Servo Motors & Servo Controllers

[Siberian Snooper]

A further question, can several of these be linked to a power source other than batteries and if so what is recomended?

 

TIA

[50031Hood]

Hah! Well spotted. At the moment friction is all that is stopping the wire from dropping out but I was thinking of substituing the wire for a pin of the correct diameter. Alternatively I might shamelessly copy your heat shrink idea!

[Grovenor]

 

Alternatively I might shamelessly copy your heat shrink idea!

You're welcome, it was second or third hand when I got it!

Best

Keith

[50031Hood]

A further question, can several of these be linked to a power source other than batteries and if so what is recomended?

 

TIA

6V DC power supply; you might even get away with using a mobile phone charger. The current draw is pretty minimal compared to DC motors so you should be able link several of them. I've purchased a Y-lead that sends a single output to two servos at the same time for cross-overs which the PICAXE handles without complaint.

[Siberian Snooper]

Many thanks for the replies gents.

 

SS

[50031Hood]

6V DC power supply; you might even get away with using a mobile phone charger. The current draw is pretty minimal compared to DC motors so you should be able link several of them. I've purchased a Y-lead that sends a single output to two servos at the same time for cross-overs which the PICAXE handles without complaint.

Video for the above: http://youtu.be/mel9rxFgwpo

[ndk]

Just to show it can be done, I have assembled 2 signals from MSE components, for my 4mm narrow gauge layout, and have got as far as assembling them on a test stand . They are powered using the Peco smartswitch kit & as a servo beginner I have found this a good way in to this type of control. If I get really carried away I might even change to a GF controller but there is still some work to do yet - i.e. plant the signals on the layout!

 

http://www.youtube.com/watch?v=ApLz0MKmcTk

[Stephen Freeman]

Hi,

 

I thought this thread needed a bit of an update.

 

1) GF Controls - Dual Control Board - very easy to set up but you can't have the 2 outputs operating in opposite directions, i.e. the bounce is at one end or other for both of them. You still cannot alter the speed or the amount of throw.

2) Megapoints - good value for money and can be configured in a variety of ways (perhaps too many) and having to count flashes can be time consuming.

3) MERG Servo4 - have to be a MERG member, no bounce on standard firmware. SEMA4 firmware is available which gives a bounce facility. You need a Windows PC to set up the board and servos. The bounce is not terribly realistic in my opinion. On each bounce the arm moves to the end point and then back to the bounce point before settling on the endpoint after the third bounce. There is therefore no way (as far as I can see) to simulate the dwell of the other 2 boards when the signalman pauses to take breathe (?). However it is possible to alter the firmware to take account of this and I have put a revised version on the MERG Forum. Of course you need to be able to write the firmware to the PIC using a Pickit2 etc.

 

I've run out of time for now but will put a link on to a Youtube video I did a little while ago asap.

[Steve Hewitt]

Hi,

 

I thought this thread needed a bit of an update.

 

1) GF Controls - Dual Control Board - very easy to set up but you can't have the 2 outputs operating in opposite directions, i.e. the bounce is at one end or other for both of them. You still cannot alter the speed or the amount of throw.

 

Hi Stephen,

 

I'm not sure what you mean about operating in two directions?

The "bounce" is always at the "Danger" end of the movement, i,e, Operating switch closed on the earlier versions.

On the latest version "Danger" can be either "switch open" or "switch closed" by changing a simple dip-switch.

Which end of the servo movement this is at is determined by how you adjust the throw.

It can be at the Clockwise, or anti- Clockwise end by just adjusting the end stops.

You can set this differently for each of the two servos. 

Indeed, the Bounce function can be switched On or Off separately for each servo.

The speed of movement is not adjustable, the time for movement being constant. If the amount of movement is larger - longer throw, the speed is correspondingly faster.

I don't understand how you "cannot alter the amount of throw", as that is the sole purpose of the adjustment knob.

 

Steve.

[Stephen Freeman]

Hi,

 

I thought this thread needed a bit of an update.

 

1) GF Controls - Dual Control Board - very easy to set up but you can't have the 2 outputs operating in opposite directions, i.e. the bounce is at one end or other for both of them. You still cannot alter the speed or the amount of throw.

2) Megapoints - good value for money and can be configured in a variety of ways (perhaps too many) and having to count flashes can be time consuming.

3) MERG Servo4 - have to be a MERG member, no bounce on standard firmware. SEMA4 firmware is available which gives a bounce facility. You need a Windows PC to set up the board and servos. The bounce is not terribly realistic in my opinion. On each bounce the arm moves to the end point and then back to the bounce point before settling on the endpoint after the third bounce. There is therefore no way (as far as I can see) to simulate the dwell of the other 2 boards when the signalman pauses to take breathe (?). However it is possible to alter the firmware to take account of this and I have put a revised version on the MERG Forum. Of course you need to be able to write the firmware to the PIC using a Pickit2 etc.

 

I've run out of time for now but will put a link on to a Youtube video I did a little while ago asap.

Hi, having a few issues with Youtube at the moment- hopefully later today

Edited April 14, 2016 by Stephen Freeman

[Stephen Freeman]

Hi Stephen,

 

I'm not sure what you mean about operating in two directions?

The "bounce" is always at the "Danger" end of the movement, i,e, Operating switch closed on the earlier versions.

On the latest version "Danger" can be either "switch open" or "switch closed" by changing a simple dip-switch.

Which end of the servo movement this is at is determined by how you adjust the throw.

It can be at the Clockwise, or anti- Clockwise end by just adjusting the end stops.

You can set this differently for each of the two servos. 

Indeed, the Bounce function can be switched On or Off separately for each servo.

The speed of movement is not adjustable, the time for movement being constant. If the amount of movement is larger - longer throw, the speed is correspondingly faster.

I don't understand how you "cannot alter the amount of throw", as that is the sole purpose of the adjustment knob.

 

Steve.

Hi, on the last one I used it was not possible for the 2 outputs to have bounce in different directions and that had a dip switch. The earlier ones had bounce in one direction only for both outputs.

 

I should have said that the amount of throw is limited.

Edited April 14, 2016 by Stephen Freeman

[Stephen Freeman]

Hi, having a few issues with Youtube at the moment- success?!  http://youtu.be/-0gyuO1DSlU

 

 

 

 

Edited April 14, 2016 by Stephen Freeman

[dhjgreen]

Link does not work for me, sorry.

[Steve Hewitt]

If anyone wants to try out alternative Servo Controllers I'll have several on my Demo Stand at the Leigh 0 Gauge show on Saturday 23rd April.

 

I'll have the old Embedded Controls units, Servo4 from MERG, The ESU SwitchPilot and GF Controls Double and Eight-Way boards on my stand.

 

Do drop by for a chat and you can try them all out.

 

Steve.

[dhjgreen]

Link does not work for me, sorry.

Thanks for fixing it

[bigal10]

2 years on, and I've just been revisiting this thread, interested in the discussion about the PICAXE modules, mention in posts 73 + 74.

 

I see the kits are still available from Rapidonline.

 

Has anyone ever managed to make them work on DCC? - they would look like a cost effective unit at £6+ for 4 servo drivers.....

 

Regards,

 

Alan.

[Gordon H]

2 years on, and I've just been revisiting this thread, interested in the discussion about the PICAXE modules, mention in posts 73 + 74.

 

I see the kits are still available from Rapidonline.

 

Has anyone ever managed to make them work on DCC? - they would look like a cost effective unit at £6+ for 4 servo drivers.....

 

 

Given that PICAXEs usually run a form of BASIC interpreter to make programming easier, it seems unlikely that code execution would be fast enough to decode DCC packets. Of course, they can easily handle the servo side of things, as that is the kind of application they are intended for.

That said - there is a project here that claims to do DCC decoding using two PICAXES: http://www.picaxe.com/Project-Gallery/PICAXE-based-DCC-Accessory-Decoder/

However, note how high the clock rates need to be to do this task, and the fact that the job is divided between the two devices to make it work.

[Suzie]

...That said - there is a project here that claims to do DCC decoding using two PICAXES: http://www.picaxe.com/Project-Gallery/PICAXE-based-DCC-Accessory-Decoder/

However, note how high the clock rates need to be to do this task, and the fact that the job is divided between the two devices to make it work.

 

Decoding DCC packets and providing an accurate servo signal at the same time is a impossible to do completely accurately in software alone in a single processor. This means that to do a good job of being responsive to DCC packets as well as accurately positioning the servos you either need a hardware PWM driver to provide the servo pulses, or two processors, or compromise somewhere.

 

This approach using two devices is a very sound one for two reasons :-

 

1. the processors can be optimised to their respective tasks (decoding and positioning).

 

2. PICs are cheaper than dedicated hardware of sufficient resolution.

 

The processor speeds used in this are a bit excessive for decoding DCC, but not too excessive when accurately positioning servos. While the 14-bit position resolution for a servo that can be acheived at 64MHz might seem a bit excessive (it equates to about 1/20th of a degree), for slow movements it does eliminate visible (and audible) stepping on quality servos that is very pronounced when only 8-bit resolution is used (yes, half a degree steps are very noticeable).

 

Most single processor servo decoders will compromise somewhere, and running a fast clock will minimise delays in packet decoding while servos are being controlled acurately by interrupt routines. At one time PICaxe did offer some SX processor based solutions which might have been quite useful in this scenario if the processors could be run at their full 150MHz clock speed - but using them would be a bit keen.

[Gordon H]

Decoding DCC packets and providing an accurate servo signal at the same time is a impossible to do completely accurately in software alone in a single processor.

 

 

Doing the whole task In software alone would certainly be difficult. That is why microcontrollers are used for this kind of thing rather than microprocessors, as they have the necessary hardware peripherals built-in. It just needs a bit of careful design and thought as to how these peripherals might be used to best effect with concerns like interrupt priorities etc. However, I am not convinced the ready-made PICAXE environment allows for that degree or resolution of low level control, instead using raw processing speed as a substitute for it.

Edited April 17, 2016 by Gordon H

[Derekm]

Hi

 

im very interested in your article and would like to try my hand at modelling semaphore signals in n gauge but was wondering if you could point me in the right direction for components for the signals and perhaps any build tips you could give me.

[Donw]

Julian Coles  a member of MERG builds decoders with either solenoids or Servos these are very small and attached to the point motor so that there is minimal wiring the connections are made to the track right by the pointmotor. Every point motor has its own decoder. Generally just uses a PIC  coded in assembler. Designs and code are available to MERG members.

 

Don 

[Stephen Freeman]

18 hours ago, Derekm said:

Hi

 

im very interested in your article and would like to try my hand at modelling semaphore signals in n gauge but was wondering if you could point me in the right direction for components for the signals and perhaps any build tips you could give me.

In 2mm/n scales, Wizard have a limited range of parts, Modelu do some lamps and finials. Signal posts are available in most of styles including tapered square post ones .

[Steve Hewitt]

On 28/07/2021 at 17:00, Derekm said:

Hi

 

im very interested in your article and would like to try my hand at modelling semaphore signals in n gauge but was wondering if you could point me in the right direction for components for the signals and perhaps any build tips you could give me.

Hi,

 

Stehen Freeman has already answered your question about supply of components very well.

As to build tips, my generfal advice would be to use Brass and Nickel Silver for the parts you have to fabricate yourself, and solder wherever possible rather than glue.

Supplies of metal sections - strip, tube, angles etc. are available on-line from such as Hobby Holidays and Eilleens Emporium.

The best starting point for any model is a good photograph of the prototype.

Understand how the prototype was constructed and operates, and do the same in miniature where possible.

For whichever Railway or BR Region you are modelling, obtain the appropriate Signalling Reference books.

Excellent guides to model signal construction have been written by Derek Mundy, Mick Nicholson, Peter Squibb and they are well worth studying.

 

Good luck, and please let us know on RMweb how you get on.

 

Steve Hewitt.

[Theakerr]

I have built about 90% of my signals (00) using Ratio signal kits.  The 90% refers not to the number on my layout but the % of the kit used.  Specifically, all counterweights and L angle brackets have been replaced with metal ones from Wizzard models.   The plastic ones just did not stand up.  Because I do not use DCC I have used Megapoints servo boards.  Economical and easy to use.   I have recently used a new micro servo with a footprint about 1/2 the size of a RC90 (derekm see your thread) therefore much easier to install as a total unit (signal + servo) and for my signals they work more than adequately

Edited July 29, 2021 by Theakerr

[KingEdwardII]

For Servo controller board, if you're looking for DCC, I can recommend the Chesterfield Model Making & Minature Electronics ("CMME") Arduino-based DCC 16 Servo Decoder - drive 16 servos for around £20. Has things like signal bounce if your servos are driving semaphores.

 

It needs a little skill with programming, but there is plenty of information available and friendly help from the folk who make them.

 

https://chesterfield-models.co.uk/product/servodecoder/

 

No connection other than a satisfied customer.

 

Yours,  Mike.