Magnetic Stirrer Project Reset

[eccerr0r]

Re: Magnetic Stirrer Project Reset

I'm glad you see the light and not fooling with the analog crap anymore.

[EasyGoing1]

Re: Magnetic Stirrer Project Reset

i thought it might make changing patterns .

If it does, there is nothing interesting about it ... take a look at the video I posted in the post before this one... you'll see what happens when it gets off track.

[EasyGoing1]

Re: Magnetic Stirrer Project Reset

Yes this is still a motor that converts electrical energy to mechanical energy. Just that the rotor is not attached to the stator housing at all. The proper term is probably a stir plate with no moving parts.

OK, more progress ... I was fairly successful using the DRV8825 stepper driver... the coils get pretty hot over time, ive been able to reduce that by setting the current choke to a lower value but if it gets too low, the pill wont track as well to the coils, but I think it might be a matter of making small incremental adjustments in the current and testing it in the kind of liquid I need to stir.

I also ordered some bigger coils but I'm now thinking that maybe smaller coils would be more suited to the task. Not sure yet ... need to experiment with that.

BUT - here is a video of the latest progress. I have my RPM formula obviously inaccurate because when the pill gets going at a decent speed in this video, I have it calculated at 70 rpm but clearly it's going much faster than that so I need to figure that out ... but it starts out at 10rpm then I turn the rotary and it climbs rpm by 2 per click ... when it gets above 50, it looses tracking until I reach 70 then it locks in pretty good and will stay that way until I turn it off.

BUT HEY! .... PROGRESS!

[petehall347]

Re: Magnetic Stirrer Project Reset

If they were placed inside a cup, I would assume they would just jump around since they don't have a north and south pole to keep them aligned with the coil pairs.

i thought it might make changing patterns .

[EasyGoing1]

Re: Magnetic Stirrer Project Reset

i suppose being called motor less is not really true as it must be a motor of sorts when its working .
this sort of thing wants me to see what happens to iron filings on a sheet of paper when placed over the coils .

If they were placed inside a cup, I would assume they would just jump around since they don't have a north and south pole to keep them aligned with the coil pairs.

[EasyGoing1]

Re: Magnetic Stirrer Project Reset

Here's another tidbit hint that probably can not be applied in this type of motor: Brushless fans have a hall effect sensor to know the position of the fan magnet so it can put the correct polarity near the poles so we don't have vibration instead of rotation when the fan initially tries to spin up. This is somewhat difficult to do with the no moving parts stir plate as magnet position would be difficult to feed back to the exciter.

It's coincidental that you happen to mention this, as earlier today I was looking at videos that broke down and analyzed the driver circuits for those electromagnetic levitator toys. They use four coils exactly like the coils I use, but what they do, is they use hall effect sensors to determine when the levitating magnet gets off center. When it does, the current through the hall effect sensor changes slightly. It is amplified through an op amp then that change is used to offset the current flow in the coil that the magnet is leaning towards and it is added to the coil to push back on the levitating magnet. All of it happens so fast, that the net result looks like a magnet is prefectly levitating about an inch above four coils. You can even spin the rotating magnet and it will spin for a long time given the almost zero friction.

BUT, that's not why I'm writing this response ...

So I started really studying the datasheets for the stepper driver that the EE guy used and most of it made sense, but then I looked at the datasheet for the 8825 driver, of which I have several, and what I learned is that internal to that chip, it maintains a table so that it knows where it is within the stepping sequence, which can be micro stepped down to 32...

There is a table in the datasheet that shows how much current is being applied to each of the two coil pairs at each step in the table. And what I noticed right away is that at step 1, it puts 100% of the current into coil pair A while only putting like 5 to 10% in coil pair B. Then as you step, it decreases the current in A and increases the current in B and it keeps doing this until you have stepped it all the way to the end (32 steps for the 32 mode, 16 for 16, etc.) and it appeared to me that after you have run all of the steps in a cycle, it would have moved the end of a stir bar from one of the four coils to the next adjacent coil - or effectively, a full set of steps would be ¼ turn of the spin bar.

I also saw that when the step pin is held low for 1.8 microseconds then set high, it's on the low to high edge that causes it to step one time.

This gradual transition of current is exactly what I've been trying to do in the first place... SO ...

I ran some calculations to figure out, based on the desired RPM, how much time I would have to wait in-between steps to get a full rotation (which at 32 step mode for example, would be 32 steps - four times for a complete rotation). Using that, I was able to come up with a formula where I could feed the microcontroller an RPM that I want and it will then calculate the delay between steps.

I need to play with the code a little but my first run looked promising ... although a little jerky so I need to figure that out (I think it has to do with the way I'm calling the stepping function that I wrote) ... but I think I'm finally on the right track here...

Mike

[eccerr0r]

Re: Magnetic Stirrer Project Reset

Problem is what the motor should do if the magnet N pole is near the S pole electromagnet and the S magnet S pole is far away from the electromagnet, you're going to be stuck with weird dancing in this situation no matter what and you'll need to manually align the rotor back on track or hope over time random movement will drop it back into proper alignment. I suspect the smaller stir bars the opposite end are closer to the "opposite" polarity electromagnet instead of being far away and have a higher chance of getting back to proper alignment.

Yes this is still a motor that converts electrical energy to mechanical energy. Just that the rotor is not attached to the stator housing at all. The proper term is probably a stir plate with no moving parts.

Here's another tidbit hint that probably can not be applied in this type of motor: Brushless fans have a hall effect sensor to know the position of the fan magnet so it can put the correct polarity near the poles so we don't have vibration instead of rotation when the fan initially tries to spin up. This is somewhat difficult to do with the no moving parts stir plate as magnet position would be difficult to feed back to the exciter.

[petehall347]

Re: Magnetic Stirrer Project Reset

i suppose being called motor less is not really true as it must be a motor of sorts when its working .
this sort of thing wants me to see what happens to iron filings on a sheet of paper when placed over the coils .

[EasyGoing1]

Re: Magnetic Stirrer Project Reset

Yeah, that's how it should be done, digital on/off signals driving the coils, not with analog signals driving the coil and certainly not analog signals driving MOSFETs to drive analog to the coils...

So an interesting discovery ... I found a youtube video of a guy who is an EE who SUCCESSFULLY made a motor-less magnetic stirrer using what we all assumed would be the best driver - a stepper motor driver, specifically an A4983 which I've never messed with. I have tried using the 8825 and you mentioned the L298 H-Bridge, which I have also messed with.

Here is one of his videos that he made to show how the stir bar cant get back on track if it gets decoupled - something he was talking about in a hobbyist chem forum where he kind of discusses his project in some detail. The main issue is, that he hasn't posted a damn thing since August of 2017 and he did mention wanting to mix "dangerous" chemicals ... explosive types ... so maybe he blew himself up or he ended up being a thwarted Unabomber and is in federal custody ... who knows ... but what that means for me is - no schematics and no Arduino code - he used an Arduino ESP-8266 to control his project - he wanted wifi for remote control ... obviously because of the nature of the stuff he wanted to stir.

So I've been assuming so far, that my specific issues in this project have been related to not being able to graduate the voltage through the coils. However, if that EE managed to get it working with a stepper driver, and ASSUMING that stepper driver pulses the coils without changing the voltage levels on them as it ramps up the speed, etc., then maybe my thought process on this has been wrong the whole time.

I'm going to look at the A4983 datasheet and see how it's different than the 8825 that I'm familiar with ... maybe something about it will jump out at me and explain why he chose that driver do do his project... which actually works.

Something else that I thought was interesting ... he said that he only set the driver for 200ma and his coils were a lot smaller than what I'm using. They also had fairly high gauge wire on them and he talked about how smaller stir bars were better suited to this project ... although he did mention wanting to use larger coils to get more adhesion between the stir bar and the coils ... but as I think about it ... sometimes going smaller means you can do stuff using less energy ... sometimes having a bigger gun isn't always the right option...

Mike

[eccerr0r]

Re: Magnetic Stirrer Project Reset

Yeah, that's how it should be done, digital on/off signals driving the coils, not with analog signals driving the coil and certainly not analog signals driving MOSFETs to drive analog to the coils...

[redwire]

Re: Magnetic Stirrer Project Reset

It depends on how much torque you want and how deep the flask is.

Three coils, L298 H-bridge, Arduino: http://blog.dzl.dk/2018/02/26/magnetic-stirrer/
https://www.edn.com/electronically-g...gnetic-fields/

Every commercial stirrer uses a motor...

[eccerr0r]

Re: Magnetic Stirrer Project Reset

Well it looks like you've run into the MOSFET problem that all H-tree designers run into at one point or another - high side control. Easiest solution is to use P-channel devices on the high side, but you STILL NEED NEGATIVE FEEDBACK if you want to drive them sinusoidally. The problem with MOSFETs nonlinear inout curve still exists whether you're running single ended or H-tree.

As for amplifier design, I've already told the solution - you need an audio amplifier. Using an existing audio amplifier or copying an audio amplifier design is sufficient.

[sam_sam_sam]

Re: Magnetic Stirrer Project Reset

would you be willing to educate me through the design process so that I can learn something?

I would be interested in this myself

I made this comment because I have no real use for this now but I have been keeping track of your progress with this project because I might have a need for something like this in the future so I try to keep an open mind for things like this and might learn something from the process you are going through to get where you need to be

[EasyGoing1]

Re: Magnetic Stirrer Project Reset

Any BJT or are you asking us to design a custom solution for you that's nonideal and uninteresting to us? Now we're talking about some design fees here, I could use some $ ...

I'm not opposed to this ... it would depend of course on the cost and whether or not your role would be an isolated one in that ... would you insist that you just design a complete working circuit for a fee or would you be willing to educate me through the design process so that I can learn something?

[EasyGoing1]

Re: Magnetic Stirrer Project Reset

I built two H-Bridges with this configuration:



And they work in that when I bring pins J1 and J4 high while J3 and J2 are low, current flows through the coils, then if I reverse the high and low state on the four outer pins, current flows in the opposite direction through the coil.

HOWEVER, I began to notice that when I had 12 volts on Vcc, the two top mosfets would get fairly hot while the two bottom ones did not. I also noticed that there didn't seem to be much voltage applied to the coils as their magnetic strength was pretty weak. That's when I decided to put an amp meter at the coil node C1(J5) to see how much current was actually flowing in the coil.

What I discovered was that it did not matter how much voltage was applied to Vcc, the current through the coil was always around 250ma. So when I applied 5V, then 12V then 19V at Vcc, the current through the coil was always the same. The voltage across the coils (that is the voltage from J5 to J6) is always 2 volts regardless of the voltage at Vcc.

What I am assuming at this point, is that the MOSFETs that are connected directly to Vcc are for some reason, "absorbing" the voltage applied at Vcc as it increases causing the rest of the circuit to remain constant and even though the current remains constant, the increased voltage at those MOSFETs increases the wattage they are dissipating which is causing them to get hot.

What I don't understand, is why are they doing this? Seemingly acting like resistors? I would expect this behavior from resistors, not MOSFETs. I also would have expected the lower MOSFETs to behave similarly to the top MOSFETs and what I mean by that is when I built this circuit, I was under the impression that since the current flowing through the circuit is the same, if the top MOSFETs were going to warm up, then the bottom ones would as well, but this is not the case and I'm stumped. It's as if the MOSFETs at the top change their properties because they're closer to Vcc but I know that's not possible.

Any ideas?

Thank you,

[eccerr0r]

Re: Magnetic Stirrer Project Reset

Oh... and also it's quite possible that your Onyko is powerful enough on its own - since most power amplifiers are bipolar (this won't work with the cheap capacitor coupled single ended amplifiers!), there's no need to bridge outputs - just use one amplifier with AC coupled input to the microcontroller output. No need for the 180 degree opposite phase, just the 90 degree phase output on the second channel, and use left on vertical coils and right on horizontal.

If anyone thinks I'm kidding about any of this, no I'm not. There's no reason this shouldn't work, just that it's inefficient.

[eccerr0r]

Re: Magnetic Stirrer Project Reset

Actually even easier, forget the 2n3904 inverter - use four different microcontroller outputs with the filtering you have.
Two outputs, 180 degrees out of phase of each other.

Other two outputs also 180 degrees out of phase with each other, 90 degrees out of phase with the previous two

Do all in software...

Alas the design aspect is still building the circuit without wasting your Marantz to power your stirrer.

[eccerr0r]

Re: Magnetic Stirrer Project Reset

BTW tongue in cheek ridiculous response:

Find two STEREO audio amplifiers that supports bridging, brand doesn't matter, Denon, Yamaha, NAD, Pioneer, Realistic, I don't care.
One stereo you bridge L + R together (i.e. connect to one pair of your up/down coils)
Use one channel directly, use a 2n3904 to invert the signal and feed opposite polarity to the other channel.

Other stereo do the same thing except hook to the left/right coils.

Feed your PWMed 90 degrees out of phase sine waves to the line input of two stereos (with appropriate inverters). You may need to AC couple them to the amplifiers.

Adjust the volume on both stereos until you get proper behavior...

Now the question left as an exercise to the reader is how to simplify this to not use full audio amplifiers. This is what the design cost goes into.

[eccerr0r]

Re: Magnetic Stirrer Project Reset

Any BJT or are you asking us to design a custom solution for you that's nonideal and uninteresting to us? Now we're talking about some design fees here, I could use some $ ...

[EasyGoing1]

Re: Magnetic Stirrer Project Reset

the positioning of your coils are probably not ideal for the pill. The magnetic fields visible to the pill are everywhere, unlike a real stepper where the fields are very compact and close to the magnet.

real stepper motors' steps are usually like 1/100th or 1/200 of one turn. Your step is 1/4 or 1/8 of one turn, this is huge difference that needs to be accounted for. Or you could make your stepper more like a real stepper - you can do this by putting in more coils.

Yeah, I basically confirmed this by scoping the output of a DRV8825 driver board and what I noticed right off the bat is that there is no variation in voltage applied to the stepper motor from the driver board ... it either sends 0 volts or it sends the full voltage from the Vsource pin... ALSO, it steps very harshly and even when I set the code so that I could manually step it, it just jerks around and my thoughts were that since a stepper motor uses magnets that are attached to a shaft and tightly coupled, then stepping in this way works just fine, whereas in my application, the magnetic pill is just free standing and will react harshly to that kind of stepping which brings me back to the need to graduate the magnetic field so that the transition from coil to coil is much smoother.

I thought about adding more coils, but doing that would not let me use a stepper driver at all ... and I would still be in the boat where I need to basically control the field with an analog "feel" to it ... so Im not sure that extra coils will help me at all unless they would give me the advantage of stepping coil to coil using digital on and off style switching because the travel distance between coils is shorter ...

When I looked at a photo of the inside of a commercial stirrer that uses electromagnetic energy to spin the pill, it only had 4 coils ... so it must be doable somehow.

I think my next step is to build an H-bridge but these MOSFETS have too much of a non-linear curve to them in terms of how they saturate from the gate voltage.

Do you know of any transistors that might have more of a linear curve to them? Or what do I need to look for in the datasheets to check for that as I do have a box of different transistors that I can look up their datasheets and see if any of them would work?

I need to see if my theory at least is on the right track where I graduate the field strength in an analog pulsating style ... if I can get the pill to spin in any way on a smooth path then I can look at other ways to accomplish it in the final circuit using some of the circuit designs you mentioned like the audio amplifier modules ... I've never worked with any of those but the logic seems right in what you're saying.