Hard drive magnets for ribbon mic

I busted open some old SCSI hard drives and they had really big and VERY powerful magnets in them.

they were perfectly connected and spaced with the gap being about 1/4 inch apart and 2 inches long. I could see a ribbon fitting perfectly in there.

can a ribbon work here? is the polarity of the magnets right? they are attracted to themselfs through the gap...like if the spacers were not there they would crush anything between them including a little skin on my hand. :shock:

anyone?

i think scott dorsey posted about this on r.a.p. a while back. you might email him about it. i'm not sure if the orientation of the field is right for a ribbon mic. those magnets are very strong tho.

ed

All HD magnets I tried are not polarized right. If you put them vertically, then on the left side, the top is N, bottom is S, and on the right side, top is S, bottom is N. So the signal induced in the ribbon will be cancelled.
I never tried it though.

Hmmm maybe I'll flip one around. It would sure look funny but hey.

I cannot picture what do you mean.

[quote author="bluebird"]I busted open some old SCSI hard drives and they had really big and VERY powerful magnets in them.
[/quote]
Good magnets. I have used it for ribbon mic.
But some mods needed.
There are 4 the some magnets and two ARMCO steel plates.
There are two magnets glued on every ARMCO plate.
You can two magnets pick up from one ARMCO (be careful and strong !!!)
On the second ARMCO plate only make space between magnets (to
space of your ribbon) AND FIX IT (by piece of wood).
Then put on the second two magnets. Then you will have
half of HDD magnet unit, but wit four magnets and then stronger.
Because you want to have magnetic field perpendicular to
magnet faces, you must use pole pieces. I used quadrats made from
ARMCO steel 3 mm thick (with sides long as ribbon length).
With ribbon side precisely finished!!!
Then I put the quadrats on the magnets, and fix all it by epoxy.
Spaces between quadrats can be made by two braas screws,
it will be ribbon width and must be very precise.
It performs good, but rather for pressure than for velocity ribbon because
of asymetry.
Have a fun.
xvlk

Xvlk
Do you have a picture of your setup? Sounds interesting.

[quote author="smilinfu"] ... picture of ... setup ... Sounds interesting.[/quote]
yes, I can give picture of my old experimental setup.
It is only with magnet pair.
Measured induction 0.80 T here, but gap is not wide and ribbon
bad instalable.
Here is link
(you must copy and paste link in other browser)
http://mujweb.cz/www/xvlkxvlk/stara.jpg

Magnets are from old (1991) Sequent mainframe

Newer unit is under construction and measurement and I want not
post infos here till I finish it.

xvlk

hey that looks pretty interesting. I now understand where the feild of the magnets are...

I was unable to access the jpeg
Would be willing to repost the link or email it to me?
smilinfu"AT"earthlink.net

Smil,

(you must copy and paste link in other browser)
http://mujweb.cz/www/xvlkxvlk/stara.jpg

Click to expand...

It works..

Jakob E.

[quote author="xvlk"] Magnets are from old (1991) Sequent mainframe

xvlk[/quote]
You mean the old parallel processing 486 chip based Unix boxes...
I have one around here somewhere....
You can pick up those old SCSI drives out of those (first "commercial" generation) Unix boxes for nothing or next to nothing

[quote author="uk03878"]You mean the old parallel processing 486 chip based Unix boxes...
I have one around here somewhere....
You can pick up those old SCSI drives out of those (first "commercial" generation) Unix boxes for nothing or next to nothing[/quote]
My knowledges of all hardware are pretty poor,
This computer for me was good as material source.
There was 4 HDD units model 97560. cca 1 G each.Three of its are mine,
one was recycled :-( ... .
There are good magnets inside and it is pity to recycle.
But it is dificult to dissect it.
There was also many of 2 GB IBM HDDs, with small magnets,
Production year 1995, In disc array, 48 V powered,
{with DC/DC convertor near each disc, Powering was
similar as in todays mobille switching centres}
I want to use it in Starmaker ribbon DIY for a future.

xvlk
P.S.: Foto: {cut and paste to other browser}:
http://mujweb.cz/www/xvlkxvlk/hdd.jpg

Basically you want the old 5.25" wide, "full height" (all the new stuff is "half height") hard drives. I used to have a bunch of big, super-powerful magnets, about the size of my thumb. They were actually dangerous to play with, because if you lost your grip they'd clamp together and maybe pinch your finger.

[quote author="Scodiddly"]Basically you want the old 5.25" wide, "full height" (all the new stuff is "half height") hard drives. I used to have a bunch of big, super-powerful magnets, about the size of my thumb. They were actually dangerous to play with, because if you lost your grip they'd clamp together and maybe pinch your finger.[/quote]
It is "super full height". Magnetic disc alone have 13 cm in diameter.
In normal "full height" HDD units for PCs was ferite magnets. Not so good.
Todays Neodymium caviars are better.
xvlk

> some old SCSI hard drives and they had really big and VERY powerful magnets

FWIW: modern drives have magnets just as powerful and much smaller. You will probably get better results with your dead last-year home-PC drive guts than with ancient drives.

> All HD magnets I tried are not polarized right.

Correct. See picture.  EDIT 12 years later-- re-attached, below.

The coil in a hard drive is a "full turn" (really many full turns). The most efficient magnet has an N and a S pole on each face, to soak both sides of the coil with flux in the right direction.

A standard ribbon mike is a "half-turn" device. It does not loop around, current is all one direction (for the simplified DC case, or a freeze-frame view of AC/audio).

There is a trick. Use two ribbons wired as shown on right. Although I show them nearly touching in the middle, actually the magnet is very weak in the middle. So don't waste ribbon there. However that means an acoustic diaphragm that is more 2-point than 1-point. Since it is already a rather large system (with typical hard drive magnets), the directionality will go crazy in the top-octave.

You could use one ribbon in just one third of the magnet: that sketches fine.

The other thing is: these magnets are about an inch front to back. In a classic bi-di ribbon, this forces one or both sides of the ribbon to be recessed deep inside a solid structure. Hold a toilet paper tube to your ear: this causes an audible coloration of the mid-range. With a 1" or 1/2" box-shape duct between the world and the ribbon, you get the same effect but shifted to the top of the audio band. While sometimes a lucky combination of dimensions gives a boost that corrects some other dip, usually recesses and ducts just mess-up the sound. And if you want messed-up sound, there are plenty of cheap/nasty mikes on the market, or even in your friends' junk boxes.

BTW: in the PC world, "full height" is 3.5 inches or about 9cm. (This size pre-dates the IBM PC.) Half-high is half that. Most current desktop hard drives fit a half-high bay, but are actually less than 1.75"/4.5cm tall, being just tall enough to hold the works. The old "5.25" width (size of some cocktail napkin, the day Shugart was complaining about 8" floppies being too big) was really about 5.5" wide on the drive; the "3.5" width used by 1.44M floppies and most desktop hard drives is about 4" wide.

Cool trick, PRR :thumb:

I'd think practically it would be easier to make one ribbon and hook a contact in the middle, as it is not much fun to machine two more clamps, and then install two ribbons.

It might be actually interesting for an omni ribbon, where the ribbon should not be recessed too much and there is not much of cavity. Tuning the back-chamber is a bitch, though. I have never been able to find a perfect tonal balance, and I spent maaaany hours trying different materials, chamber size, spacing, etc. But you can actually find some cool sound effects here, like excellent top and no bass, or excellent bass and no top. Very good mids and no either bass or top would be cool for old broadcasting effect.

On the other hand, I just don't understand why to go for all these troubles and complications, when nice and "right" Neodimiums can be had for a few shekels from ebay.

[quote author="Marik"]
It might be actually interesting for an omni ribbon, where the ribbon should not be recessed too much and there is not much of cavity. Tuning the back-chamber is a bitch, though. [/quote]
Yes, it will be my starmaker DIY
Why to tune back cavity. That hell was done in Schottki - Gerlach
Siemens mics? Why not to use damped tube as proposed by Olson?
xvlk

> interesting for an omni ribbon... excellent top and no bass

> use damped tube as proposed by Olson?

Right. The theory works best with a pure resistance on the back. A physical form is an infinite tube the same area as the ribbon. A practical form is a long tube with cotton or wool fuzz to kill the reflection at the far end and in the bends that you probably want to make the pipe fit a case.

Simple theory suggests a tube that is many wavelengths long at all audio frequencies, or dozens of feet. But the ribbon is small, so the pipe has small diameter. There is significant wall friction, which damps the pipe. As a first guess, if the pipe length/diameter ratio is much-more than 30:1, maybe 100:1, it will hardly ring at all. A typical pipe for a ribbon might be 0.3" diameter, so a 30" (1-meter) pipe might work.

That is still rather big, so we fold it 5 or 6 times to fit in a case. The bends will cause reflections: make them as big as possible. Even so, you probably need to add thin wool fuzz at many places along the length. That is hard to do in a long bent pipe: Olson used separate straight sections and bends so he could get the fuzz in.

what about two steel pieces standing up at 90 degrees from the flat side of the magnet ?