Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[Larrchild]

Never mind the Servomotors, here come the Piezoresistive elements!

[Sahib]

compasspnt wrote on Thu, 05 October 2006 23:13

Hey Cemal. Say hi to Scotland.  I had hoped to be there starting next week for a couple of weeks, but sadly work here prevents.

Done Terry. Please let us know when you are coming over next.

Quote:

The motorised potentiometer appears that it could sound really smooth in it's initial gain reduction. How would you properly control the hold and release so that different types of sounds could be obtained?

The microcontroller compares the incoming signal against the threshold set by the threshold potentiometer. If they are both equal then it applies the release as set by the release potentiometer. If it is still above then it holds until it falls back to threshold level and then applies the release.

Quote:

For instance, if you wanted things to be very, very transparent and smooth, might not this system be able to recover so nicely that one would never know compression was being applied?  Sort of like the perfect human fader-riding robot?

Perfectly possible. Assume that we have a 1mS release time. And the potentiometer will sweep from A to B. We can control and manipulate that speed in various ways within 1mS time frame which may produce interesting audible results.

For example we can apply acceleration and descelleration to the sweep. In which case the potentiometer will gradually reach to its maximum speed and then gradually slow down and stop at the desired point. Or we can accelerate, hold constant and  descelerate.  

Quote:

And then say you wanted some real "pumping" compression sounds.  Do you think the motors would be capable of simulation of such, and if so, how might that be accomplished?

That would be easy. You would move the potentiometer at its full speed and start/stop it dead. This would definitely produce pumping effect. The amount would depend on the maximum speed that the potentiometer can move.

Quote:

So, might there be possible Presets that would perfectly emulate various types of known compression, as well as that of "none at all, yet perfect gain control?"

Theoretically this is possible as I explained above  but I am not sure of the result we may get. Trying to emulate an electronic process with a mechanical system is an unknown teritory. I have not been there before. It is completely against the way the technology is progressing. May be we should get down and try this thing. We have everything needed here. I will start with modifying an RC servo and see if I can get the response time below 1mS.

Cemal

[Sahib]

I have done some tests today.

I have modified an old RC servo and driven it with another one. I have comfortably managed to get the modified servo do a full sweep of a potentiometer in 1uS.

Without being too ambitious I can say that mechanically I could get the potentiometer respond to a transient of about 5uS comfortably using an off-the-shelf RC servo.

Electronically, however, this is subject to modifying the servo amplifier and driving it with a minimum pulse width of 1uS on time (and 1-2uS off) to see if it accepts it. At the moment with the current set up it only accepts a minimum of 0.5mS pulse width. If it does not then we'll drive it straight from the microcontroller which is much easier and better. What I am trying to do is to get an existing RC servo to respond to 1uS.

Cemal

[ssltech]

Sahib wrote on Sun, 08 October 2006 08:55

...a full sweep of a potentiometer in 1 MDM (maxdimario) wrote on Fri, 16 November 2007 21:36 I have the feeling that I have more experience in my little finger than you do in your whole body about audio electronics.. Sahib Sr. Member Offline Posts: 429 Re: Need info on different compression methods. Yes Keith, with no problem. I reversed the servo by completely removing the brushes/windings of the driver motor. It functioned as the output shaft. I drove it with another servo from its output. I applied 10uS to the driving servo and the output turned just under 10 turns. Main problem is that naturally the torque output is not sufficient because the servo I used only had 3Kg/cm torque rating. So the reverse torque is very low and would not drive a potentiometer. However, what I can do with it is to manipulate a resistive bend sensor which can be the attenuator. I did not try it with a heavier torque output servo because they are slower. Anyhow I have options. I can use a faster servo with a torque rating of around 3 kg/cm and replace the drive motor with a ultra high speed motor. I just have to look at what's available. I can also replace the gear train to increase the ratio. We worked with a good gear manufacturer in far east and the guy cuts anything you want at a very reasonable price. Unfortunately we are moving the studio into new premises within the next month or so and I do not think  we'll be able to do anything before new year. But let's keep the idea alive. Cemal ssltech Hero Member Offline Posts: 4780 Re: Need info on different compression methods. Ah, okay... -I thought that when you said you 'made a servo do a full sweep of a potentiometer' that you actually had it connected to a potentiometer. In which case it would naturally have already overcome the inertia/friction issues. In fact, I wonder what sort of power this sort of solution is likely to consume... Keith MDM (maxdimario) wrote on Fri, 16 November 2007 21:36 I have the feeling that I have more experience in my little finger than you do in your whole body about audio electronics.. Sahib Sr. Member Offline Posts: 429 Re: Need info on different compression methods. Yes Keith that's why I said in my previous post that I can get to 5uS response. It looks more realistic to me at this stage. The sweeper of the potentiometer is just brushes and its friction is totally negligable. Almost all of the friction  is due to the spindle working inside a metal bush.  However, that could easily be improved by replacing the bush with a ptfe plastic version which have almost zero friction. We can also go to the extereme of using micro metal ball bearing. I can calculate the frictions etc. and come up with a required torque rating but its all unnecessary. We have done these things hundreds of times before and, I am confident that with few modifications and improvements we can respond to 5uS. Cemal Sahib Sr. Member Offline Posts: 429 Re: Need info on different compression methods. Blunder. I made a huge error in my readings. The pulse that I was sending was not 10uS it was 100uS. My whole scientific career is ruined.  I feel like an idiot.   The controller sends out a maximum of 2mS in 8 bits. And I sent 13 steps which is 101uS. We have just made another test. Again and again. I got Mahcem check it over. This is definite.  When we sent 100uS the turn ratio varies. The reason is the increase torque ratio because we are driving the servo in reverse direction. So the driver servo is sometimes jamming.  Anyhow in average we can get the full sweep of a pot in 50uS. Sorry about this. I got very excited at the time too but I was wrong. However, this is still encouraging. Anyhow, I am still certain that responding to 5uS will not be difficult. This time I'll knock something but not with a rotary motion. We should have a micro lead screw somewhere that was left over from a job and plenty of DC motors. Coupling something together should not take much. To be continued. Print Pages: 1 2 3 [4]   Go Up « previous next » R/E/P Community > R/E/P > R/E/P Archives > Bruno Putzeys (Designer) - Dave Hecht (Master Tech) > Need info on different compression methods.   SMF 2.0.19 | SMF © 2021, Simple Machines | Terms and Policies XHTML RSS WAP2 Site Hosted By Ashdown Technologies, Inc. Page created in 0.06 seconds with 16 queries.

[Sahib]

Yes Keith, with no problem.

I reversed the servo by completely removing the brushes/windings of the driver motor. It functioned as the output shaft. I drove it with another servo from its output. I applied 10uS to the driving servo and the output turned just under 10 turns.

Main problem is that naturally the torque output is not sufficient because the servo I used only had 3Kg/cm torque rating. So the reverse torque is very low and would not drive a potentiometer. However, what I can do with it is to manipulate a resistive bend sensor which can be the attenuator.

I did not try it with a heavier torque output servo because they are slower. Anyhow I have options. I can use a faster servo with a torque rating of around 3 kg/cm and replace the drive motor with a ultra high speed motor. I just have to look at what's available. I can also replace the gear train to increase the ratio. We worked with a good gear manufacturer in far east and the guy cuts anything you want at a very reasonable price. Unfortunately we are moving the studio into new premises within the next month or so and I do not think  we'll be able to do anything before new year. But let's keep the idea alive.

Cemal

[ssltech]

Ah, okay... -I thought that when you said you 'made a servo do a full sweep of a potentiometer' that you actually had it connected to a potentiometer. In which case it would naturally have already overcome the inertia/friction issues. In fact, I wonder what sort of power this sort of solution is likely to consume...

Keith

[Sahib]

Yes Keith that's why I said in my previous post that I can get to 5uS response. It looks more realistic to me at this stage.

The sweeper of the potentiometer is just brushes and its friction is totally negligable. Almost all of the friction  is due to the spindle working inside a metal bush.  However, that could easily be improved by replacing the bush with a ptfe plastic version which have almost zero friction. We can also go to the extereme of using micro metal ball bearing.

I can calculate the frictions etc. and come up with a required torque rating but its all unnecessary. We have done these things hundreds of times before and, I am confident that with few modifications and improvements we can respond to 5uS.

Cemal

[Sahib]

Blunder.

I made a huge error in my readings. The pulse that I was sending was not 10uS it was 100uS. My whole scientific career is ruined.  I feel like an idiot.   The controller sends out a maximum of 2mS in 8 bits. And I sent 13 steps which is 101uS.

We have just made another test. Again and again. I got Mahcem check it over. This is definite.  When we sent 100uS the turn ratio varies. The reason is the increase torque ratio because we are driving the servo in reverse direction. So the driver servo is sometimes jamming.  Anyhow in average we can get the full sweep of a pot in 50uS. Sorry about this. I got very excited at the time too but I was wrong. However, this is still encouraging.

Anyhow, I am still certain that responding to 5uS will not be difficult. This time I'll knock something but not with a rotary motion. We should have a micro lead screw somewhere that was left over from a job and plenty of DC motors. Coupling something together should not take much.

To be continued.