn't, I am learning sooo much. I'm not necessarily looking at the end result, it's more the journey of getting there.

I ask that everyone have patience and allow me to redo everything over again.

Thanks
J

 

praon,
I have built the stand alone random portion and by the end of today I should have the logic portion done.

I still have the 4 channel scope, can I test and adjust the pots on just the random part without harming anything on the circuit?

to clarify, I have everything from the 40106, 40175 and the 258 on this circuit board. the only wire that will connect to the next board will be the output pin (1) of the 258.

thanks again for your wealth if knowledge.

JC

 

I still have the 4 channel scope, can I test and adjust the pots on just the random part without harming anything on the circuit?

Yes, the output voltage on LM258 should be changing in "random" steps as explained earlier (not really random of course)