looking for a trigger shaped potentiometer

[Simon]

Hi guys, i want to build a PWM speed controller for a micro scalextric. the standard controllers seem to have a 60 ohm reostat so I've not got much chance on reusing that and they don't look very well made anyway.

Where can i buy without breaking the bank a trigger shaped potentiometer ?

[bilko]

What about an old battery drill ?

[insurgent]

I guess it depends on what your bank's breaking point is
As yachtronics mentioned, one from a drill would be an option: VSR drill triggers range from expensive to *very* expensive. The Dewalt VSR trigger for my drill set me back $40 and I've seen them as high as $100 (Bosch). Finding an inop junker at a swap meet for $5 might be a better option if you don't mind used.

[ivan747]

What about an old RC controller? I have one and it has 4 pots and some good circuitry. You can make a board of the exact same dimensions and put whatever you want in it.

[Simon]

What about an old RC controller? I have one and it has 4 pots and some good circuitry. You can make a board of the exact same dimensions and put whatever you want in it.

excellent, yes used is fine, my other idea is a slide pot coupled to a large lever like a speed boat control

[Zero999]

Hi guys, i want to build a PWM speed controller for a micro scalextric. the standard controllers seem to have a 60 ohm reostat so I've not got much chance on reusing that and they don't look very well made anyway.

You could Use one in a potential divider with an op-amp to give the required voltage. It should last much longer if run at much less than its rated current.

How are you planning to do the PWM?

What supply voltage are you using?

I have a few circuit ideas I could post, if you can provide more information.

[Simon]

well I'll either use a 555 IC or most likely one of the many PIC12F615's i have "in stock" , use the trigger to give the control signal to an ADC inputand output a corresponding PWM drive to a power N channel mosfet. The voltage is err, not sure, everything is rated at 12V but the input transformer is 16V AC with rectifier etc in the box on the side of the track so really it is 14.5V minimum.

PWM will give better control and probably provide more torque at low speeds than the reostat (potential divider) method and so add to that accurate speed control that at the moment is just not there.

[Zero999]

I have an idea using a dual single supply op-amp (e.g. the LM358) and a quad comparator IC (e.g. the LM339). The comparator can be used to do PWM (much more cheaply and effectively than the 555) and an op-amp is required to amplify the tiny voltage created when the rheostat is used as part of a low current potential divider (this is required whether you use an MCU or not). I'll post a schematic tomorrow.

[Simon]

Yes i had thought of using the original reostat on a low voltage or via a voltage divider and using a low voltage reference for the MCU's ADC, but I don't like the cheap reostat anyway

[sub]

Sparkfun sell 10k trigger pots for US$2.75: http://www.sparkfun.com/products/10314.

[Zero999]

Here's the simplest solution I could think of. For one channel, the  LM393 could be used. The LM339 could give you three channels.

I got the PWM circuit from the link below and modified it so it will work with a 60 Ohm pot.
http://www.4qdtec.com/pwmmod.html

The downside is, it's not amplified so uses a lot of power, at 12V 36mA per channel, plus the current required by the comparator. It's possible to reduce the current by increasing R5 and using an op-amp to amplify the voltage.

[Simon]

Here's the simplest solution I could think of. For one channel, the  LM393 could be used. The LM339 could give you three channels.

I got the PWM circuit from the link below and modified it so it will work with a 60 Ohm pot.
http://www.4qdtec.com/pwmmod.html

The downside is, it's not amplified so uses a lot of power, at 12V 36mA per channel, plus the current required by the comparator. It's possible to reduce the current by increasing R5 and using an op-amp to amplify the voltage.

err, is it me being thick or is U1A not actually doing much for the output ?

[Simon]

Sparkfun sell 10k trigger pots for US$2.75: http://www.sparkfun.com/products/10314.

looks good, thank you. I'll just have to work out how to fashion the control body

[Zero999]

Here's the simplest solution I could think of. For one channel, the  LM393 could be used. The LM339 could give you three channels.

I got the PWM circuit from the link below and modified it so it will work with a 60 Ohm pot.
http://www.4qdtec.com/pwmmod.html

The downside is, it's not amplified so uses a lot of power, at 12V 36mA per channel, plus the current required by the comparator. It's possible to reduce the current by increasing R5 and using an op-amp to amplify the voltage.

err, is it me being thick or is U1A not actually doing much for the output ?

Yes, it's you being thick.

The circuit won't function without U1A.

There's a full description of the circuit on the page, I linked to in my previous post.

[Simon]

right I'm thinning out a bit now  yes i see the link

[Simon]

I think I'm going to end up with a micro-controller design as I'm getting ambitious.

I want to have two channels so that the cars can share the same power rails. I can do this easily by using 24V AC power drive and putting diodes in the cars so that they work on one polarity only. each controller can control the duty cycle of a different polarity.

My only worry now is that I want to have water very close to the track (the wheels should go through it) so I need to factor risks of the water and how much I am willing to accept. I don't know how conductive water is to 24V AC

[Zero999]

Water is a bad idea because it will cause corrosion. The conductivity of the water depends on the ionic content.

The simplest and most effective way to perform speed control on AC is to use phase control. Normally a TRIAC is used but you'll need to use an SCR as you want half wave rectification. You can then connect the speed controllers in reverse parallel.

I assume you're going to be power this off a transformer? Be ware, using half wave rectification for large loads causes DC flux in the core which can cause it to saturate, excessive current flow and overheating. You could get round this by overrating the transformer and not driving only one car on the track for long periods.

Another solution is to use 24VDC and an h-bridge to control the duty cycle.

There are also more exotic solutions to this problem which involve keeping the track powered continuously and superimposing a digital signal on it which is decoded inside the cars which contain a built-in PWM controller. The advantage of this is more than two cars can be run on the same power rails.

[Simon]

yes I'd love to do full blown digital multi channel control but I'm not sure where to begin there, my plan for the moment is to have a +/- 24V DC supply and use a H bridge to alternately send power in the right polarity. So at 50% 24V is equivalent to full 12V (100% duty if it were 12 V control), that way I'd just put a diode in each car and channel selection is acheived.

How would I super impose a "signal" on a power rail ? I might just have room for some electronics in the cars (1:64 scale) A mini is 35 mm wide over the wing mirrors

[Zero999]

There are two ways I know of.

One is to superimpose an AC digital signal on the DC supply. The signal is just a squarewave which is AC coupled on to the power supply rails. Manchester coding is used to ensure there's no DC content so it can pass through AC coupling capacitors. The disadvantage is the supply's impedance would need to be high at the AC signal's frequency which means putting chokes on the power supply rails and the devices running off it.

Another is to use a squarewave AC signal for the whole power rail and rectify it before any of the devices. If the duty cycle is kept near 100% (don't forget a +/-12V square wave has the same RMS voltage as 12VDC),   Again, Manchester coding could be used to make sure there's no DC content in the signal. The disadvantage is power will be lost in the diode drops at the rectifiers.

What ever distribution system is used, each car would have to be given a binary code so it knows you're talking to it an the controllers would have to be polled sequentially as it's not possible to transmit at the same time.

If you're only using two cars, why torture yourself with microcontrollers? Phase contol has got to be the easiest way to do it.

How much power so the cars use?

Just make sure the transformer is much larger than you need, double the power rating of both of the cars should do.

Attached is an example which might need a bit of tweaking but should work.

[Simon]

hm yes on two cars i suppose an AC drive will be the easiest. I doubt the motors take more than a few hundred milliamps so no problems on the transformer. As for the diode drop I can just increase the supply voltage

in the case of a superimposed control signal how would it be injected onto the 12V rail ? just put it in parallel via a coupling capacitor ?

[Zero999]

The AC signal can be injected on to the 12V rail via a capacitor. You just need to make sure there aren't any AC coupling capacitors on the 12V rail, otherwise it will be shorted to 0V. Inductors can be used to ensure the 12V rail has a high impedance to AC.

The diode drops are less of an issue for the 24V phase control method than running from 12VDC and using an h-bridge and full wave rectification in each car.

[Simon]

well I assume that a 100nF cap on the motors will not be an issue ? I could put an inductor in series with the motor.

For the moment I think I'll just stick to 24V AC control and worry about sofistications later

[Zero999]

The 100nF capacitor is there just for EMC. Adding an inductor would make it high impedance to the AC signal and improve EMC further.

Yes, I think 24V phase control is the besty solution for now. This method has been used to control the speed of universal motors and the brightness of lamps for years and has been proven to work well.