Really big VU meter

GroupDIY Audio Forum

Help Support GroupDIY Audio Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.

mobyd

Well-known member
Joined
Jan 27, 2006
Messages
203
Location
Auckland NZ
I gotta build a really big VU meter. It's for a clapometer style application. I've built lots of them before with columns of LEDs, striplights etc, but the issue here is that it has to look like a classic mechanical tachometer. 180 degree swing with a needle about a metre long. Preferably with the same sort of ballistics as a regular d'Arsonval meter. The electronics are trivial but the mechanicals are not. The only thing I have been able to come up with is a big grunty stepper and a pot stuck on the back to keep track of where it is, and a micro (PIC or something similar) to sort it all out. I cant help thinking there's gotta be a better way (preferably not involving a motor, a magnet and an aluminium cooking pot i.e. monster speedo). Any ideas ?
Cheers
M
 
mobyd said:
The electronics are trivial but the mechanicals are not.
Do it optical? Think overheadprojector or a ccd camera to beamer/TV-panel, depending on surrounding brightness might do the trick.
 
How about a camera and monitor showing the blown up image of a small meter?

oops... Harpo beat me to it...

Otherwise PIC driving a stepper motor seems reasonable technology but perhaps too much development work for a one-off.

JR

 
To start with a stepper is not the one you would use. By the way if you are going to control it with a microcontroller you do not have to servo it with a potentiometer (close loop) as you simply count the number of pulses you send to it (open loop). You would only use the potentiometer feedback if there is a danger of stepper being jammed.

A DC motor is more like what you are looking for. You indicated that the electronics is trivial, therefore I won't go in there but if you are stuck I'll throw in some ideas.

However, directly coupling the pointer/needle onto the motor's shaft will produce a  very violent movement, and highly likely that it will oscillate . To produce a smooth movement you will need to damp the motor shaft. This could be achieved by coupling dampers like the ones used on top loading CD  mechanism. Except you will need a larger one. I have got  few of those that I removed from a couple of photographic printers, that a friend of mine whom had a photo lab scrapped. And the DC motors I got from them are also top quality. So this gives you some idea of sourcing cheap components. Check out e-bay.

The other alternative method for damping is to use a gear train. A standard R/C servomotor with a torque output of 3-5Kg/cm will be good. Remove the electronics and couple the output shaft to the shaft of your vu meter. Of course you are not applying any voltage to the motor. Now you will be driving it in reverse and this will nicely damp the movement.

Obviously if I was doing this job I would properly calculate the forces involved and the torque output required. Trial and error is other option.

I have just read the other posts.

R/C servo is a good choice but to drive a pointer/needle 1m long will require a lot of torque. The highest torque output servo that is currently available gives 25Kg/cm which is Hitec HS85BB or equivalent brand and that may struggle. At some point Supertec made 35Kg/cm but these were discontinued because they were not reliable. Also required hell of a current.  Also higher the torque output, the slower the servo. Hitec HS85BB I think moves to 60 degrees at 0.25 sec. So to 180 degrees you got almost a second. This does not account for the additional torque required for the friction that the damper creates.

Tonegawa-Seiko has industrial versions of these R/C servos (which are widely used in military drones etc.) which provides 100Kg/cm but prepare for a bank loan. Most importantly not everybody sells them.

 
a really big VU meter. ... has to look like a classic mechanical tachometer. 180 degree swing with a needle about a metre long.

"VU meter" is usually 90 degrees. Tachs are often 270 degrees. The VU movement is really up-to-180 deg, using the most linear center of the range (it gets very lame at the 0/180 extremes). National once offered a chip to drive a 2-coil rotor for 270 deg needles, with obvious car-use, but I think that died long ago.

Just thinking too loud....

Mass and mag-flux/force scale with size.

It *may* be valid to scale-up and see how that looks, what parts may be available.

I can't find a VU meter in this mess. IIRC, a cheap (non-Standard) meter has a 1.5" radius pointer and a 0.5" diameter coil around a medium-strong magnet.

So a 40" pointer should have a magnetic motor which is all copper rotor, medium-strong magnet, 13 inches (0.3m) diameter.

Say a sauce-pan covered with HDD magnets, and a coil wound around a Frisbee. What you proposed and do not like. Gearing trades size for speed and torque. You could make the actual motor more compact, but probably not a lot less material (copper and magnet).

Hard drives have a head-arm swinger, used 2GB drives are trash-price, and a DC-coupled audio amplifier will push them fast. ~~10mS. They can move faster than the drive-specs say: they limit maximum seek speed because they have to read track IDs on the fly to know when to slow. Also to stay within cheap power supply limits, and to handle heavy-duty long-time data-access chores. You could probably smack 18V across the coil and get 5mS (naked) seek, if you let it cool between rounds of clapping. However the max swing is under 90 degrees. And the actual seek will be far slower with a meter of tin on the arm.

> same sort of ballistics as a regular d'Arsonval meter

They come many speeds. What do you need? A car tach does not have to be much faster than the engine, or the driver's eye. A loaded engine can only rise RPM so fast. 100% in 2 seconds? The meter should be faster, but not so fast that individual sparks flick the arm. A VU can be faster, but we know that if we are not trying to peak-catch, we often like a slower integration. And audience clap-build is not speedy compared to electronic time-scales.

In any case, at this size and speed, torque is essential. Not really Tom Swift's Electric Catapult torque, but at/beyond the upper edge of R/C and mini-motor designs, even with gearing.

Get a small RC car. Hack the motor. Wind a string around the wheel axle shaft, tie the ends to your needle. These small motors have rapid acceleration and high peak torque. The gears and wrap-string can lever 600 turns into 0.6 turns, a meter motion. String-slip can limit needle-bending. You do need a pot on the needle to indicate absolute angle. You might want hard/simple cut-off switches to break the power if the needle swings past 100% or 0%.
 
Good ideas.

One thing as important as the motor. The needle. 1m is a hell of a length and it will flex and oscillate even up to 5mm diameter. Forget metal, it will have to be glass or carbon fiber and shaped flat rather than round.

Now, try to swing that with your hand and you'll feel the amount of force it creates. You have to drive it in  one direction, stop it at a point and drive it on the opposite direction. Think about the force you require. Starting is not a problem but stopping is, without the descelleration. 

Not a trivial task.
 
In olden times a mirror galvanometer would be used for tasks like this.  I remember a surplus shop that had a drop dead 48 inch (or more) meter mounted on the wall as decoration.  Google came up with pictures of about 100 mirror galvanometers, but none as good as the one in that shop.
 
OK this is a little Rube Goldberg, but what about making a crude projector out of an open back normal sized meter, stuck in front of a projection lamp. Some kind of a lens system and screen or light box, would be required for sharp focus after it is blown up...

JR
 
Well it finally happened. Somewhere along the way the needle length got upped from 900 to 1500mm
throwing out all my calculations by a mile, so it ended up a suckitandsee approach. Used a big stepper
and a PIC. Ballistics were (and still are) the biggest bugbear. First cut of the code when I shouted at
it, it threw itself (15Kg) off the stand and thrashed around on the floor - completely destroying the needle.
With acceleration and decelleration tables it is a lot tamer now and good enough for the job at hand
(a billboard). Obviously there is work that could be done, especially at the bottom end of the decay
but hey, the sun is shining.
http://www.youtube.com/watch?v=Brmx_owJpeE
M
 
Fantastic project!

...and if you sat in that chair, you'd have no problem knowing when you'd hit +3...

you'd feel the draught!

Great stuff,

8)

Mark
 
I realy like that!!!

In the 70's, an engineer I knew wanted a VU on the wall, right besides the rythmo strip in a film post studio.
He took a standard VU with a transparent scale and simply pit it on a retro-projector ans was done with it.
But I like this one wayyyyy better!!!  ;D

A stereo pair would be something in any control room!
Plus the placement would have to be judicious to avoid knocking off anyone!!

Love it!  :)
 
mobyd said:
Well it finally happened. Somewhere along the way the needle length got upped from 900 to 1500mm ...

First cut of the code when I shouted at it, it threw itself (15Kg) off the stand and thrashed around on the floor - completely destroying the needle.

http://www.youtube.com/watch?v=Brmx_owJpeE


Wow, Simon's really slashed the budget this time hasn't he - is the next series of X-Factor going to look like Opportunity Knocks then?

Seriously, I love it to bits. This gets my Harry Hill award for 'Mental Project Of The Year'. Awesome!  ;D

 
sahib said:
Good stuff. Did you do the billboard too?

Nope, just the electricals - the billboard reminds me of the extreme fun I used to have
in Minis in decades past. Fubulous little cars, esp the Cooper 1275s.
M
 
It looks way way cool. Though I wouldn't like to be trying to sleep anywhere near that sign. :D

The mic: I say it's suspended to a lamp, right?
 
Back
Top