Signal voltage problem

Thread Starter

servo

Joined Aug 11, 2011
31
Hi, everyone. I have a simple, but thus far very frustrating problem with a signal control circuit I've designed.

The circuit sims fine in Multisim and works fine in one direction on the breadboard, but when I flip the switch, it stops working like I flipped the off switch. The signal needs to be ±10VDC and the amperage is negligible. I don't understand what's causing it. It works fine when I hook it up directly to a variable power supply, but when run through the regulator and pot it doesn't work right. Please help.
 

Thread Starter

servo

Joined Aug 11, 2011
31
Yes. That is the case with this setup. I've been running this valve with multiple power supplies and there have been no issues, but I'm trying to build something that will allow me to run it off of just one supply. The design sims fine as all I'm doing is reversing the voltage through the load. I've rewired it and I get ±10V running it through a multimeter, but when attached to the valve it only works the one way. The other way only causes a reaction when the pot is fully open and then the current draw jumps up to 500mA and the servo starts vibrating. Also, when the pot is not fully open, I'm reading 20V+ off of the wiper. The regulator is still putting out 12V so where is the extra voltage coming from?
 
Last edited:

Thread Starter

servo

Joined Aug 11, 2011
31
I can only imagine that I've got voltage coming into the ground terminal of the pot. There's no other reason to have that much voltage on the wiper. So why am drawing voltage off the ground? Should I have my grounds separated more?
 

Thread Starter

servo

Joined Aug 11, 2011
31
All my grounds are tied together, as they are connected to the negative bus on the breadboard which connects to the negative lead on the power supply. What I meant is should I have them farther apart on the bus? Would that make a difference?
 

SgtWookie

Joined Jul 17, 2007
22,230
It could be that the grounds ARE tied together, and one side of the valve servo is grounded.

[eta] whoops, cross-posted.

At this point, I think we really need to see a schematic of the side you're trying to control. If either input connection to the servo is grounded (it's not floating) then you're going to connect 10v to ground when you flip the switch, which would explain high current.

You might have to use optocouplers/optoisolators to completely isolate your controller board from the valve servo.
 

Thread Starter

servo

Joined Aug 11, 2011
31
I'm not familiar with optoisolators/optocouplers. My power is drawn from the same 24V supply as the valve, and is brought down to 12V using the regulator. At the end, the board is connected directly to the negative lead on the power supply. Because it can be controlled directly by the power supply, I know that it does not have to be a dual supply. One input to the servo is grounded and the other is connected to the pot. Both the power and the ground go through the switch which is between the board and the valve. I'll attach a pic of my setup.
 

Attachments

strantor

Joined Oct 3, 2010
6,782
I'm not familiar with optoisolators/optocouplers. My power is drawn from the same 24V supply as the valve, and is brought down to 12V using the regulator. At the end, the board is connected directly to the negative lead on the power supply. Because it can be controlled directly by the power supply, I know that it does not have to be a dual supply. One input to the servo is grounded and the other is connected to the pot. Both the power and the ground go through the switch which is between the board and the valve. I'll attach a pic of my setup.
better to post your images in PNG format. your BMP is so tiny, we can't make anything out. this is what it looks like to us:

What terminals are on your valve?
is it +24V, 0V, sig -, sig+, - or - is it +24V, +sig, 0v?
 

Thread Starter

servo

Joined Aug 11, 2011
31
It's an 11+PE connector. The relevant pins are +24V supply, +24V enable, 0V ground, "command +" and "command -" which are ±10V. The valve schematics are proprietary, so I don't have them. Rest assured, as long as the valve is powered and enabled, only 10V and practically no current needs to go through the command pins to move the spool. I've been doing this for months with a separate power supply so I know it works.
 

strantor

Joined Oct 3, 2010
6,782
It's an 11+PE connector. The relevant pins are +24V supply, +24V enable, 0V ground, "command +" and "command -" which are ±10V. The valve schematics are proprietary, so I don't have them. Rest assured, as long as the valve is powered and enabled, only 10V and practically no current needs to go through the command pins to move the spool. I've been doing this for months with a separate power supply so I know it works.
did you test to see if either of the signal pins are connected internally to the 0V pin?
 

Thread Starter

servo

Joined Aug 11, 2011
31
Yes, the 24V goes into the power and enable pins, as well as the branching off to the board. The black is the ground where the 0V pin, the board, and the command ground are connected.
 

strantor

Joined Oct 3, 2010
6,782
Yes, the 24V goes into the power and enable pins, as well as the branching off to the board. The black is the ground where the 0V pin, the board, and the command ground are connected.
ok, disconnect the command ground from there. leave it floating. Also, if that red banana jack is +24V then your breadboard is not set up like your drawing. You have the wiper (orange wire, center of the pot) connected to the red banana jack (or at least thats the way it looks).
 

Thread Starter

servo

Joined Aug 11, 2011
31
Ohhh, I see. No, the banana jacks on the board are going out to the command pins and the switch. The power and ground are the red and black alligator clips. The orange lead coming in from the left is the power and the yellow going up top is the ground. The white wire coming off the black banana jack is the command ground.
 

strantor

Joined Oct 3, 2010
6,782
Yes, the 24V goes into the power and enable pins, as well as the branching off to the board. The black is the ground where the 0V pin, the board, and the command ground are connected.
ok, disconnect the command ground from there. leave it floating. Also, if that red banana jack is +24V then your breadboard is not set up like your drawing. You have the wiper (orange wire, center of the pot) connected to the red banana jack (or at least thats the way it looks).
just to clarify , when you say command ground I assumed you were talking about position "B". is this correct? if not, disregard what I said about disconnecting it.
 

Attachments

SgtWookie

Joined Jul 17, 2007
22,230
I've been running this valve with multiple power supplies and there have been no issues, but I'm trying to build something that will allow me to run it off of just one supply.
So, you are running both the servo and this 12v regulator from the same supply, right?

Then they have a common ground, via the single power supply. Just because you stuck a linear regulator between your single power supply and the servo does not mean that it is isolated. Now, if you had used a 1:1 transformer from the AC to isolate the power, you would have a different situation, and it would be working.
 
Top