voltage "clamp" when input analog signal removed

[skippyV]

Hello,

We have a box that has its output controlled via an analog input.
The box spins a shaft dependent upon the analog input.
If the analog input is 5 volts - the shaft does not spin.
If the analog input is 0 volts - the shaft spins at top speed.
And of course the speed varies between those end-points as you would expect.

Now the problem arises where power may be lost to the analog input signal - which would drop it's output to zero. And subsequently rotate the shaft at full speed.

So my question is what kind of circuit should we use to ensure that if the analog input were to drop out of the picture all of a sudden (due to unexpected power/communication lost to that device) - how could we make the input of the box "see" 5 volts and stop the shaft?

Thanks,

Skippy

 

[Harald Kapp]

Hello Skippy.

The solution depends on the input resistance of your box and of the way the control signal (0V...5V) gets lost.

If the input signal goes to 0V with a low resistance (depending on the output circuit of the source for the control signal) you will need a circuit that detects 0V and switches over to 5V. It can be done.
If the input signal goes to 0V by a haigh resistance (broken wire, unplugged connector etc.) and the input resistance of your box is reasonably high, you can simply add a resistor from the input to 5V. As long as the source drives a control signal between 0V...5V, the input will see this signal and your box will act accordingly. If the input signal is lost, the resistor will force 5V to the input and the shaft will stop, This so called pull-up resistor should be at least 10 times smaller than the input resistance of your box.

I think the design of the box is unfortunate. If you can, have it redesigned such that 0V=stop and 5V=full speed. that would be safer in my opinion.

 

[skippyV]

This box design is a$$ backwards but for now we're stuck with it.
I think I see your idea. You mean to change the range I think with the additional resistors.

Current model:
analog out --> shaft speed
5v -> stop
4v -> slow
3v -> medium
2v -> fast
1v -> faster
0v -> fastest

New model:
5v -> stop
4v -> slow
...
1v -> faster
0.1 v -> new fastest
0 v -> stop

That way if the analog output circuitry lost power - it'd shut off.
Is that what you mean? This might work.
Thanks for the advice!

 

[skippyV]

After pondering this some more, I'm unsure of how to do it.
I understand the idea of a pull up resistor to put 5 volts across the box input - when there is no signal. But I don't know how I'd do that at the same time I'd drive the analog signal on the input. It couldn't go above 5 volts.
I could use the same source for the pull up resistor that we're using for the box itself. No power to the box - shaft won't turn. Power to the box and pull up resistor - 5 volts at input and shaft won't turn. That makes sense.

But how would you do that in conjunction with a analog signal at the box input?

My electronics knowledge is too far gone in memory. I need some "spoon feeding" if anyone is up to the task.

Thanks

 

[KrisBlueNZ]

Hi Skippy. As you say, a pullup resistor would work if the connection is broken, but it may not work if the voltage source loses power. It depends on the characteristics of the voltage source. One way to find out would be to test it. Connect a pullup resistor, make sure it doesn't (significantly) affect the control voltage, then kill the power to the voltage source and see what happens.

If that's not going to work, we can design a circuit for you. It would be very useful to know more about your setup though.

What do you know about the voltage input? What's its input resistance? What happens if you apply more than 5V to it? Try connecting, say, +9V from a battery through a 10k resistor to the input. That's very unlikely to damage it because the resistor will limit the current. What voltage do you measure on the input? Does the shaft stop turning?

What power sources do you have available? Can you get +5V from the shaft controller box? Can you get any other higher voltage supplies from it?

Can you provide manufacturer names and model numbers for the voltage source and the shaft control box? Links to data sheets?

 

[skippyV]

Hi Skippy. As you say, a pullup resistor would work if the connection is broken, but it may not work if the voltage source loses power.

I may not have described that portion accurately.

The box that controls the spin rate of the shaft is kind of a 'voltage/current-stepper-upper'? (Excuse my sad terminology).
The shaft required a certain level of power. By using this box we could vary the -5 to +5 input to the box which would then fluctuate the power to the shaft accordingly. And since we spin the shaft in only one direction - we only changed the input signal from 0 to 5 volts.

Right now I don't have specific parameters of the box on hand. What I remember from what little documentation it had was that the input signal can not exceed 5 volts.

So we used a simple analog output device and limited it's output from 0 to 5v. And used it to drive the box. However once we powered on the box, we had to have a 5 volt signal already driving the input to keep the shaft stationary. Then we'd reduce the analog signal to start the rotation.

So I should be more specific when I refer to a voltage source. If the source that is driving the analog circuit fails - but the source used by the box is still on - then the shaft will go full speed. But if the pull-up resistor is tied to the same power source used by the box - than 5 volts would be there and stop the shaft from rotating.
No power to the box - no power to the shaft or the pull-up resistor. No motion -> all good.

As for input resistances I can find out more about the box's tomorrow when I look at the single page of documentation we have for it.

But the analog circuit we were planning on replacing anyways. What we were working with didn't necessarily lose power - but would lose communication. Which had the same effect of losing power - sort of. So we're trying to design now for all circumstances. Our new analog source will be driven by a RMC75E module from Delta Motion. Which should eliminate the 'dropped communication' scenario but because of this backward model used by this "box" - we gotta ensure something doesn't happen if power to the signal dies and the shaft just starts spinning unexpectedly.

Because if the signal is at zero - the shaft is spinning at full speed.

As the previous gentleman said - it's a poor design. But we spent $300 on this box and we're crunched for time - so kind of stuck with it.

Now with the RMC module the power source to the signal will be completely separate from the power source used by the box (and pull-up resistor if that is the direction we take). But the potential is there to create some simple circuit to integrate with the signal-to-box connection.

 

[KrisBlueNZ]

Can you draw up a diagram please. Include as much information as you can - part numbers, especially.

 

[skippyV]

Yes, I'll do that. Will try to post it by tomorrow night. Thanks for the interest and help, Kris

 

[Gryd3]

I'm looking forward to the diagram/schematic...
My concern is how you say the source is failing... If you loose power both devices go down, no problem. If the source gets disconnected the pull-up resistor will present 5V to the 'box'.
If the source fails... ie locks up, glitches, etc. and it shows 0V, it sounds very likely to me that it will pull the voltage on the line to 0V even with the pull-up resistor.

The other item that catches my attention... is that this box used to operate from -5 to 5... but was changed to 0-5 as only one direction is desired. It sounds to me that this change 'broke' something.. as 0V is full tilt and 5V is off.. what was the -5V used for? It may be worthwhile to revert back to the -5 to 5 model if it means that -5 is one direction, 0 is off, and 5 is the other direction. These are just observations I have made though. I'll look again when more info is presented.

 

[KrisBlueNZ]

The other item that catches my attention... is that this box used to operate from -5 to 5... but was changed to 0-5 as only one direction is desired. It sounds to me that this change 'broke' something.. as 0V is full tilt and 5V is off.. what was the -5V used for? It may be worthwhile to revert back to the -5 to 5 model if it means that -5 is one direction, 0 is off, and 5 is the other direction.

Yes, I was wondering about that as well. I didn't follow the description very well but that part seemed more confusing than the rest. I hope the diagram will clarify this!

 

[skippyV]

Hi Guys. Unfortunately I wasn't able to get a hold of that box's spec sheet. Our RMC unit came in today while some unexpected things happened (e.g. forklift broke and home-brew repairs were attempted). I'll try again tomorrow. Tomorrow being a holiday there should be less activity and more opportunity for me to rummage around.

You see I'm just the "software guy" so I have to tread lightly in areas that are not in "my domain".
But I'm privy to much of the design. And I remember my almost forgotten days of electronic hobbying to remember that message boards like this can provide very valuable knowledge. Too often I see people who don't ask for help because... I guess egos get in their way. I don't have that problem.

As for the +5 to -5 range, we need the shaft to spin in only one direction because of its purpose. But that does lead to the question of - if it's at full speed at zero, what would happen if it went a little negative? Would it suddenly reverse speed in the opposite direction? Guess it'd have to. Not a good idea from a mechanical perspective. But as far as the box is concerned it's just voltage polarity.

When we originally got the box, the guy who spec'd it and dreamed up the connections was just going to use the box in a different configuration. Because the spec sheet said - hook it up one way and the input of zero would create an output of zero. Hook it up a different way (the way we ultimately did) and the logic is reversed. He originally wanted it to be zero in -> zero out. But we couldn't get it to work that way for some unknown reason.
So he called the manufacturer and told the tech there about it - and the tech said try it the other way. I know. Crazy. But we were under the gun and had to make it work "yesterday". So I changed the software to tell the analog output to drive +5 volts for stationary and... you know the rest.

As for writing this out on a diagram - I'm game to building what I know. I doubt it will tell you much more but sometimes a picture helps. The box's spec sheet would undoubtedly help the most.

Is there a free and easy-to-use app that you guys use to throw together a circuit diagram?

 

[Gryd3]

LTspice was recommended in another post, and I have yet to play with it in too much depth, but I like it, and it's free!

 

[KrisBlueNZ]

A bit of Googling turned up these possibilities. They are all free.

ProfiCAD: http://www.proficad.com/ - looks ideal. The "home license" version is free, and has some fairly minor restrictions.

Google/Trimble Sketchup: http://sketchup.google.com/intl/en/product/gsu.html - easy to use; many tutorials

DraftSight: http://www.3ds.com/products/draftsight/overview/ - free; compatible with AutoCAD

DraftIt: http://www.cadlogic.com/products/draftit/featurecomparison.aspx?Product=DRAFTIT - "very basic"

Visio: http://microsoft-visio-2010.en.softonic.com/ - trial version only, I think - user rating only 5.9/10

LTSpice is excellent, but it is really only for electronic schematics. It's not appropriate for general block diagrams.

Please let us know your opinion of anything you try out.

 

[Gryd3]

A bit of Googling turned up these possibilities. ...
DraftSight: http://www.3ds.com/products/draftsight/overview/ - free; compatible with AutoCAD
...

I highly recommend DraftSight. Free to use Commercially or personally.
Has almost all of the same commands as AutoCAD and can import and export DXF, DWG, and many other formats.
I've introduced this program to a couple of cabinet shops in my area who don't want to buy AutoCAD

 

[skippyV]

Here is a primitive attempt of a block diagram using DraftSight.
I like the program - just need more time to learn it.
I combined the before and after models into one picture.
The first version has no RMC unit in it. The new plan has the RMC and will not have the SBC or its analog out device.
The RMC will have its own analog out.
And I finally got my hands on "The Box"'s spec sheet. It's a little rough for wear.
So I'm uploading a pdf for the crude block diagram and a pic of the spec sheet I took with my cell phone.

 

[KrisBlueNZ]

Sorry about the slow response. This issue is becoming a bit complicated.

So the motor controller is a Midwest Motion Products MMP 25A? The one described at http://www.midwestmotion.com/products/speedcontrols/25A-24V.html

You mentioned the Delta Motion RMC75E. This is documented at http://www.deltamotion.com/products/motion/rmc70/cpu75e.php. Are you replacing the MMP 25A with the RMC75E and a new motor controller board?

What does "SBC" (in post #15) mean?

What have you found out from using the RMC75E that arrived about ten days ago?

What's the overall status of this project?

 

[Gryd3]

Sorry about the slow response. This issue is becoming a bit complicated.

So the motor controller is a Midwest Motion Products MMP 25A? The one described at http://www.midwestmotion.com/products/speedcontrols/25A-24V.html

You mentioned the Delta Motion RMC75E. This is documented at http://www.deltamotion.com/products/motion/rmc70/cpu75e.php. Are you replacing the MMP 25A with the RMC75E and a new motor controller board?

What does "SBC" (in post #15) mean?

What have you found out from using the RMC75E that arrived about ten days ago?

What's the overall status of this project?

According to his Draftsight Diagram (PDF) The SPC appears to be a 'Single Board Computer'.


Just so we get things perfectly straight here (as it sounds as though there is near 0 room for error)

Current setup is an SBC with an analogue out circuit attached that is fed into 'The Box' which accepts 0-5 VDC as an input. The current analogue circuit is wired into the 'W' connection on the box.
It is 'The Box' that currently runs the motor at full tilt when 0VDC is present on the 'W' line? When you feed it 5VDC on the 'W' line it stops the motor?

You plan to remove the SBC and it's analog circuit and replace it with a C# app on a different computer that will feed into the RMC. The RMC is capable of putting out +-10VDC as an analog output... I am assuming you are programming your app to make the RMC provide 0-5VDC to 'The Box' exactly like the original setup?

From the sounds of it... although 'The Box' design is half-assed backward, you are more interested in implementing a safety feature more than anything.

Can you please clarify if this is the case?


Also, some additional info.
You want this safety feature to be present for both setups?
Can you post any info on the analogue circuit currently in use?
Can you please let us know the current expected use of 'The Box' and the shaft transducer? Do you ever have it run at 100% or is it generally below that limit?

 

[(*steve*)]

Sorry for poking my head in here, but the 5V pullup should be on the controller end powered from the same source as the controller that will make the motor run at full speed. If the 5V here fails, nothing will work.

Of course, a failure causing the remote end to short out will still cause full speed.

 

[Gryd3]

Sorry for poking my head in here, but the 5V pullup should be on the controller end powered from the same source as the controller that will make the motor run at full speed. If the 5V here fails, nothing will work.

Of course, a failure causing the remote end to short out will still cause full speed.

The 5V pullup would need to be placed between the Analog Circuit and 'The Box' to pull the voltage to 5V if the analogue circuit fails.. Part of the problem I think is that no one here knows exactly how it fails or has not said yet. From what I understand, the SBC provides input to the Analogue circuit currently in use, which in turn feeds into 'The Box'... If the SBC fails, the analogue circuit will most likely pull the voltage to 0, even with a pullup resistor... If the Analogue circuit fails, the line would either need to be open, or in a high impedance state for a pullup resistor to work. Otherwise you end up with a voltage divider that locks the motor at a speed other than full, or a 'frozen' analogue circuit that keeps the line pulled low...
It would be really nice to get our hands on more details from the Analogue circuit currently in use, as there may be something else in there we can tap into for a safety to lock that line at 5V, or cut power from 'The Box'.
Of course that solution would only help for the analogue circuit currently in play, and most likely would not work with the new RMC that they plan to use to replace the SPC/AnalogueCircuit.

 

[skippyV]

Gyrd3's assessment is correct. Yes to both of your question, Gyrd. And the analogue circut (and it's accompanying SBC (not SPC -> typo)) has been removed from the picture. And I've been using the RMC device with the intent on it providing the same output to 'The Box' as you so aptly described, Gyrd. (Perhaps we should refer to it as the MMP since Kris found the source - thanks for that Kris. I should have looked it up myself but I've been burning the candles at both ends.)
To provide some more background info without saying too much that might possibly tick off my employer - this thread was begun on my own time in an attempt to come up with solutions to this problem of outputting "full power" in case whatever is driving the MMP suddenly stops driving it. Causing the shaft to receive 0 volts and sending it into full speed. The situation is even more complicated from our production environment. I'm the software guy. The last time I asked a question from the "hardware-guy" in an attempt to better understand a different component, he said "that's not your problem - how does that affect you?!" Even though my software may not be affected by any particular hardware implementation change - I like to understand as much of "everything" that I can. But I'm old enough to know you can't always help people with solve their problems. Especially people with particular... let's call them personality aspects.
At present I've been working 7 days a week to get the software working to the latest set of requirements - which change every time there is a discussion. Sometimes multiple times a day! (Try to capture that kind of decision making in a document!).
And I have to be done with one revision that adheres to the latest "requirements" (I quoted that because they are not documented - just discussed) by the morning.
So I'm happy that you guys have given your time and thoughts to this problem. It helps buoy my spirits in reflection of human nature and the idea that at some fundamental level - we do want to try and help each other.
And if you had an idea to a solution I'd be curious to know it because I like to learn! But I wouldn't dare propose it to the 'hardware-guy' because in his eyes I don't know.... never mind.
So if you decide to let this thread die that's perfectly understandable.

I have to go clock in now and get coding.
Thanks to all you guys for all your input.

Skippy