How to wire a DIY heating mantle?

[Nicodem]

Building a magnetic stirrer – instructions for the layman (or laybee)

You can get an idea on how to build a magnetic stirrer by checking the drawing. Therefore I will rather concentrate the instructions on where to get the material and how to make the RPM regulator.

The magnet

Finding a strong enough magnet should not bee a problem. They are an essential part of many machines. Notably, the strongest magnets are found in loudspeakers and DC electromotors (in cars etc.). These can be broken in order to obtain a small ~2.5cm piece.
It is however a little more difficult to get a piece of more symmetrical dimensions appropriate for a magnetic stirrer. The magnet in the bicycle dynamo is round and its magnetic field expands in four directions. It could be used but it is quite heavy and you would need a stronger electromotor to drive it. I found the best magnet in an old analog electricity consumption counter. Some contain a quite strong and square magnet (like shown on the drawing). In the case you can’t get a magnet of appropriate form you can use a piece broken from a loudspeaker that is put in a plastic round stopper and fixed with wax. It has to bee fixed so that its magnetic field expands orthogonal to the axis of rotation. This field can bee felt and checked by using another small magnet to find the attraction and repulsion sites. Try your best to fix the magnet in the center otherwise the whole stirrer will vibrate extensively when rotating. In such misfortune use some wax to balance the eccentric weight.
Avoid weak magnets like the ones used for billboards – check if your magnet is capable of moving the stirring magnet at a distance of at least 7cm before deciding that it is strong enough.

The electromotor

Small electromotors are everywhere around you. Find an old cassette player, an old electromechanical toy, a batteries powered mini-fan or whatever else and take out the e-motor. Make sure it is not a brushless or step motor as these need different speed regulation (a useful motor will only have two connecting wires). Don’t take off its transmition wheel, as you will need its upper surface to glue the magnet on it. Check also what is its working voltage U_m (if nothing is written on it assume the voltage needed is the same as the one the whole machine/cassette player/toy worked on). It shouldn’t bee wider than ~3cm so it will consume less than ~0.3A of current (I_m) – this will bee enough and will make the RPM regulator easy to make.

The electricity source

Nowadays it is so easy to buy an adjustable 3 to 12V electricity adaptor. If not adjustable, chose one that will give at least the voltage of the electromotor but not more than ~2V more. Assuming you got yourself a 3Velectromotor from a toy, you can also use the adapter of your old mobile phone battery charger or some other 3 to 5V source, as the power needed is truly small (~1W). Avoid using higher voltage than the one used in calculating the RPM regulator parts or the transistor might overheat. You can use a connector like shown on the drawing or, if you will use the adaptor only for the stirrer, you can simply skip the connector and solder its wires directly on the RPM regulator (cut the wire and find the polarity). The stirrer will only work if the polarity (+ and - side) is correctly orientated so pay attention to this.

The RPM regulator

These e-motors can have their rotation speed regulated from min to max by a simple voltage regulation circuit. The circuit described here is the simplest possible of the useful types and it only requires a variable resistor (R1), a transistor (T), a capacitor (C) and a resistor (R2). The circuit shown on the drawing is built on an etched circuit board, but this is of course not worth the work for such a small thing. Better use a 2×3cm piece of an already etched and drilled universal board that can bee bought in any electronics shop. Actually you can even simply solder the few elements on the variable resistor if you don’t care for its “state of the art” look. The electronic parts can bee obtained from old cassette players, TV or other electronic accessory. Or you can simply buy them where you buy the board (it costs a few $ - the board and R₁ being the most expensive pieces). Choosing the types is wide open.
The transistor (T) used should support the electromotor current (~0.1 to 0.3A) and dissipate ~1W power – BC219 or similar should do (ask the vendor for analogue ones). You can either ask the vendor to mark you the position of collector, emitter and base of the three contact wires or find the specifications on the net using the type code in a search engine. Read about transistors on the net anyway in order to understand more. It is very important to connect it properly or it will not work.
The capacitor (C) should bee anything from 22microF to 220microF for 16V (or more, but they get bigger in dimensions). Mind the +/- signs on the chassis and orientate it so that the “–“ wire is connected on the minus lead on the board. Again, use the search engine if you know nothing about electrolytic capacitors.
The R₁ and R₂ properties are more specific.
Choose R₂ by assuming your e-motor works at U_m, need up to I_m and solve the equation: R₂=120*U/I_m. Lets say you have a 3V e-motor (U=3V) that need max 0.3A of current (I_m=0.3A) as these are the extremes for the most common small e-motors (like in toys etc.). This gives R₂=1200ohm (1/4W size will do).
Lets say we have a 4.5V adapter (U_a=4.5V; an old mobile phone charger for example). The voltage drop on R₁ will have to bee about U_m or just a little more. We can therefore calculate the needed resistance: R₁=R₂*U_m/(U_a-U_m). For our case this would bee 2400ohm but it should bee a little more in order to use R₁ also to switch of the stirrer at the min position. I think the nearest standard is 2700ohm, but in practice 3900, 4700 or 5000ohm would work as well (the only difference will bee the widening of the “switch off’ threshold). Now we have all the specifications needed.



That’s all

When you have all the material it will take you a few hours to solder the RPM regulator together and glue the parts inside a plastic box (depends on your skills and imagination). Enjoy.

Edit:
PS: I just realized that the BC219 transistor is quite old and might bee hard to come by. Therefore just ask the vendor to give you any switch or low frequency transistor that can handle 0.3A and can dissipate 1W of heat. I think something like BC140 or BC 141 are much more than enough (

http://noel.feld.cvut.cz/semi/philips/acrobat/7365.pdf

). There are hundreds of other types useful like 2N1613 and others. If using a really small e-motor that consume less than 0.12A (e-motors in walkmans) you can even use BC107, but keep in mind that you should use a reduction wheel with an elastic bend (also from a walkman) as these e-motors don’t have much torque.

[barkingburro]

how are you going to heat the mantle at this point?

[Organikum]

Stirrer:

The best source for strong magnets which are already perfectly shaped and come in identical couples of two are old harddisk-drives. Neodym. Best.

I personally wouldnt try to "fix" anything by using wax, this is absurd. Two component epoxy is the way to go.

RPM regulation by reducing the voltage on DC motors is plain bullshit as one looses all torque and the stirrer will not stirr anymore at lowered rpm´s. There are cheap rpm refulators available at Radio Shack (US) or Conrad (EU) which regulate rpm without this drawback and also hold the rpm "fixed". (they just interrupt the current periodically and have a feedback included). This is to be found in all drilling machines nowadays and available for AC and DC motors. Go better for this.

Anyways if DIY isnt yours dont fuss around but buy the device for this will be cheaper and better in the end.


Heating Mantle:

The advantage of a good electric heating mantle is that it has only a very low heat capazity by itself, say if you turn heat off there is not much restheat conserved in the mantle which heats on for some time. This is the main advantage over heating baths. Building a heating mantle which has a high heat conserving mass makes not much sense at all. If one does this, as less as any possible cement etc. has to be used. If the heating mantle is driven by DC less than about 50 V, it is easy to build a very light device which basically consists only of the heating wires and two layers fiberglass fabric. 12 V are recommended - transformators are readily available for use with halogen-lamps often with power output regulation included.

Not enough power?
Nonsense.
A 150W heating mantle beats every 800W hotplate - perhaps not in heating up time but everywhere else.


The "air-bath":

This is a nice way to achieve even very high temperatures. Basically the flask is contained in a (pyrex) glass or metal container and the air is heated. In old times this was done with open flame and a metal-net to protect the flask, but nowadays hot-air pistols with 300°C to 600°C are readily available for cheap. Such a hot-air bath is instantly improvised with some thick Al-foil. It is very useful for processes like the LOOMIS - benzychloride synthesis (hypochlorite + toluene) where the quick removing of the heat source is essential.
Dont direct the output of the gun directly against the flask please - the input should be tangentially. Also their has to be a little space left open at the neck of the flask where the air can escape - thats understood.

[Sredni_Vashtar]

I've passed the day looking into motor control. Exciting.

The idea is to design a cheap stirrer based on a 120 mm PC fan with the blades cut off and a pair of rare-earth magnets epoxied onto the hub. Experiments work so far, but can't control the speed very well. Want full torque across the whole speed range with good control. A 5W PC fan is powerful, quiet and reliable for long refluxes, and is low profile.

Basically, we are looking at 2 types of motors in this game: brushed and brushless. PC fans are further complicated because in addition to being brushless, they have some active circuitry to control speed.

With a standard brushed motor, when you apply load, it slows down. As it slows, it's resistance drops. If you use a  series resistance to control the motor, the drop in motor resistance causes a fall in voltage across the motor. This reduction in power reduces available torque.

The way to avoid this effect is to use a Pulse-Width-Modulated power supply for the motor: it always gets full voltage, but the voltage is applied in varying width pulses. This way, you get maximum torque at all motor speeds. Typical PWM controllers run at 5-20 KHz pulse frequency.

Another problem is that common brushed motors operate at too high a speed for a magnetic stirrer. Typical low voltage DC motors are 8000 rpm or more. You want say 150 - 1500 rpm for your stirrer with a typical 400 rpm speed. Unless you want to operate your motor at 10% capacity all the time, you need a reduction gear or pulley.

Brushless motors (including most PC fans) can be controlled by a PWM controller. PC fans tend to operate at a lower speed to start with (2500 rpm) so you can get away without the gearing. However, PC fans require a low frequency PWM to allow the internal electronics to function properly, say 20 - 160 Hz. I haven't found a commercial controller that runs at this speed. I've ordered some parts to make one and will post results. Ordinary PWM controllers probably work on PC fans but not efficiently - has anyone tried?

PWM   circuit

(http://casemods.pointofnoreturn.org/pwm/pwmtheory.html)

Finally magnets. The most powerful available are Neodymium-Iron-Boron (NdFeB). These are fairly cheap but have a low operating temperature (some well below 100 C). Get above this and the magnet dies. Need to be careful with the magnets close to the hotplate. Best for hot areas are SmCo (Samarium Cobalt), but these are not as powerful, uncommon and expensive. I'll go for the NdFeB and see how it goes.

Wonder Magnets - types of magnets and uses

(http://www.wondermagnets.com/cgi-bin/edatcat/WMSstore.pl?user_action=category&category=Magnets_and_Magnetism)

Organikum: hard-drive magnets sound a good source for this.

[Nicodem]

The best source for strong magnets which are already perfectly shaped and come in identical couples of two are old harddisk-drives. Neodym. Best.

I agree with Organikum. The NdFeB magnets from HD-drives are the best. I used them in one of the three stirrers I built, but in my experience it is very difficult to get them off the metal jacket where they are fixed on. I broke all of them trying to get them off. So I considered them as the “non-symmetric” ones. Though I only tried with three old HD-drives and they all had magnets of slightly different shapes and fixed in different metal jackets so you might bee lucky and obtain a good one if you smash enough drives. Maybe there some fixed only with a screw? Maybe the best thing is to simply bend the metal jacket in order to expose the two magnets upward. At least this is how I see Organikum’s idea, if understood correctly?

I personally wouldnt try to "fix" anything by using wax, this is absurd. Two component epoxy is the way to go.

Maybe I did not express my self clearly enough. I said the magnet (if unsymmetrical or impossible to glue on the e-motor) should bee fixed in a round plastic container with wax (or some other flexible resin). Then, the plastic container should bee glued to the e-motor with very strong glue.
You can’t use epoxy resins to fix the magnet in the container. How are you going to re-fix the magnet in order to remove the vibrations caused by mass unbalance when it is fixed with hardened glue?
I used this trick with one of my mag. stirrers and it works perfectly with the broken magnet from a HD. It was also the easier one to fix the vibration problem which is the main trouble when you don’t find a magnet of symmetrical shape.

RPM regulation by reducing the voltage on DC motors is plain bullshit as one looses all torque and the stirrer will not stirr anymore at lowered rpm´s.

Organikum, you are exaggerating and complicating way too much. In all my three stirrers I used this basic voltage regulator and it works like a charm even with the minimum possible speed. The one described can nicely rotate a 3cm stirrer at ~2 rotations per second and goes to a max where you can’t see the stirrer anymore. It can make rotate 1/2L of liquid in a really strong vortex pulling air at the center. I can’t say if the other two regulators would also work as slowly since I use them for the other two stirrers which are of different design using reduction wheels connected by an elastic band. I can’t find anything faulty in these home made stirrers that would make the commercial ones superior. The only thing they truly lack is a hot plate.

DC electromotor properties:
If a DC motor does not have much internal ohm resistance it will keep enough torque even if the voltage lowers and consequently the RPM’s drop. At lower RPM the counter-potential drops as well and consequently the motor can draw more current which give a stronger mag. field resulting in more torque. This is exactly why electric vehicles usually don’t require a transmission gear if such motors are used. It is a self equilibrating process. I know this from theory as well as from experience. But all this apply only to the normal DC e-motors. Brushless or step-motors require a completely different and more complex circuit for regulation.
Those interested should check the link Sredni found where the pro/contra of varius RPM regulations are described:

http://casemods.pointofnoreturn.org/pwm/index.html

Anyway, you are right. Most bees here know shit about basic electronics so it might bee more practical to simply buy a regulator, but make sure it applies to your e-motor as there are many different types. The regulator in the AC drilling machines, for example, is only a simple rheostat and will work only with an AC brushed motor. Many precise RPM regulators that can bee bought only work for step-motors (like those found in 3.5 or old 5-inch drives). So make sure it will work before you waste your money. Or simply buy the motor and the regulator together.
I would not humiliate myself by doing so, though. There is already so much garbage material around that can bee recycled into useful equipment. Besides electronics was my hobby when I was a kid and I like doing these things from scratch.

Sredni:
If you “use a series resistance to control the motor” you will never attain any useful RPM control. This is nearly not the same as controlling voltage which allows the motor to draw as much current as it needs. A resistor limits the available current causing the motor to fault. It’s no wander why you loose torque. Try using a transistor and you will see how huge the difference is.
I’m afraid that anything above 10Hz will negatively interfere with the PC fan internal circuit. It would probably still work at high frequency PWM regulation but in the same fashion as if you would just limit its current (loosing torque). The frequency should bee at least three times lower than the frequency of the steps in the rotating mag. field. This is of course impossible when you would want for example only 5 rotations per second. Assuming there are six coils inside the motor the frequency of the PWM should bee 5×6/3=10Hz. However, try and report back (but use a lower power transistor than the one described in your link – a PC fan only consumes less than 200mA!).
The heating caused by a hot plate or oil bath is not such a problem. I attached two aluminum disks (from old HD-drives) separated by ~2mm in on top of the stirrer. This insulates and dissipates excess heat from the oil bath perfectly. Hot air goes up so you don’t have to worry for the stirrer to heat up (most certainly not up to 100°C).

[Organikum]

but in my experience it is very difficult to get them off the metal jacket where they are fixed on. I broke all of them trying to get them off.

You dont need to get them off, they are usually placed on two identical metalpieces and may be used together with these. The metalpieces also close the magnetic field and the magnet is much more powerful with than without.
If you really desire to get them off you may bend the metal where they are fixed on - the metal is soft and easy to bend - and they will go off with ease. You cannot pull them off anyways, you have to shove them.
Talking of magnets from 3,5" drives and 2,5" drives. Magnets from very old 5 1/4" are another story though  

The question of torque just depends on WHAT you stirr. It seems that I stirr mostly stuff which thickens or is heterogenous - probably thats my reason for preferring strong stirrers (in the end nothing beats an overhead stirrer/vibrating stirrer anyways).

And hey Nicodem, you wrote WAX.
And I dont think thats a reasonable way to fix ANYTHING.
Maybe I am wrong.
Dont think though....


ORG

[Sredni_Vashtar]

I knocked up a PWM controller as per the website design. I tried 3 PC fans and a 2.5V brushed motor from an electric toothbrush.

With all motors, high driving frequencies were much better than low. Up to several hundred Hz, the motors made funny noises and stalled easily when driven slowly; torque was also low.

40 KHz was was inaudible and gave good results. A 120 mm PC fan can be smoothly controlled from around 250 rpm up to 2500 rpm. Decent torque from about 400 rpm upwards.

You don't need to build a circuit. Buy a pulse-width-modulated low voltage motor controller - can be had from various hobby and electronics suppliers. Ready made controllers operate at high frequencies and should work find with PC fans.

[PsuedoPsycho]

Nicodem I dont know what country you are from, but your
 knowledge and vocabulary is excellent! You would get my vote for Noble cladestine chemistry pillar of the community
award! Keep up the good work!
AFOAF SWIM met recently built a contraption from magnets
ghettoed out of a pc hard drive and a Apple floppy disc
drive from 1984 (shame)It would have sold good at the ever so popular auction site in antique pc hardware!?
Cheer's and keep up the good work!!