Maker Pro
Maker Pro

What size cap do I need?

windinmysails

Mar 12, 2012
26
Joined
Mar 12, 2012
Messages
26
Hi all,

I'm not sure I've posted in quite the right part of the forum, so please feel free to move me if I'm wrong - I've had a look through and can't see anyone else asking this question, so here goes:

I'm running a inLog remote DAQ and control box at the end of quite a long piece of cable to turn on and off some 24V motors that take about 2A. The DAQ needs about 12V so I've popped in a voltage regulator and a small smoothing cap, but something strange happens which I can only attribute to voltage drop on my long cables. I've measured and without the motors running I'm down to about 20V at the entry to the voltage reg. When I fire up one of the motors (controllled by the DAQ) I'm guessing there's a surge that causes the voltage to drop so that the voltage regulator can't maintain its output. That causes the DAQ to reboot. If I isolate (i.e. give the DAQ it's own isolated supply) everything is fine. So I'm guessing I could use a decent sized capacitor across the output of my voltage regulator to keep up the voltage (the DAQ doesn't drain a lot of current) - but what size should I aim for?

Any suggestions very gratefully received!

Many thanks

Neil
 

gorgon

Jun 6, 2011
603
Joined
Jun 6, 2011
Messages
603
If you add a diode and a capacitor before the regulator, you'll 'isolate' the regulator from collapses on the voltage supply. Put the diode in the V+ lead and the capacitor after it. If you already have a capacitor before the regulator, just add the diode.

TOK ;)
 

windinmysails

Mar 12, 2012
26
Joined
Mar 12, 2012
Messages
26
Thanks guys,
I've actually already got a diode on the input to the voltage regulator (to protect the regulator from damage if it's connected backwards). In addition, I've got a 22uf 50V cap across the input (as described in the datasheet for the Traco Power TSR-1, 1A series device), but it seems this isn't enough to stop my problem. Do I just need a bigger cap?

Rleo, I have the device right next to the regulator (about 40mm away). I can't get any closer.

Thanks for your help people!

Neil
 

davenn

Moderator
Sep 5, 2009
14,039
Joined
Sep 5, 2009
Messages
14,039
Thanks guys,
I've actually already got a diode on the input to the voltage regulator (to protect the regulator from damage if it's connected backwards). In addition, I've got a 22uf 50V cap across the input (as described in the datasheet for the Traco Power TSR-1, 1A series device), but it seems this isn't enough to stop my problem. Do I just need a bigger cap?
Rleo, I have the device right next to the regulator (about 40mm away). I can't get any closer.
Thanks for your help people!
Neil

no almost definately not enough. The rule of thumb is 1000uF per Amp of current supplied but for general operations of 1 amp or less a 1000uF would do well. That's across the input of the regulator. You could stick that 22uF across the output of the reg.

Datasheet

interesting they are saying only 22uF on the input. try the higher value and see how it goes

Dave
 

gorgon

Jun 6, 2011
603
Joined
Jun 6, 2011
Messages
603
You could try to keep the 22uF and add another bigger capacitor in parallel. What's the current drawn here, from the DAQ?

Have you got anything over the motor, like a diode to short any reverse voltage spikes?

Could also be that the DAQ itself trigger due to a inducted noise from the motor.

TOK ;)
 
Last edited:

Harald Kapp

Moderator
Moderator
Nov 17, 2011
12,610
Joined
Nov 17, 2011
Messages
12,610
Decouple regulator and DAQ from the motor like this:
attachment.php


Thus the surge from the motor will be decoupled from the buffer capacitor at the regulator's input.

Harald
 

Attachments

  • Regulator.gif
    Regulator.gif
    3.2 KB · Views: 246

jackorocko

Apr 4, 2010
1,284
Joined
Apr 4, 2010
Messages
1,284
Thus the surge from the motor will be decoupled from the buffer capacitor at the regulator's input.

why wouldn't you want to put it here like this?
 

Attachments

  • Regulator.gif
    Regulator.gif
    3.7 KB · Views: 140

windinmysails

Mar 12, 2012
26
Joined
Mar 12, 2012
Messages
26
Thanks for all the replies.
So in answer to Dave's response: I'll have a go at that, I'll try a bigger cap there and see if things improve.
To Gorgon: No, there's no diode across the motor because it is sometimes run one way and sometimese the other. There is a small ceramic cap on there, but nothing much. Current drawn by DAQ is low - I've not got it here to measure but it's certainly less that 1A as that's all the voltage reg is capable of and I'm guessing it's a few hundred ma. I'll check when I'm with the box again tomorrow.
To Harald: That's basically what I was thinking of doing when I wrote this morning. If I put something across that would maintain the voltage for long enough that the inital voltage drop caused by the motor starting could be surpassed without pulling down the DAQ.
To Jackorocko: I'm not sure what you mean by putting it there, do you mean on the other side of the diode, across down to the ground? Wouldn't that mean that the cap was helping to support the motors as well? Or am I misunderstanding?

Thanks for all your help! I'll get on a try a few of these tomorrow and let you know how it goes!

Neil
 

Harald Kapp

Moderator
Moderator
Nov 17, 2011
12,610
Joined
Nov 17, 2011
Messages
12,610
@jackorocko:
The placement you suggest is possible, but any current surge from the motor will drain the capacitor. Thus the voltage across the capacitor decreases and the regulator possibly has not enough input-output voltage differential to generate a stable regulated output.
My placement has the advantage that should the surge from the motor current reduce the input voltage to the circuit below the voltage on the capacitor, the diode will block any reverse current. Thus the full charge of the capacitor is available for the regulator.

Harald
 

windinmysails

Mar 12, 2012
26
Joined
Mar 12, 2012
Messages
26
Hi all,
Finally I go around to trying the larger capacitors yesterday.
So, I first put a 470uf in where Harald suggested - this worked some of the time but other times it failed, so I changed it for a 1000uf and the situation seemed to improve - I now am able to turn on and off the motors without causing the DAQ to reboot.

One thing I was going to say was that I don't actually have the capacitor on the right hand side. The circuit diagram with the datasheet doesn't suggest that this is necessary, but is it here? If so, what size should I use please?

Another question I am pondering is how long the voltage drop could be maintained for and weather it would be better for me to create a seconardy circuit with a battery that was charged from the incoming supply. That way, either the motors drained too much for a little while, or indeed the main supply was running low (they are big batteries), this could be sensed and I could use the DAQ to shutdown the system cleanly. Does anyone have an idea for a simple circuit that would do this? I'm guessing a relay, a cap, a couple of diodes (to reduce the incoming voltage to the relay so that when it dropped below the activation point the relay would drop, causing an open or close to the DAQ for which I could control an action). Does this sound plausabile?

Thanks again for everyone's help!

Neil
 

Harald Kapp

Moderator
Moderator
Nov 17, 2011
12,610
Joined
Nov 17, 2011
Messages
12,610
Glad it works.

Generally a small capacitor (~100nF or so) is good for stabilizing the voltage regulator. If the data sheet doesn't mention it, it may not be necessary. It improves regulation but will have no measurable effect on your problem (motor current surge).

Harald
 

(*steve*)

¡sǝpodᴉʇuɐ ǝɥʇ ɹɐǝɥd
Moderator
Jan 21, 2010
25,505
Joined
Jan 21, 2010
Messages
25,505
If the problem is transient voltages, you may be better off with a number of capacitors, say 1000uF electrolytic, 10uF tantalum, and 0.1uF ceramic.

This is because the frequency response of these capacitors vary (due to ESR and internal inductance). Transients, by their nature, have a lot of energy at higher frequencies.

Having said that, I haven't followed the earlier conversation in detail.

edit: perhaps a small capacitor directly across the motor will help too. Say 0.1uF 50V. You may be getting inductive/capacitive pickup of these transients directly into your circuit. The capacitor will reduce the rate of change of voltage across these wires.
 
Last edited:

windinmysails

Mar 12, 2012
26
Joined
Mar 12, 2012
Messages
26
Thanks for all your help people.
I am now suffering from another, perhaps more peculiar problem!
The 1000uf cap that I've put in where Harald suggested seems to be keeping my DAQ alive and well. I've done a number of operations now to turn on and off the motor and all is good. The strange thing is that I've got another DAQ / control board which sits across the power cable nearer to the battery but which doesn't run from the power source - it's USB powered. This board uses a couple of relays to turn on and off a motor in the same way as the other, but these are 12V relays, so I'm using a 12V regulator to take the 24V feed down to 12V. The motor itself is 24V at about 3A, I can't see any cap across it so perhaps as Steve suggests I need to put something here. Basically, sometimes when I turn on one of the relays it causes the board to "reboot" but this is more of a USB disconnect / reconnect. I can't actually see why it would do this anyway as the board isn't powered from the same supply, but it's definitely related to me doing something on the power (turning something on or off). Any ideas people?

I'll try to get a decent drawing together to share with you all.

Many thanks,

Neil
 

Harald Kapp

Moderator
Moderator
Nov 17, 2011
12,610
Joined
Nov 17, 2011
Messages
12,610
This sounds like an EMC problem to me: When switching the load an electrical disturbance is created which can, among other effects, lead to a "reboot" of the circuit. It would help to have a schematic.

Does this really happen when you turn-on the relay, or at turn-off? It it happens at turn-off, a missing or defect freewheeling diode across the relay coil cut be the culprit.
If it happens at turn-on, the problem lies elsewhere. Maybe the inrush current to the motor induces a stray voltage via its electromagnetic field or ... There are alomost endless possibilities. You can try to put a ferrite (such as you can find e.g. on monitor cables) around the wires leading to the motor. And layout the wires such that they are as far away from the controller as possible, and by no means should they be parallel to the USB cable. Also using a higher quality USB cable might help.

Harald
 

Rleo6965

Jan 22, 2012
585
Joined
Jan 22, 2012
Messages
585
Can you use separate power supply for 24V motor? Is that AC or DC? IF AC motor, you placed a zero voltage crossing detector to trigger turning on of motor to avoid high current emf.
 

windinmysails

Mar 12, 2012
26
Joined
Mar 12, 2012
Messages
26
Hi Harald,

I'll have a go at making sure the USB cable is nicely away from the power side of things and see if it helps. I still haven't got around to doing a schematic. I have to say that quite a lot of the wires are rather close together as I've had to put everything into quite a small plastic box so it wouldn't surprise me if it were interference of some kind. Maybe I can put power and the USB stuff in different boxes.

Rleo,
I'm considering using a separate supply - actually, I was thinking of having a smallish sealed lead acid battery to run the motor which trickle charged from the 24V line thus mitigating the surge as the motor cut it. It's a DC motor. I would need to make a small circuit to charge the battery (probably 2 batteries actually to give me the 24V). What's the best small circuit to do this sort of charging? I'm guessing I'd have to make it disconnect the charging when the motor kicked in otherwise I risk blowing up my charging circuit!

Thanks again for your help,

Neil
 

windinmysails

Mar 12, 2012
26
Joined
Mar 12, 2012
Messages
26
Hi people - sorry for the delay in posting. I've been trying to work out how best to solve these problems I was having.
OK, so the good news is that it was an interference problem most of the time on the USB device. Actually, I think what was happening was that as the relays were turned off, the collapsing field was enough to cause some kind of shock in the USB board. Despite the face that the cables were well away from one another, the relays and the USB board were in the same small plastic enclosure. By putting the relays about 100mm away the problem has gone! The next time I do this for real I think I might put the control board in a metal (earthed) box to shield it a bit better.

I still get some issues with power drop off when the motors come in, but I'm not entirely surprised given the length of low quality cable I'm using. As part of what I'm doing anyway I will need to ensure that there is a power source much closer to the motors that is accessible should the main power be disconnected, so I'm going to have a battery which is charge from that 24V line, but the motors will run from the batteries. I think that'll take out some of the drop I see when they start. Does that sound like a sensible idea?

Thanks again for everyone's help.

Neil
 
Top