Topics Topics Edit Profile Profile Help/Instructions Help Member List Member List  
Search Last 1|3|7 Days Search Search Tree View Tree View  

Analog 0-10v

:: EPE Chat Zone ­:: ­Radio Bygones Message Board :: » EPE Forum Archives 2007-2009 » Archive through 28 November, 2008 » Analog 0-10v « Previous Next »

  Thread Last Poster Posts Pages Last Post
  ClosedClosed: New threads not accepted on this page        

Author Message
Top of pagePrevious messageNext messageBottom of page Link to this message

individual
Frequent Contributor
Username: individual

Post Number: 88
Registered: 05-2005

Rating: N/A
Votes: 0 (Vote!)

Posted on Friday, 14 November, 2008 - 06:57 am:   Edit Post Delete Post Print Post    Move Post (Moderator/Admin Only)

Hi....
What is the best way to filter and convert a 0-10v analog signal, to a 0-5v so the pic can handle it?

Do resistor dividers do the job?Or will I be needing more than that?
Top of pagePrevious messageNext messageBottom of page Link to this message

chuckieboy
Frequent Contributor
Username: chuckieboy

Post Number: 121
Registered: 05-2005


Rating: N/A
Votes: 0 (Vote!)

Posted on Friday, 14 November, 2008 - 07:00 pm:   Edit Post Delete Post Print Post    Move Post (Moderator/Admin Only)

Yes you can use a resistor divider as long the little extra current draw will not affect your application. Been only 10 volts this should be to much of a problem or you could use an op-amp has a buffer/divider if it a nosiy voltgae
Top of pagePrevious messageNext messageBottom of page Link to this message

individual
Frequent Contributor
Username: individual

Post Number: 89
Registered: 05-2005

Rating: N/A
Votes: 0 (Vote!)

Posted on Saturday, 15 November, 2008 - 04:04 pm:   Edit Post Delete Post Print Post    Move Post (Moderator/Admin Only)

What I really want is a nice 0-10v input, and a 0-5v output to feed it to the PIC's A-D.
Immune to emi, and reliable.

Any ideas?
Top of pagePrevious messageNext messageBottom of page Link to this message

individual
Frequent Contributor
Username: individual

Post Number: 90
Registered: 05-2005

Rating: N/A
Votes: 0 (Vote!)

Posted on Sunday, 16 November, 2008 - 07:35 am:   Edit Post Delete Post Print Post    Move Post (Moderator/Admin Only)

In other words.
I want to interface my PIC to a 0-10v output from a PLC, either by using the pic's A-D or by reading the 0-10v digitally by using another A-D.

What is the best way to do this?
Top of pagePrevious messageNext messageBottom of page Link to this message

hackinblack
Frequent Contributor
Username: hackinblack

Post Number: 324
Registered: 09-2006

Rating: N/A
Votes: 0 (Vote!)

Posted on Sunday, 16 November, 2008 - 11:46 am:   Edit Post Delete Post Print Post    Move Post (Moderator/Admin Only)

here is a useful page,on signal conditioning
http://www.industrologic.com/aninput.htm

remember the inputs must be filtered to avoid any mains hum and high frequency stray affecting the actual voltage;ferrite beads in the leads and a RF shielded cable (co-Ax) and case (metal)

A2D accuracy can also be increased by oversampling and then discarding several bits (it is mentioned in several Microchip App.notes)
Top of pagePrevious messageNext messageBottom of page Link to this message

gajjer
Frequent Contributor
Username: gajjer

Post Number: 163
Registered: 05-2007

Rating: N/A
Votes: 0 (Vote!)

Posted on Monday, 17 November, 2008 - 05:46 pm:   Edit Post Delete Post Print Post    Move Post (Moderator/Admin Only)

Hi individual
you really need to be more specific about what you are monitoring. If its a temperature sensor then its not going to change very quickly and you can afford to use a large time constant for the low pass filter. On the other hand if its an audio signal its going to be a much smaller time constant.

Also what environment it is operating in. Under the bonnet in a car is going to be noiser than in a metal box in a field. How long are the leads to the signal source? If its a pot in the same box as the pic, you don't have much problem, but if its 100yds away then you will need to do more.


Can't really advise without more information.

cheers
gaj
Top of pagePrevious messageNext messageBottom of page Link to this message

individual
Frequent Contributor
Username: individual

Post Number: 91
Registered: 05-2005

Rating: N/A
Votes: 0 (Vote!)

Posted on Tuesday, 18 November, 2008 - 08:26 am:   Edit Post Delete Post Print Post    Move Post (Moderator/Admin Only)

Thanks for your responce gajjer.
To be more specific, I'm going to monitor the output of a pressure sensor (10 bars, 24vdc, 0-10v output) which is going to to be installed in a screw type air compressor.
The distance is like 2 feet from the sensor to my pcb, and of course , the signal cables will be isolated from the power and noisy cables.
I want to read this 0-10v using the PIC's A/D, so I needed a reliable way to filter and convert it to a 0-5v signal.

Any ideas?
Top of pagePrevious messageNext messageBottom of page Link to this message

gajjer
Frequent Contributor
Username: gajjer

Post Number: 164
Registered: 05-2007

Rating: N/A
Votes: 0 (Vote!)

Posted on Tuesday, 18 November, 2008 - 05:53 pm:   Edit Post Delete Post Print Post    Move Post (Moderator/Admin Only)

Hi individual
I would keep it pretty simple. Not sure what accuracy you are after but with a fan you might find that you pick-up pressure pulses from the fan.
Personally I'd start with a divider made up from two 10k resistors with >0.1uF across the lower resistor.
The cap being fitted close to the input of the PIC.
That still passes frequencies >200Hz and personally I would think you need it lower than that. Check it out with a scope if you can. I'd use twin screened wire to the sensor and ground the screen at the PIC end only.

If you need to use a polarised capacitor be careful that the leakage is low for the one you choose.

As I say you may see pulsing because of the fan blades so it is difficult to say what cut-off frequency you need. If you see ripples because of the fan you could average in software. But that will reduce the response time in addition to the low pass.

Hope that helps
cheers
gaj
Top of pagePrevious messageNext messageBottom of page Link to this message

individual
Frequent Contributor
Username: individual

Post Number: 92
Registered: 05-2005

Rating: N/A
Votes: 0 (Vote!)

Posted on Wednesday, 19 November, 2008 - 07:48 am:   Edit Post Delete Post Print Post    Move Post (Moderator/Admin Only)

Thanks gajjer. I'm gonna take your advice and start with the resistor divider.(simplest).

But, are there any other reliable ways that could do the job, I mean if I need higher accuracy, and a more filtered out signal, what would be the best soloution then?
Top of pagePrevious messageNext messageBottom of page Link to this message

gajjer
Frequent Contributor
Username: gajjer

Post Number: 165
Registered: 05-2007

Rating: N/A
Votes: 0 (Vote!)

Posted on Wednesday, 19 November, 2008 - 05:47 pm:   Edit Post Delete Post Print Post    Move Post (Moderator/Admin Only)

Hi individual
the problem you have with the divider is that the leakage current in the PIC ADC is +/- 500nA which will be temperature dependent too. Thats why I suggested relatively low values in the divider. The leakage will be like a resistor in parallel with the bottom resistor of the divider. If it was fixed you could perhaps compensate for it. But it wont be.
So the next thing you could do is to put an op amp buffer in between the divider and the PIC. So you could use a 5V rail to rail op amp. That is one that can have both its inputs and output go from 0V to 5V. Your classic op amp will not have that ability and some of the rail to rail ones don't have a particularly high drive capability. But enough to overcome the leakage of the ADC.
You could then use bigger resistors in the divider. And if you really wanted to get the frequency response to roll off faster you could use a two stage RC or even a Sallen Key filter ( google it ).
I would hold fire though. Think about what you are measuring. Fan pressure is not a precise measurement. It depends where the sense tube is situated. Pressure will vary across the outlet area of the fan. Unless you stick it into a plenum chamber to filter out the pressure variation, you aren't really sure what you are measuring.
And is fan pressure what you really want to know, or is it mass air flow. In which case, pressure is not a very good way of doing it.

You need to give more information about what you are trying to achieve.

cheers
gaj
Top of pagePrevious messageNext messageBottom of page Link to this message

chuckieboy
Frequent Contributor
Username: chuckieboy

Post Number: 122
Registered: 05-2005


Rating: N/A
Votes: 0 (Vote!)

Posted on Wednesday, 19 November, 2008 - 09:00 pm:   Edit Post Delete Post Print Post    Move Post (Moderator/Admin Only)

individual read what Gajjer has said and read it again to make sure you understand it becasue you seem to be asking the same queation everytime someone replies, either what Gajjer has said or like I said op-amp has buffer and divider, how close do you want to get the PIC can only do a 10bit resolution if you need a higher accuracy then why not use a A/D converter say with 22 or 24 bit resolution then you would be close

Why not do a Goolge there are plenty of samples and ways of doing it
Top of pagePrevious messageNext messageBottom of page Link to this message

zeitghost
Frequent Contributor
Username: zeitghost

Post Number: 1233
Registered: 01-2006

Rating: N/A
Votes: 0 (Vote!)

Posted on Thursday, 20 November, 2008 - 08:17 am:   Edit Post Delete Post Print Post    Move Post (Moderator/Admin Only)

You can get 22 bit converters that are monotonic now?

I'd have thought that 10 bit was adequate for measuring pressure...
Top of pagePrevious messageNext messageBottom of page Link to this message

individual
Frequent Contributor
Username: individual

Post Number: 93
Registered: 05-2005

Rating: N/A
Votes: 0 (Vote!)

Posted on Thursday, 20 November, 2008 - 08:17 am:   Edit Post Delete Post Print Post    Move Post (Moderator/Admin Only)

Thanks for your responses guys.

The application is about measuring air pressure, in an air compressor.There is no fan, and didn't mention a fan in the first place.(or did I?)
So, the air is compressed in a 50 litre chamber, which I'd like to measure its pressure.
To make things simple, I bought a pressue transducer (max. 10 bars), which is powered from a 24dc supply, and has an output span of 10 volts. (0 bar = 0 volts, 10 bars =10 volts).
This 0-10v signal output is pretty well conditioned and filtered, and I want to read it using the PIC.
That's all.
I hope that I made things clearer now.

Why I want to work on the 0-10v signal? Because it is the industrial control standard just like the 4-20mA signal.

I can use this 0-10v signal readout to measure other things as well, (VFD speed feedback, proportional valve position, reading the outputs of other sensor transducers).

I hope that I succeeded in explaining my point of view.
Top of pagePrevious messageNext messageBottom of page Link to this message

gajjer
Frequent Contributor
Username: gajjer

Post Number: 170
Registered: 05-2007

Rating: N/A
Votes: 0 (Vote!)

Posted on Thursday, 20 November, 2008 - 05:35 pm:   Edit Post Delete Post Print Post    Move Post (Moderator/Admin Only)

Hi individual
yep my mistake you said compressor. I'd still go with the simple divider. The tank will act as a low pass filter and the transducer should give a pretty good indication of actual pressure. The same thing should be good for the other things you are monitoring too.

Let us know how you get on.

cheers
gaj
Top of pagePrevious messageNext messageBottom of page Link to this message

hackinblack
Frequent Contributor
Username: hackinblack

Post Number: 330
Registered: 09-2006

Rating: N/A
Votes: 0 (Vote!)

Posted on Thursday, 20 November, 2008 - 06:01 pm:   Edit Post Delete Post Print Post    Move Post (Moderator/Admin Only)

0 to 10V through a resistor divider to give 0 to 5 volt swing; a 10 bit A2D e.g. 12F675 will give 256 steps divided by 5 (effectively)giving 51.2mV per volt or per bar.

a screw type compressor will give a steadier increase in pressure over a piston type (which would send the sensor haywire with pulsations;especially as the receiver is such a small volume one)
though it would,i imagine,still need some software 'damping' to give steady readings; taking an average over several samples and sampling several seconds apart.

a thing to remember is most pressure meters have mechanical/oil damping to prevent needle flutter;unless it's a 'clever' electronic sensor this wont be included...check the output voltage with a sensitive voltmeter or ideally a scope;and see if the reading rises smoothly or erraticaly,this should give you an idea of how much tweaking you need to do in the firmware.

Administration Administration Log Out Log Out   Previous Page Previous Page Next Page Next Page