Megapoints power supply

[Izzy]

The Merg Servo4 boards that I built can also accept a range of input currents, which I presume are then regulated down to that needed. I guess the idea was that any suitable spare power supply could be used, an old controller transformer etc, but does then hit problems over the start-up current draw issues, which I don’t think were fully realised initially. 

 

This is probably one of the biggest problems to have surfaced with the attempts to use them for model railway control, the sheer numbers of servos sometimes involved in contrast to other r/c applications. Whereas with ‘ordinary’ solenoid/slow action point motors one decent couple of amp supply can generally cover as many as needed, this is not the case with servos.

 

Personally I don’t like the idea of multiple high power supplies just to serve this one aspect of a model railway, but maybe that is just me being over cautious, and especially as I prefer to keep things as simple as possible.

 

Now I just use ‘hacked’ servos, which can be run off just a single AA battery if required.

 

Izzy

[clecklewyke]

I’ve been having problems booting myMegapoints system and Dave Fenton suggested it might be a power supply problem. So I did a wee experiment, booting with just seven rather than eight servos connected to the servo controller board. And it worked!

 

I can confirm that if I reboot with all eight servos it normally fails but if I disconnect one servo, leaving just seven connected it always reboots successfully.  It does not matter which servo I disconnect so it is the number of servos that counts. It is not that one servo is faulty. I can live with the workaround of disconnecting one servo for the moment but not when I have the signals connected as well, which will need a further eight servos.

 

 So i think I’ll need to find a beefier power supply which is silent. I could not bear the fan whir of the supply Dave recommends.

 

The power supply I use is supposed to deliver 12v 6a regulated, so meets  Dave’s recommendation.of 2a per servo board, see https://www.amazon.co.uk/gp/product/B074HX4GJ3/ref=ppx_yo_dt_b_asin_image_o09_s00?ie=UTF8&psc=1

 

Is this unsuitable or does the recommendation of 2amps per servo board need reviewing?

 

Can anyone suggest a suitable  SILENT power supply?

 

Ian

[Crosland]

5 hours ago, clecklewyke said:

I’ve been having problems booting myMegapoints system and Dave Fenton suggested it might be a power supply problem. So I did a wee experiment, booting with just seven rather than eight servos connected to the servo controller board. And it worked!

 

I can confirm that if I reboot with all eight servos it normally fails but if I disconnect one servo, leaving just seven connected it always reboots successfully.  It does not matter which servo I disconnect so it is the number of servos that counts. It is not that one servo is faulty.

 

Do the servos move when booting?

 

If you reconnect the 8th servo after booting does everything work as expected?

 

If the servos are all moving or twitching simultaneously when booting then it's not just the maximum capacity of the supply that matters, but how it responds to the transient current draw at boot up.

 

 

[RobjUK]

These are fanless and quite cheap - 12V 10A for just under £15.

 

https://www.robotshop.com/uk/switching-power-supply-12v-10a.html

 

Or even a 15A which also looks to be fanless for a slightly lower price;

https://led.me.uk/12v-15a-constant-voltage-power-supply-for-led-strip

 

[clecklewyke]

1 hour ago, Crosland said:

 

Do the servos move when booting?

 

If you reconnect the 8th servo after booting does everything work as expected

 

 

 

In answer to your questions:

 

If I boot up with eight servos they briefly twtch then make a whistling noise like a modem or fax machine then go silent and don’t respond to switches

 

If I boot with seven the system powers up normally, responding

 

And, RobjUK, thank you for your suggestions. Have you(or anyone else) any experience of these units?

 

The first looks promising but I see that the Anotes on the second say, “Requires professional installation. For indoor use only not water resistant.

PLEASE NOTE: This unit may need to be mounted in an enclosure to meet Electrical Safety Standards.”

 

Ian

[RobjUK]

51 minutes ago, clecklewyke said:

Have you(or anyone else) any experience of these units?

Not those exact ones; other makes / types that are generally similar.

 

They are in the "open frame" class, which means they should be mounted inside some other enclosure of some sort, so people (or children) cannot get to the terminals.

The same applies to both, but one supplier does not bother with the disclaimer to cover themselves.

 

 

 

 

[AndyID]

6 hours ago, Crosland said:

If the servos are all moving or twitching simultaneously when booting then it's not just the maximum capacity of the supply that matters, but how it responds to the transient current draw at boot up.

 

Should it not be possible to avoid the problem entirely by having the firmware sequence the initiation of the control pulses fed to the servos, or am I missing something?

[Nigelcliffe]

12 hours ago, clecklewyke said:

I’ve been having problems booting myMegapoints system and Dave Fenton suggested it might be a power supply problem. So I did a wee experiment, booting with just seven rather than eight servos connected to the servo controller board. And it worked!

.......

 

 

I thought earlier you said there were two Meagapoint boards.  If that is the case, can you power up one, then power up the second with the existing power supply, without things dropping out ?   A simple on-off switch before the second board would be enough to try this. 

 

If that works, then more automated options to switch the boards could be proposed, such as a very simple time-delay circuit before the second board was given power.

 

 

- Nigel

 

 

[clecklewyke]

11 hours ago, Nigelcliffe said:

 

 

I thought earlier you said there were two Meagapoint boards.  If that is the case, can you power up one, then power up the second with the existing power supply, without things dropping out ?   A simple on-off switch before the second board would be enough to try this. 

 

If that works, then more automated options to switch the boards could be proposed, such as a very simple time-delay circuit before the second board was given power.

 

 

- Nigel

 

 

I will need two servo control boards when I instal my signals. For the moment I need one board to power eight point servos but  I will need to add another board to power eight signals. The power supply will boot seven servos but not eight.

 

Ian

[Crosland]

20 hours ago, clecklewyke said:

In answer to your questions:

 

 

You didn't answer "If you reconnect the 8th servo after booting [with 7] does everything work as expected?" If the answer is yes, then that certainly points to it being a problem with transiennt current at power up (caused by the servos twitching) upsetting the power supply.

[Crosland]

15 hours ago, AndyID said:

 

Should it not be possible to avoid the problem entirely by having the firmware sequence the initiation of the control pulses fed to the servos, or am I missing something?

 

That's certainly possible. The design should also avoid the servos twitching at start up.

Edited April 14, 2019 by Crosland

[njee20]

Yep, but it doesn’t, so you have to buy hardware to overcome a software issue!

[clecklewyke]

21 hours ago, Crosland said:

 

You didn't answer "If you reconnect the 8th servo after booting [with 7] does everything work as expected?" If the answer is yes, then that certainly points to it being a problem with transiennt current at power up (caused by the servos twitching) upsetting the power supply.

I intended to say, and indeed typed, that it boots with seven servos and continues to respond properly to the switche’s when the eighth is connected but it somehow got lost from my post - I find my new iPad does that, naughty tablet!

 

Sorry,

 

Ian

[Crosland]

2 hours ago, clecklewyke said:

I intended to say, and indeed typed, that it boots with seven servos and continues to respond properly to the switche’s when the eighth is connected but it somehow got lost from my post - I find my new iPad does that, naughty tablet!

 

Sorry,

 

Ian

So simply buying a higher capacity supply may not solve the problem, depending on how the supply behaves with the megapoints at power up. It might be a case of try it and see if you cannot find an alternative.

 

This is where old fashioned transformer based supplies can be an advantage, as they do not cutout quickly like a switch mode supply.

 

[clecklewyke]

I thought you should know that the system is now booting every time, with all eight servos. I’ve done nothing to it since my last report. The system just seems to have settled down and is working perfectly. Mind you I don’t have any spare servos so the problem might return when I connect another servo controller board and start fitting signals.

It’s a mystery!

 

Ian