ribbon mic:xvlk solution/DIY hall probe /plz elaborate

hello all
xvlk has offered a wonderful solution for a DIY hall probe for measuring magnetic flux ..

questions:
what is ... smd - transistor?

what gets inserted into the magnetic gap? this smd device OR some kind of wire probe ??

known magnetic flux of FDD for comparison/calibration?

betamax device??????

here is his explaination:



Hall probe is on every older Floppy disc drive.
it is some SMD-transistor looking device under magnet glued onto
inertor.
it have four legs - one grounded , one connected via resistor to +5 V
and two goes to the big IC.
that two connect to the multimeter and
measure mV.

the GND and + connect via resistor of several kOhms to the 4,5 or 9 V battery.
Then you must calibrate it. I use magend gap from HDD which I
calibrate at school metrology lab.

I never use FDD hall probe. I use Hall probes from old Betamax recorder.

xvlk

Click to expand...

can we colaborate and elaborate on this design ... so that we can all have this info for the future ..?

i had often wondered if a DIY hall probe was possible .. now ..
thanks to xvlk we have a solution ...
PRR don't you have a hard drive guru for assistance on this subject?
i thought i remembered you post something about your friend that is HD guru..

thanks again xvlk... :thumb:

looking forward to everyone's input on this so we can have a complete design with options pertaining to readily available materials.

thanks
ts

dude, you really need to cut back on the coffee... and not start a new thread everytime a thought pops into your head. your thread titles look like some kind of secret code... where's my decoder ring? :razz:

SMD == Surface Mount Device

That's about all I can tell you.

just google


http://www.allegromicro.com/hall/

this might help you get started

What xvlk is talking about is not an SMD transistor, it just looks like one, It is the hall effect sensor in the drive. Or I assume that is what he is talking about. With how readily available hall effect sensors are and easy to get I am not sure why he did not just quote a part number or what values to look for in a hall effect sensor. Maybe the ones in diskdrives is just the right range for this purpose. I do enjoy scanveging for parts, but I do not even know where to get a diskdrive other then buying on. Have not had one in atleast 6 years.

adam

gus thanks ... nice reference :thumb:

adam

What xvlk is talking about is not an SMD transistor, it just looks like one, It is the hall effect sensor in the drive. Or I assume that is what he is talking about.

Click to expand...

YES exactly ..

I am not sure why he did not just quote a part number or what values to look for in a hall effect sensor. Maybe the ones in diskdrives is just the right range for this purpose.

Click to expand...

the floppy drives must be perfect for this application and
even a beginning of a path to a new option is better than no option at all
..thanks again xvlk .. :grin:
part numbers would be nice too ... maybe we can look through this link that gus has offered .. i looked at it quickly but it might be beyond my current knowledge .. :sad:

also consul... i half assumed that smd was surface mount something ... but thanks for clearing that up

solder ... no coffee ... not a new idea / a new solution to an old idea ...
first word in the title is ribbon mic: ...
thenk theres .. hall probe ...
i try my best to make the description applicable to the content ... it makes sense to me .. sorry
i give away free decoder rings .. just red through my "ribbon mic" posts and you will have it .. :grin: lol
nice to have your kind comments and interest.. :wink:
anyway
i still have these questions :

what gets inserted into the magnetic gap? this smd device OR some kind of wire probe ??

known magnetic flux of FDD (floppy disk drive) for comparison/calibration?

thanks
ts

The hall effect sensor would go in the gap, Just run a piece of copper wire from house wireing down with the leads from teh sensor and shrink wrap it all together. Since this is a very much uncalibrated hall sensor I am not sure if pulling apart a floppy is worth it. Unless you have a floppy drive jsut sitting about that has no use. then go for, otherwise I am pretty sure I have seen Hall Effect sensors at Radio shack in the past, and of corse any decent electronics supply house will have loads to choose from.

adam

:grin: :shock: !!MASSIVELY COOL REFERENCE!! :shock: :grin:

http://www.lakeshore.com/mag/hs/hsm.html

later
ts

OK :shock: i just found out why xvlk has recomended the hall generators from FDD's ... just look at the prices for these puppies .. :shock:

here:

http://www.lakeshore.com/mag/hs/hsoi.html

$100-$400 :shock: just for the generrator without regulated PS or cucuitry (or a voltmeter)

so... lets do some exploritory surgery on some floppy drives ...
:grin:

lots of great info at this www ...above ... enjoy

later
ts

grabbed this from Wikipedia ) free online encyclopedia

Frequently, a Hall sensor is combined with circuitry that allows the device to act in a digital (on/off) mode, and may be called a switch in this configuration. Commonly seen in industrial applications they are also used in comsumer equipment; for example some computer printers use them to detect missing paper and open covers. When high reliability is required, they are used in keyboards.

Click to expand...

wow these sensors are everywhere .. hehe .. were surrounded ..lol

later
ts

Rather theoretical: There is method to calibrate Hall probe
at home.

source of homogenous magnetic field is loop (made from thick copper
wire, like used for welding transformers).

If some part of loop is narrow, it is two wires in paralel,
and between them is homogenous field.
H = 2I /(Pi*D)
B = mu * H = 4*Pi*1E-7 * H
where I is current trough and D is distance between.

If that loop is made from copper, you can compute resistance.
You can measure AC (mili) voltage on it - know current.
And you can measure AC voltage on ( still DC baterry powered Hall probe)


And you can obtain high current by transformer soldering iron.
Problem is in mechanical stability and cooling. But possible to made in the garage.
and ... Absolut.
Second method to obtain high DC current is to use starter as dynamo...

xvlk

[quote author="ToobieSnack"]

...the floppy drives must be perfect for this application...
[/quote]

No. As I wrote in another thread, they are way too sensitive. Their linear range is probably up to 200--500Gauss, at the most.
I made a few probes from HDD, some samples, and AD element. The latest was the least sensitive, with linear range up to 0.2T, IIRC. It still gave a serious error in the 0.5T range.
It is quite hard to find the Hall element with sensitivity needed, so after spending some time playing with that, I just bought a commercial calibrated gaussmeter.
There is one local company in SLC, which sells 3T probes separately for $80. That would be the cheapest and easiest solution I found so far, but I'd just buy their Gaussmeter, calibrated and ready to go. They do a lot of custom job and you might try to give them a call and ask if they have one second hand, or some model which is cheaper than on their website:

http://www.trifield.com/gauss_meter.htm

No. As I wrote in another thread, they are way too sensitive. Their linear range is probably up to 200--500Gauss

Click to expand...

Marik, thanx a lot. Mean, that it is not pure Hall probe, but something
with the amp.
But there are current setting resistor on the FDD drive, why?
(On the my FDD drive, different drive can have diferent probe)
If it is complex system with the amp (like in computer magnetic
keyboards) why +5 V is connected directly ???????
Normal hall probes are linear up to 3 T.
But, Hall probe is multiplier. Then sensitivity is
much affected by current trough it.
What was current (or Voltage/limiting resistor) when you had measured it ?????

xvlk

There is kind of sensitive Hall probe,
which have butterfly layout; but order of sensitive ????
Normal Hall probes are linear to 0.6 T, can be linearized to 1T
by small resistor (10-100 Ohms) connected paralel to output terminals.
But it is monotonous (and very lineary) to 5 T.
xvlk

[quote author="xvlk"]

No. As I wrote in another thread, they are way too sensitive. Their linear range is probably up to 200--500Gauss

Click to expand...

Marik, thanx a lot. Mean, that it is not pure Hall probe, but something
with the amp.
But there are current setting resistor on the FDD drive, why?
(On the my FDD drive, different drive can have diferent probe)
If it is complex system with the amp (like in computer magnetic
keyboards) why +5 V is connected directly ???????
Normal hall probes are linear up to 3 T.
But, Hall probe is multiplier. Then sensitivity is
much affected by current trough it.
What was current (or Voltage/limiting resistor) when you had measured it ?????

xvlk[/quote]

It's been awhile since I played with the Halls, so to be honest I don't remember all details and will need to read about it again. IIRC, the Hall sensor depends on its voltage rating. So if you apply a voltage to +V of the sensor, the output voltage goes up as the magnetic flux increases. There is a certain amount of increase per Gauss (basically this is a sensitivity) and cannot exceed maximum feeding voltage rating of the particular device.
I have somewhere left an AD22151 sensor. It has 8 pins, and with external circuit can be set for different sensitivity:

http://www.analog.com/UploadedFiles/Data_Sheets/5074535122786366627AD22151_a.pdf

I remember having hard time to source right element.

[quote author="Marik"]
It still gave a serious error in the 0.5T range.
[/quote]

What is serious error.
It may be 20 - 30 % error. It is several dB on gain of mic.
But I tell all the time, that you must somewhere callibrate
hall probe.
If probe have 60 % error at 1.5 T, it not signs, that you can
not measure with it.
You must only callibrate it within all 1.5 T range
and then measure trough that callibration curve.

And because you measure voltage (or millivoltage) onto Hall probe,
you must still have curve to translate it to the Teslas.

And if this curve is not linear . For 10 of 20 % measurement it can
be rather nonlinear.

And you still must callibrate altough it is perfectly linear, and
numbers of callibration points for 10 % measurement is not so astronomical.

And, by the way. Know someone some about general dependance of Hall
voltage to the magnetic field. I somewhere seen, that it is arcustangent ,
PRR, look to Shea, Shockley, or what we have at the US help me.

xvlk

[quote author="xvlk"]

But I tell all the time, that you must somewhere callibrate
hall probe.
If probe have 60 % error at 1.5 T, it not signs, that you can
not measure with it.
You must only callibrate it within all 1.5 T range
and then measure trough that callibration curve.
[/quote]

The problem is if you calibrate it in 1.5T range and the element is linear lets say till 500G, then in 0.2T range it's gonna be about the same 60% (maybe litlle less) unlinear, compare to 1.5T.
I have a digital Gaussmeter, which is calibrated within 1% in 0.2T-2T range and within 5% in 200G--0.2T. I have some DIY probes, I made out of HDD and sample elements. Of course, they'd give you some idea about flux density. Having Hall element curves you could calculate and compensate errors. But compare to a real thing I found they are basically unusable for work in a wide flux range.

Listen to Marik. He is smart. All Russian tech people seem to be genius.
Hall sensors are a pain in the ass.

First, they are not that sensitive, so you have to amplify the heck out of them, which means noise amplification.

Second, they are tempeture sensitive. If you buy a commercial gaussmeter, this has been taken care of for you.

Third, they are directionally sensitive. They must be oriented in the field the exact same way, other wise your readings will vary. Even the earth's mag field can have a minor effect. They make compasses out of hall sensors, ya know.

Fourth, they are hard to source. F.W Bell has them, but hard to sample.
Only a few people make them.

Fifth, your bias current needs to be highly regulated, if your power supply drifts, your sensor readings drift.

If you still want to mess with them, drop me a pm.

cj

[quote author="CJ"]Listen to Marik. He is smart.
[/quote]
I know it.
[quote author="CJ"]
Hall sensors are a pain in the ass.
[/quote]
Disagree, it is simple device, like transistor or other.
It is widely used from keyboard to silicon measurements whether p or
n type.

Yes, it is difficult to made linear device with laboratory parameters,
but for mic design +- 0.1 T is O.K.

I am also fan of simple measurements and then I personally
use not one calibrated Hall probe with seal from metrological institute Sevres, Paris; but have callibrated source (magnet with gap) of 0.8 T
induction.
And then for each measurement I can recalibrate probe (to temperature, current , etc) and measure. Because all my mic have 0.6 - 0.7 T in the gap, this is enough for me.

It is good to have set of callibrated gaps (0.5 , 0.8, 1 , 1.2 ) and
before measurement check it.

I think, that this gap is more stable under sun, than every
sealing - wax.

[quote author="CJ"]
If you buy a commercial gaussmeter, this has been taken care of for you.
[/quote]

What ?? If I buy Ohmmeter, I got set of normal resistors,
to make callibration.
If I buy Gaussmeter, I must also check it, not only believe...

[quote author="CJ"]
Third, they are directionally sensitive. They must be oriented in the field the exact same way, other wise your readings will vary. Even the earth's mag field can have a minor effect. They make compasses out of hall sensors
[/quote]
This is property of magnetic field, not Hall probes.

[quote author="CJ"]
Fourth, they are hard to source.
[/quote]
Not think so.
[quote author="CJ"]
Only a few people make them.
[/quote]
Yes, and one of them lived three houses near me, and is Dead. (Not yoke)

[quote author="CJ"]
If you still want to mess with them, drop me a pm.
[/quote]
I not understand american phrasal - word to mess with,
maybe I had not read A. Ellis (that man who invented your language) carefully, but I mean, that Them is Marik.

No, it is not disputation, only discusion, or I intended to be.

xvlk

OK, go for it.
But cheaper and less time and hassle to buy meter.
I deal with hall effect sensors everyday, and they are fussy.
You need a diff input for rejection.

http://www.sypris.com/stm/content.asp?page_id=764

http://www.sypris.com/library/documents/hallcatalog.pdf

The FW Bell GH-810/820 was their best one but they pulled the plug on it. :twisted:



So we found a replacement in China that had better tempeture and cost 75 cents instead of $4.95. Look out for China.

Hall chips come in two configs, three lead and four lead, so make sure you get the right one.