EEVblog Electronics Community Forum
Electronics => Projects, Designs, and Technical Stuff => Topic started by: apelly on February 13, 2014, 06:56:00 pm
-
I've had it in mind for some time now to build a stirring hotplate; they're useful but expensive. I'm after some ideas. Please note that I'm as much interested in learning some things as in the project itself, because I'm quite inexperienced. Also note that (hopefully) ignorance <> stupidity.
By itself a hotplate is pretty simple, so let's not worry about that.
It's the stirring that intrigues me:
There are a bunch of people on the web attaching magnets to computer fans, but that seems likely to end in tears around reflow temperatures. I thought about an aluminium "fan" with the motor a bit further away, which seems plausible, except that I'm imagining a rather sturdy piece of aluminium as a hotplate and I'm wondering how effective some magnets will actually be.
Something else I considered was an array of coils to drive the flea. This is a lot more interesting, but it raises a bunch of other issues though:
Can I get wire with insulation that will withstand the temperature (say 300 degC or so) NoCan I get some sort of potting compound with a similar heat capacity as my hotplate so the coils can be close enough to the surface Not required with plan B- What must the orientation and location look like? (As I write this I'm imagining a ring of holes containing magnets, each oriented perpendicular to the top of the hotplate. Thinking this through, the preceding question is probably not worth considering...)
- How well will a magnetic field propagate through aluminium?
- How many turns will they need (magnet theory?)
- How vigorously must I drive these coils?
- Can I tell if the flea is present, absent, following or lagging the coils?
- Do I care? (Probably: if the stirred medium is too viscous to stir at the set speed, it'd be better to stir as fast as possible, and accelerate. But would the flea accelerate over time to try and match the coils anyway?)
- How closely must they be balanced to keep the flea in the middle?
edit: Just more questions. Sigh.
edit again: Now the electronics is just a thought excercise. Sigh. Talked out of another pointless endeavour. I will simply use a couple of HDD magnets on a shaft.
-
I took apart a magnetic stirrer about 20 years ago. The stir bars are just normal magnets sealed in plastic with a sort of raised bearing area molded around the center. They are driven by electromagnet coils in side the base unit, but that is about all I remember.
-
Answering one of my own questions:
No. It is not practical to get insulated wire that will withstand those temperatures. Therefore It seems smarter to have insulation between the plate and the coils and introduce some light airflow.
is it practical to calculate, or even estimate, the thermal effect of having a big cavity in the middle of a hotplate?
-
I work in a research lab, and have had to repair a fair number of hotplate/stirrers for our chemistry dept. over the years. The most reliable ones just use a strong bar magnet mounted on the shaft of a variable speed motor, a half cm or so from the underside of the heating surface. The magnetic flux has no problem coupling through the aluminum or ceramic plates typically used. All that typically goes wrong with these designs is tight/dried up bearings on the motor, or the occasional dirty speed pot.
The newer ones use embedded electromagnet coils, and a sequential electronic switching arrangement to create a rotating magnetic field. Lots more to go wrong with these, and they just don't last as long (at least in our environment) as the simple mechanical ones. Have seen a few with burned up drive electronics, coils corroded by chemical spills that got inside, etc.
-
As I was posting this I see your reply N2IXK. I understand. Your comments, along with my magnet research over the last hour or so are quite discouraging. I will abandon the electromagnet idea. Any idea if the chemists notice any difference?
Just saw this thread http://www.sciencemadness.org/talk/viewthread.php?tid=23989 (http://www.sciencemadness.org/talk/viewthread.php?tid=23989)
It seems even quite a large slab of aluminium is no match for a hard drive magnet. In comparison an electro magnet will probably be quite wimpy.
(http://www.doranaerospace.com/Images/scimadness/hotplate/2013-06-10%2023.35.53.jpg)
-
The hotplate I use in my lab has magnets mounted on a platter which is connected to what looks like an a.c. fan motor.
(http://img.tapatalk.com/d/14/02/14/avymu6uz.jpg)
(http://img.tapatalk.com/d/14/02/14/esumeput.jpg)
The top is a porcelain ceramic type of material. Very hard, chemical resistant, low thermal mass. Basically they have heating wire directly in contact with the ceramic top. The low thermal mass is important to allow rapid switching of temps, at the expense of higher wattage needed for rapid heating. Similar to a soldering iron. If it's a bulky block, like that aluminum block, it takes forever to change temps. The thermal detection is minimal though I've had devices which had a connector for an external thermocouple, fir better temp control.
Sent from my EVO using Tapatalk
-
This one is pretty simple. Magnets on an electric motor, with a hall effect sensor for speed feedback. The white stuff on the bottom of the plate is insulation.
(http://i.imgur.com/EIFDav0.jpg)
(http://i.imgur.com/bdYAYJz.jpg)
I'm actually having some issues with this one after extended use - the display gets all garbled, but everything else still works fine. Goes back to normal after cooling off. There is a PIC on the main board and a 7 segment driver on the display board, isolated from each other by optocouplers. Reflowing every joint I can think of didn't help - my best guess is the optocouplers are being overheated and putting out thermal noise. Scilogex assures me it works fine and I'm imaging things. One of these days i'm going to try installing a fan to cool the PCB.
The design heatsinks 4 TO-220 ICs (I haven't checked part numbers) into the bottom plate, and between it and the hotplate itself the case forms an oven where the PCB is slowly baked. :o
(http://i.imgur.com/5LlQDpx.jpg)
-
Silicon Chip had a magnetic stirrer project one some time back, google for it.
-
I work in a research lab, and have had to repair a fair number of hotplate/stirrers for our chemistry dept. over the years. The most reliable ones just use a strong bar magnet mounted on the shaft of a variable speed motor, a half cm or so from the underside of the heating surface. The magnetic flux has no problem coupling through the aluminum or ceramic plates typically used. All that typically goes wrong with these designs is tight/dried up bearings on the motor, or the occasional dirty speed pot.
The newer ones use embedded electromagnet coils, and a sequential electronic switching arrangement to create a rotating magnetic field. Lots more to go wrong with these, and they just don't last as long (at least in our environment) as the simple mechanical ones. Have seen a few with burned up drive electronics, coils corroded by chemical spills that got inside, etc.
The "Steri-stirrers" that the UWA Chemistry Dept used had a bar magnet mounted on the rotating shaft of what looked a heck of a lot like an old record player drive motor.
About all that ever went wrong with them was students spilling various infernal concoctions into the guts of the thing! ;D
-
Silicon Chip had a magnetic stirrer project one some time back, google for it.
Thanks. I'll buy it and see what they have to say.
-
After reading this, the first thing that comes to mind is a F&P Smartdrive motor that I imagine is the principle you wish to replicate.
http://www.unicopter.com/1916.html (http://www.unicopter.com/1916.html)
Only the poles need to be facing inward?
-
After reading this, the first thing that comes to mind is a F&P Smartdrive motor that I imagine is the principle you wish to replicate.
Yes. Pretty much.
I had imagined poles facing up/down though. Now you bring this up, I don't know which would get better coupling.
-
Just as a note, its probably worth looking into accelerating the flea gradually, synchronous motors (that is essentially what this is) don't stall well so may not have the torque to accelerate it to the required speed.
-
I agree. There are also potential issues with liquid viscosity. I'm not sure how or if to address this yet.
In another thread someone suggested a stepper motor controller. I'm yet to follow up, but that looks like it makes sense, and may reduce component count too.