The Steampunk Forum at Brass Goggles
October 18, 2017, 09:34:05 pm *
Welcome, Guest. Please login or register.
Did you miss your activation email?

Login with username, password and session length
News: Brassgoggles.co.uk - The Lighter Side Of Steampunk, follow @brasstech for forum technical problems & updates.
 
   Home   Blog Help Rules Login Register  
Pages: [1]   Go Down
  Print  
Author Topic: Converting mechanical gauges to electric?  (Read 4715 times)
alfa1
Zeppelin Captain
*****
Australia Australia



« on: January 07, 2009, 09:27:35 am »

Electrical gauges seem to only have about 70 degrees of needle travel, whereas mechanical gauges (pressure gauges etc...) appear to have at least 270 degrees of needle movement.    I think they look better, and also a mechanical gauge seems 'older'.

A year or two ago I created an electrical powered gauge with a mechanical appearance for my http://www.brasswings.com/autocalc.htmlcasemod by using a small DC motor that wound up a spring.    It sortof worked but not very well.   The DC motor likes to be running fast or not at all, which means the needle was never reliably in the same place any time I turned the machine on.   A shame, because I wanted to use it to display actual case temperatures.
So heres my question...   before I put my wireless router in a new case and display traffic usage on a gauge, does anyone have a good mechanism/way of modifying a real actual old gauge to be run on small electrical currents?
Logged
Professor Fzz
Zeppelin Admiral
******
United Kingdom United Kingdom



WWW
« Reply #1 on: January 07, 2009, 10:23:00 am »

I used a servo motor to drive the original mechanical linkage when I did this last year.  It works pretty well though it isn't silent, and you have to be careful not to drive the gauge upwards too rapidly or it gains too much momentum and shoots pass the target value crashing into the stop. 

Here's a picture:

« Last Edit: January 07, 2009, 10:25:34 am by Professor Fzz » Logged

Try to learn something about everything and everything about something. -  Thomas H. Huxley
alfa1
Zeppelin Captain
*****
Australia Australia



« Reply #2 on: January 07, 2009, 10:54:42 am »

I used a servo motor...


Actually I had considered using a servo motor, but for no specific reason, I had reservations about a servo motor buzzing away continuously 24 hours a day 7 days a week for probably several years on end.
I might end up doing that anyway if I cant find any other method.
Logged
Professor Fzz
Zeppelin Admiral
******
United Kingdom United Kingdom



WWW
« Reply #3 on: January 07, 2009, 11:36:52 am »

Actually I had considered using a servo motor, but for no specific reason, I had reservations about a servo motor buzzing away continuously 24 hours a day 7 days a week for probably several years on end.
I might end up doing that anyway if I cant find any other method.

Yes, I share your concern.  Mine has worked fine since October, but I don't know whether the servo is taking any significant wear or not.  I'll probably only know when it fails.  I did some fine tuning so that the servo is silent at the zero position on the scale.  In principle a servo shouldn't draw any power (at least not via the motor) when it's not under load and is at the target position.  In practice, sometimes they buzz, even when they're at the target position - presumably hunting back and forwards.  So you want to make sure the zero point on the scale is at a stable target position for the servo where it isn't buzzing.

Then I added a little damping in software - the pointer isn't moved from the zero position for very low traffic levels - only things that would actually be visible on the meter.  This avoids constant servo movement from background chatter on an otherwise silent network.  On my home DSL link, the meter is at zero 95+ % of the the time.  If you ran a lot of P2P apps all the time though, the servo would have a lot more work to do.

In general, any motor-based solution is going to be running a motor whenever the gauge needs to move.  A servo is just a motor with a built in control loop, so it's probably not too different.

I did think of using an electromagnet instead of the servo, but I didn't have one handy.  If the servo dies, that's plan B.  I can drive an electromagnet just as easily from USB (over an I2C bus via a DC motor control board I've got lying around).  The advantage of an electromagnet is that it has no gears to wear out, so should be longer lasting.  The disadvantage is it needs an external power supply, whereas the servo draws so little current I can power it off USB.
Logged
JingleJoe
Zeppelin Overlord
*******
United Kingdom United Kingdom


The Green Dungeon Alchemist


WWW
« Reply #4 on: January 07, 2009, 05:46:10 pm »

You can get electric meters with over 270o of movement, I have one right here above my desk in my mad scientist lamp Smiley it measures 0 to 260 volts AC Grin

Logged

Green Dungeon Alchemist Laboratories
Providing weird sound contraptions and time machines since 2064.
Professor Fzz
Zeppelin Admiral
******
United Kingdom United Kingdom



WWW
« Reply #5 on: January 07, 2009, 06:03:13 pm »

You can get electric meters with over 270o of movement, I have one right here above my desk in my mad scientist lamp Smiley it measures 0 to 260 volts AC Grin

I'm curious - how does it work?  Normal meters use a coil between the poles of a permanent magnet, which fundamentally limits you to not much more that 90o of movement.  So yours must be doing something different.
Logged
JingleJoe
Zeppelin Overlord
*******
United Kingdom United Kingdom


The Green Dungeon Alchemist


WWW
« Reply #6 on: January 07, 2009, 06:24:14 pm »

Nope, I'm looking at it now and I'm pretty certain it has the usual magnet and coil, here have a look, it's missing the glass front so it's easy to see in;

« Last Edit: January 07, 2009, 06:26:47 pm by JingleJoe » Logged
Jake of All Trades
Zeppelin Admiral
******
United States United States


Brass addict, inventor, and scoundrel with a heart


WWW
« Reply #7 on: January 07, 2009, 06:24:47 pm »

*Sigh* I guess I'll share my secret...

Automobile tachometers!  The modern ones are all electric, varying needle position in proportion to the frequency of a square wave put in on their input pin Smiley
Logged

"...it's a form of fiction, and as such, while there may be times when it's considered a worthy vehicle for pointing out some of society and individual flaws - I still want a side that will let there be lighthearted adventures in the clouds, on mars, or under the sea."
--Tinkergirl
Professor Fzz
Zeppelin Admiral
******
United Kingdom United Kingdom



WWW
« Reply #8 on: January 07, 2009, 06:34:43 pm »

*Sigh* I guess I'll share my secret...

Automobile tachometers!  The modern ones are all electric, varying needle position in proportion to the frequency of a square wave put in on their input pin Smiley

Genius!  And suddenly the price of tachometers jumps on Ebay...
Logged
Jake of All Trades
Zeppelin Admiral
******
United States United States


Brass addict, inventor, and scoundrel with a heart


WWW
« Reply #9 on: January 07, 2009, 06:41:39 pm »

*Sigh* I guess I'll share my secret...

Automobile tachometers!  The modern ones are all electric, varying needle position in proportion to the frequency of a square wave put in on their input pin Smiley

Genius!  And suddenly the price of tachometers jumps on Ebay...
Haha, it's all part of my economic stimulus program.  Don't forget to check junkyards too!
Logged
alfa1
Zeppelin Captain
*****
Australia Australia



« Reply #10 on: January 07, 2009, 09:13:11 pm »

Voltmeter...
When I was a kid, I had an old voltmeter that used two plates close together.  One fixed, one moving and attached to the needle.    A bit of reading tells me now it is called an 'electrostatic' voltmeter.    Its probably the type jinglejoe has, but is really only useful to show high voltages.
http://knowledgepublications.com/history/images/Mechanical_Movements_Electrostatic_Voltmeter.gif

Tachometer...
Seems that these devices, and speedometers, use an internal electric motor with magnet attached to induce eddy currents into the metal attached to the needle.
http://www.tpub.com/content/construction/14273/css/14273_87.htm
http://auto.howstuffworks.com/speedometer3.htm
http://www.innerauto.com/Automotive_Systems/Drive_Train/Speedometer~odometer/
Maybe this is the way to go.   To insert the mechanism from an automotive meter into a pressure gauge.

Logged
Jake of All Trades
Zeppelin Admiral
******
United States United States


Brass addict, inventor, and scoundrel with a heart


WWW
« Reply #11 on: January 07, 2009, 10:48:03 pm »

Tachometer...
Seems that these devices, and speedometers, use an internal electric motor with magnet attached to induce eddy currents into the metal attached to the needle.
http://www.tpub.com/content/construction/14273/css/14273_87.htm
http://auto.howstuffworks.com/speedometer3.htm
http://www.innerauto.com/Automotive_Systems/Drive_Train/Speedometer~odometer/
Maybe this is the way to go.   To insert the mechanism from an automotive meter into a pressure gauge.

I don't think many, if any, modern tachs work this way.  All that I've seen are either mechanical or they use air-core "motors".  This gentleman gives a lovely rundown of this technology, as well as a way to make your own.  ON Semiconductor makes a lovely chip for controlling air-core instruments, and they have samples! 

Now the mechanical speedo's and tach's work on the induction principle you cited, but the motive force comes from a spinning cable connected to the drive train.  I once tried to motorize a similarly-constructed bicycle speedo, but the motor needs to spin rather quickly and the contraption turned out to be as prohibitively noisy as the servo method.  Note also that such an arrangement is not as precise, easy to control, or as responsive as an air-core (or servo, for that matter) setup.
« Last Edit: January 07, 2009, 10:51:10 pm by Jake of All Trades » Logged
alfa1
Zeppelin Captain
*****
Australia Australia



« Reply #12 on: January 07, 2009, 11:53:04 pm »

I don't think many, if any, modern tachs work this way.  All that I've seen are either mechanical or they use air-core "motors".  This gentleman gives a lovely rundown of this technology, as well as a way to make your own. 



Actually I'd briefly considered going down this path as well, even though I didnt realise that this was the technology inside modern instruments.    A few years ago I made a laser display unit, with home made galvo's that are actually air core motors with a feedback mechanism.
http://www.concordeagreement.com/galvo.html   
So I still have wire and appropriate cylindrical magnets.   It is mechanically fiddly though, and I'd have to spend time and money on making a custom built circuit to run it.   I'd initially dismissed it out of hand because I thought I'd need an extra beefy power supply to get a usable amount of current through the coils, but now I think maybe not.
Worth a second look. 

Another crazy idea I had overnight was better for packaging inside a gauge.   If I built an air core rotor with very low friction, and it rotated towards the pole of an electomagnet run from a low current like 100mv, then I could just wind 10 of those electromagnets and arrange them in a radial fashion around the core and run each of them off an output from a Dot/Bar driver chip (like a LM3914).
Quick test: 300 turns of SWG33 wire around a small wooden former, running a measly 30mA, does indeed exert a usable amount of force on a cylindrical magnet of the type used in aircore motors.   Now, how does one optimise an electromagnet?

As always, feedback is appreciated.     If I get a good solution, I'll probably end up using it in every electronic project from now on.    Gotta have gauges.
« Last Edit: January 08, 2009, 12:36:13 am by alfa1 » Logged
Jake of All Trades
Zeppelin Admiral
******
United States United States


Brass addict, inventor, and scoundrel with a heart


WWW
« Reply #13 on: January 08, 2009, 12:38:30 am »

A few years ago I made a laser display unit, with home made galvo's that are actually air core motors with a feedback mechanism.
http://www.concordeagreement.com/galvo.html   

Blimey, that's tops!  I bet one of those ON Semi chips'll control your motors with little trouble.  They work dandily with the Pontiac units I'm using, and they don't appear to have much more manufacturing precision than yours Wink

I'm thinking a straight stepper motor (from a CD drive, perhaps) might work too, though some feedback mechanism would be necessary to avoid major positioning headaches...
Logged
Mr. Consciousflesh
Snr. Officer
****
Poland Poland



WWW
« Reply #14 on: January 08, 2009, 11:20:28 am »

I'm thinking a straight stepper motor (from a CD drive, perhaps) might work too, though some feedback mechanism would be necessary to avoid major positioning headaches...

I think this is a bad idea ... The main motor in the CD drives is not exactly a stepper motor but a brushless DC motor . It has rather large "steps" making it very hard to control . The second motor - the one used for positioning the laser head , is better but usually requires external bearing ( its shaft is supported by the laser assembly ) .

My suggestion is to use a normal , brushed  DC motor with a clock spring on its shaft. The torque of the motor is proportional to the current and the torque generated by the spring is proportional to the torsion angle so you can control it by varying current flowing through the motor.
Logged

The reason we chase is lost in romance.
And still we try to justify the waste for a taste of man's greatest adventure.
Professor Fzz
Zeppelin Admiral
******
United Kingdom United Kingdom



WWW
« Reply #15 on: January 08, 2009, 01:28:08 pm »

My suggestion is to use a normal , brushed  DC motor with a clock spring on its shaft. The torque of the motor is proportional to the current and the torque generated by the spring is proportional to the torsion angle so you can control it by varying current flowing through the motor.

Won't you get discontinuities in the current and hence the torque, as different segments of the motor come under the brushes?  I'd have thought this might work better with a geared motor, so you have a number of turns of the armature to drive the pointer to full scale deflection.
Logged
Jake of All Trades
Zeppelin Admiral
******
United States United States


Brass addict, inventor, and scoundrel with a heart


WWW
« Reply #16 on: January 08, 2009, 08:18:42 pm »

My suggestion is to use a normal , brushed  DC motor with a clock spring on its shaft. The torque of the motor is proportional to the current and the torque generated by the spring is proportional to the torsion angle so you can control it by varying current flowing through the motor.

Won't you get discontinuities in the current and hence the torque, as different segments of the motor come under the brushes? 
Yes, you certainly will.  I abandoned that method after discovering that problem for myself, as (I suspect) did Mr. Alfa1. 

The second motor - the one used for positioning the laser head , is better but usually requires external bearing ( its shaft is supported by the laser assembly ) .
D'oh!  That's the one I was thinking of, but I forgot about its single bearing...
Logged
Mr. Consciousflesh
Snr. Officer
****
Poland Poland



WWW
« Reply #17 on: January 08, 2009, 11:26:16 pm »

@Professor Fzz  :
    I was using the method with the spring and DC motor to move a light sensor and it was working quite well . There were no discontinuities in the current , only some nonlinearities caused by the commutator . I believe it was working because I used a rather large motor ( probably over 30 Watts  ) . Your idea of using a geared motor should improve operation of this kind of actuator.
   
Logged
alfa1
Zeppelin Captain
*****
Australia Australia



« Reply #18 on: January 31, 2009, 04:16:19 am »

Update...   as of a few minutes ago I now have a functional (breadboarded) working circuit of an aircore motor, controlled by a variable voltage into a CS8190 chip.

The homebrew motor.   A cylindrical magnet from supermagnetman on a bamboo shaft, through some small ball bearings from a computer fan, set inside a brass tube, and wound with wire that was too large but was what I had lying around but works anyway...


Thanks for all suggestions.    Work on ripping my wireless router apart can now commence.
Logged
Jake of All Trades
Zeppelin Admiral
******
United States United States


Brass addict, inventor, and scoundrel with a heart


WWW
« Reply #19 on: January 31, 2009, 04:45:14 am »

Good show!  That hodge-podge assortment of materials definitely gets the Jake Seal of Approval Grin
Logged
Pages: [1]   Go Up
  Print  
 
Jump to:  

Powered by MySQL Powered by PHP Powered by SMF 1.1.20 | SMF © 2013, Simple Machines Valid XHTML 1.0! Valid CSS!
Page created in 0.356 seconds with 17 queries.