Pickup Tester, not a driven coil

[SteikBacon]

I builded a testguitar where I can test any pickupcombination possible with 12 coils, with any volumpot (25, 250, 500, 1m), 12 different tonecap, different resistance in tone, 12 different cap in highpassfilter, resistance in series and parallell, and pickups can be mounted anywhere between neck and bridge, and have access to modify anything easily if needed. A very usefull tool.

Any test with strings and try to have them vibrate without picking the strings, like e-bow, have not told me anything about different pickups. We have done some measurment, but I donīt even think about doing it more because it really doesnīt work. And I donīt believe setup with computer either. For now we have needed to pick the string to get the high freq to be able to estimate anything. So I donīt believe in a setup with regular stringvibration...

[David Schwab]

workbench steel?

I've always wanted to make a large framed string instrument. Like a piano, but with a fingerboard... but big and heavy. You have to stand next to it to play it.

I bet it would sound interesting!

[DrStrangelove]

Tranducer I/O plots?

I've done a bit of searching on the forum here and on the net, and this one never came up. There's a lot of topics on using drive coils for testing pickups, but then you have the problem of the drive coil's inductance and the actual drive level, coil-to-coil magnetic coupling or lack thereof.

Every transducer can be characterized as an input-output device
having an energy transfer curve.

Microphones and loudspeakers have sound level vs. voltage;
strain gauges have stress vs. resistance plots.

Guitar pickups stand out because they are rarely so described
except in subjective terms.

I propose the characterizing of guitar pickups as input-output devices.

Charting magnetic field frequency at constant amplitude against
output voltage seems like a no-brainer to me.

Measure the delta-gauss at the drive coil with a hall effect sensor
and adjust for an arbitrary field strength (say, +/-20 gauss) at a given
frequency. That way, the mutual inductance between drive coil and
pickup-under-test is allowed for in a highly reproducible method.

It doesn't need to be perfect -- it only needs to work adequately.

You could also test dynamic range and linearity at an input frequency
by varying the magnetic amplitude. This test might distinguish different
magnet alloys and perhaps whether or not they are completely charged.

Comments welcome.

-drh

[R.G.]

I builded a testguitar where I can test any pickupcombination possible with 12 coils, with any volumpot (25, 250, 500, 1m), 12 different tonecap, different resistance in tone, 12 different cap in highpassfilter, resistance in series and parallell, and pickups can be mounted anywhere between neck and bridge, and have access to modify anything easily if needed. A very usefull tool.

Good for you.

So you have a setup much like I described, but with a lot of other things you can swap onto and off it.

Any test with strings and try to have them vibrate without picking the strings, like e-bow, have not told me anything about different pickups. We have done some measurment, but I donīt even think about doing it more because it really doesnīt work. And I donīt believe setup with computer either. For now we have needed to pick the string to get the high freq to be able to estimate anything. So I donīt believe in a setup with regular stringvibration...

Belief is a strange thing - it can both open your eyes and blind them.

So how do you provide a repeatable picking? Is it like my calibrated picking accessory? Or something different?

[R.G.]

I've always wanted to make a large framed string instrument. Like a piano, but with a fingerboard... but big and heavy. You have to stand next to it to play it.

I bet it would sound interesting!

Have you seen Junior Brown's Guit-steel?

[R.G.]

Every transducer can be characterized as an input-output device
having an energy transfer curve.

Microphones and loudspeakers have sound level vs. voltage;
strain gauges have stress vs. resistance plots.

Guitar pickups stand out because they are rarely so described
except in subjective terms.

I propose the characterizing of guitar pickups as input-output devices.

Charting magnetic field frequency at constant amplitude against
output voltage seems like a no-brainer to me.

Measure the delta-gauss at the drive coil with a hall effect sensor
and adjust for an arbitrary field strength (say, +/-20 gauss) at a given
frequency. That way, the mutual inductance between drive coil and
pickup-under-test is allowed for in a highly reproducible method.

It doesn't need to be perfect -- it only needs to work adequately.

You could also test dynamic range and linearity at an input frequency
by varying the magnetic amplitude. This test might distinguish different
magnet alloys and perhaps whether or not they are completely charged.

Comments welcome.

-drh

That's a good approach, and certainly worth pursuing.

The reason I didn't get into Hall effect sensors on pickups is that magnetic fields being what they are, leakage effects have an enormous effect on coupling. From designing transformers I know that even where you have an almost total flux path in iron you're still going to get leakage. In a pickup/string where there is an air gap measured in inches you're going to get different amounts of coupling to different parts of the coil, and I couldn't come up with anything better than finite element modelling to take that stuff into account.

That's why I gave up and went for an approach that actually moved a string in the flux field of the pickup. It's not perfect - nothing is! - but it is realistic, and it can be made repeatable.

I think...

I'd be very interested in how you do in building one of your Hall-sensor driver coils.

[SteikBacon]

So how do you provide a repeatable picking? Is it like my calibrated picking accessory? Or something different?

I use the best tool I got at hand. Which still is playing experience and use of ear, and (try to) understand whatīs going on. Me and some friends with very different approach to guitars, music and style do tests, but they donīt know what they hear, only I do. So they do blindtest, and I document what they think and compare to mine. Communication is also something we had to learn. Both in what they think of something, and what they expect and want.

The only sad thing for me is that I have not made a pickup yet. Iīm really looking forward to start making them. Need parts.... got the winder ready.

[R.G.]

I builded a testguitar...12 coils...any volumpot... 12 different tonecap, ...and have access to modify anything easily if needed. A very usefull tool.

Any test with strings ... have not told me anything about different pickups. We have done some measurment, but I donīt even think about doing it more because it really doesnīt work. And I donīt believe setup with computer either. For now we have needed to pick the string to get the high freq to be able to estimate anything. So I donīt believe in a setup with regular stringvibration...

Then -

The only sad thing for me is that I have not made a pickup yet. Iīm really looking forward to start making them. Need parts.... got the winder ready.

So you've built a test guitar that can do anything on anything, have beliefs that prevent you from running certain tests, don't believe in setups with computers, only believe that picked-string tests are useful, and so you only test by ear and have never wound a pickup. Is that correct?

Some questions come to mind:
( a ) How many pickups have you tested, and what kind?
( b ) What sort of data does your testing methodology let you collect on each pickup?
( c ) Could you give us a sample of the data you have collected? That would be very interesting.

[Sock Puppet]

I've always wanted to make a large framed string instrument. Like a piano, but with a fingerboard... but big and heavy. You have to stand next to it to play it.

I bet it would sound interesting!

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

[SteikBacon]

Iīve experimented with lots of things... just out of curiosity. Learningprosess. Have tested about 30 pickups, and some have been manipulated in different ways. But the big issue with the testguitar was not only pickups, I used to think of SD as the great ones, but to research the possibilities within guitarelectronics.

Documenting data is just writing down the opinions, and things we find out. I get much info, comparing, different positions, who like it, who donīt, why,... any small variations. If I didnīt write it down I would get confused about "what did which, eh, which did what....". My memories seems to be limited. But Iīm aware that we play, and donīt measure the data. Which is why I liked this thread. How can we get more consistent data.

I do (of course ) believe in using computer. I see you can misunderstand my comment, sorry. I ment the limit in reaching consistent data using strings, vibrating or picked, and compensate with computer.

Strangelove, your method looks interesting.