Rev 2 of my DIY Stereo Cutterhead

[opcode66]

I think we should start a thread or a whole new section!

I am also tinkering heavily with my MakerBot Replicator 2. I've already made a bunch of useful stuff, and some not so useful but fun. But, I think there are enough of us here to comprise a collective effort to make an open source Trolls cutterhead.

I could see several variants. Mono. Stereo non feedback mostly plastic. Stereo non feedback metal. I've been working on winding coils a lot. We could collaborate on a properly driven design instead of leveraging speakers. Feedback down the road.

Anyone interested? I figure, why should a bunch of us try to tackle this individually?

[emidisc]

Opcode66
Your idea sounds good to me and a positive way forward
has I have mentioned earlier I can offer some help with "Inventor 3D Modeling"
I also have done a little coil winding and attempted a couple of DIY cutter heads,
The only problem I see without trying to be negative is that I guess some people will just want to take without contributing to the concept? Either way I am interested in being involved......

Emidisc

[opcode66]

Would you be willing to work in Sketchup Make? It is free. That way we would have a free dev platform that is consistent.

[emidisc]

I will take a look at it, do you know what the the output file type is?
Chances are Inventor can export to the file type I will take a look.....

[opcode66]

I can import DWG files.

[chaosbc]

sketchup can export 3d models in dae or stl.
stl export through sketchup is not very good but with the basic version of netfabb it is really consistant.

[45rpmdude]

what kind of speakers did you use. I take it you didnt use a coil.. Im trying to power a parkard bell fone-o- chord. Im getting good vibration from my custom coils. but im winding them tightly. Ive tried winding them loose. I wind my speakers with a peice of crimped tin.. then i add a magnet to it and i get sound.

Im just having difficulty mounting the sharpened nail.. evrything else works.. but the head is dead.. i perfer not using crystal elements.. i would like mine magnetic.

[tragwag]

I'm quite interested in helping any way I can with a "trolls cutterhead" of any sort.
I have some spare 1C pieces I might be able to 3D scan if we need.
let me know what I may be able to do to help!

[opcode66]

I'm pretty sure a caliper is all that is required.

3d scanning makes a huge mess of a file. Unless you are dealing with a surface that would be impossible to draw like a face, it is best to draw via measurements from a very accurate caliper.

I now have Autocad 2015. So, I can work with DWG.

[Ciuens]

Hello Mark, could you explain the cutting process with white noise and the analysis and correction with equalization?
Thank you

[Ciuens]

Here's my 3d printed cutter.
The electronics is controlled by an Arduino and a small stepper motor with reduction gears, the control software is being created in Labview.

[Ciuens]

Follows a test cut and the reference audio. The cutoff frequencies of the test are very different from the original reference.

Cut test
http://www.3dzez.com.br/teste%20cut%2001.wav

Reference audio
http://www.3dzez.com.br/reference%20cut%2001.wav

[markrob]

Hello Mark, could you explain the cutting process with white noise and the analysis and correction with equalization?
Thank you

Hi,

There are a number of ways to do this. The approach I used is as follows

1. Cut IRIAA pre-equalized white (or pink) noise to a disk. Cut 30 seconds to 1 minute to make sure you have large sample. Don't over do the level, but make sure you have high frequency level enough to overcome the noise floor of the medium.
2. Playback the disk and capture to the cut to a computer file format of your choice (I use WAV)
3. Using the un-equalized noise file as a reference, load this in to a program like Har-Bal or some other EQ matching/ spectrum analysis software.
4. Use the softwere to develop an EQ filter such that the reference spectrum and the captured playback spectrum match within the desired bandwidth. Save this filer for re-use.
5. Use the derived EQ filter to process any program material in addition to the IRIAA EQ prior to cutting.

Note that is it not advisable to push the match at the high frequency extremes as this will likely force you to apply too much power to the head. As you gain experience with the head, you can push it more. I use an older version of Har-Bal as it is simpler than the current version, but it does have some major limitations. The amount of allowed cut or boost is limited in the older version and there may not be enough boost to match the spectrum. There are now some other software EQ's similar to Har-Bal that may be better.

Hope that makes sense.

Mark

[EpicenterBryan]

The electronics is controlled by an Arduino and a small stepper motor with reduction gears, the control software is being created in Labview.

What is LabView being used for?

[Ciuens]

Hi Bryan

Labview is used to control commands to the Arduino to control the stepper motor, forward, backward, stop and go, and at the correct speed for the spacing between the grooves.

[EpicenterBryan]

I made some response measurements on the bench using a Kaman Eddy current position sensor I got cheaply from eBay.

Mark, I'm very interested in your eddy current measurement.
First, can you tell me where the measurement was made? At the stylus or the holder or other?
Did you buy a sensor with modulator / demodulator ... something like:
https://www.hofstragroup.com/product/kaman-kd2310-eddy-current-displacement-measuring-sensor/
This one says "Frequency Range: Static to 50 kHz", and Output Full Range: 1.2-3.0 VDC.
I have not used these before, so do you just decouple this with a cap and get AC out over that range assuming centered in the range of motion?

Here is an interesting question...
Is it possible to use an eddy current sensor in an application where there is also a large static magnetic force?
Pretend there is a non magnetic disc on top of a fixed magnet. Can you still measure distance to the non magnetic surface by adjusting some offset? The question comes from a possible application to implement feedback when using a surface transducer as a driver for a head. Below is a video that shows the guts near the end. The idea is if an eddy current sensor could still work in an environment with a large static magnetic field, it could be used as a non contact feedback pickup for this transducer.

After playing with a few demonstrations related to magnetic dampening and using eddy currents / magnets, I'm starting to see a few things in a new light and see applications. My gut says these drivers use the non magnetic front spring / mounting combination to generate eddy currents in opposition for damping. It will be interesting to see how they work open loop... I have two and will start messing with them next week.

Bryan

[markrob]

Hi,

The Kaman sensor I used is exactly the same one you found on ebay. Its even for sale by the same seller. They must have had quite a few of these from an old application. Kaman still makes these, but this model is no longer (might be 10 years old).

http://www.kamansensors.com/
http://www.kamansensors.com/html_pages/Kaman_Sensors_frequently_asked_questions.html

There is an interesting video on the site that shows the sensors' ability to resolve 1 nanometer.

Its a great buy. I measured at my stylus holder, since it is already made of aluminum. The Fairchild head used a similar measurement method to obtain feedback.

The surface transducer looks like a nice candidate for a driver. I don't believe they rely on eddy currents to damp the transducer. Its designed to be mechanically coupled to some sort of radiating surface. I suspect that provides any damping. Does SparkFun provide any specs on the driver?

[Bahndahn]

Great to hear updates!

Does SparkFun provide any specs on the driver?

Limited specifications are provided, I recall reading somewhere that 'more specs are said to be coming'.

Here is what is currently available: http://dlnmh9ip6v2uc.cloudfront.net/datasheets/Components/General/GD-003.pdf

[chaosbc]

I am no way affiliated to aliexpress but : http://www.aliexpress.com/item/Large-Surface-Transducer-with-Wires-4-Ohm-5-Watt-4K-5W-Free-Tracking-Number/2020210299.html

[EpicenterBryan]

I am no way affiliated to aliexpress but : http://www.aliexpress.com/item/Large-Surface-Transducer-with-Wires-4-Ohm-5-Watt-4K-5W-Free-Tracking-Number/2020210299.html

That's a great price. I think I paid $24 each for the ones I'm about to play with.

Bryan