ampage Tube Amps / Music Electronics	For current discussions, please visit Music Electronics Forum.

ampage archive

Vintage threads from the first ten years

Output Measurments

4/26/2006 6:04 PM
stan	Output Measurments Hi all, it was recently brought to my attention by a customer that LD has a spec on his site for output of his pickups (i.e. output: 270). I can only assume that this is ampres, or actually miliampres. It only stands to reason that the higher the output of the pickup, the more current it will produce, and vica-versa. My questions are, how true an indicator do you think that this reading is and how would one set up to take the measurment?

4/26/2006 6:04 PM

stan

Output Measurments

Hi all, it was recently brought to my attention by a customer that LD has a spec on his site for output of his pickups (i.e. output: 270). I can only assume that this is ampres, or actually miliampres. It only stands to reason that the higher the output of the pickup, the more current it will produce, and vica-versa. My questions are, how true an indicator do you think that this reading is and how would one set up to take the measurment?

4/26/2006 6:23 PM
Dave Stephens	Joe can probably answer this but there is another side of this which to me makes it all kind of nonsensical. We're really talking about LOUDNESS here, not output. And the truth about loudness is that treble pickups sound louder than hot wound pickups with the treble clipped off from too much wire. I have a '51 repro style tele bridge pickup and its wound real low, it sounds louder than my hotter wind tele bridge pickups because it really cuts through. Output and loudness are two different things really so what do you want, a loud pickup or a high output pickup that isn't going to anything but crunchy cream?

4/26/2006 6:23 PM

Dave Stephens

Joe can probably answer this but there is another side of this which to me makes it all kind of nonsensical. We're really talking about LOUDNESS here, not output. And the truth about loudness is that treble pickups sound louder than hot wound pickups with the treble clipped off from too much wire. I have a '51 repro style tele bridge pickup and its wound real low, it sounds louder than my hotter wind tele bridge pickups because it really cuts through. Output and loudness are two different things really so what do you want, a loud pickup or a high output pickup that isn't going to anything but crunchy cream?

4/26/2006 10:31 PM
Luijo	That's millivolts.

4/27/2006 12:39 AM
Steve A.	Here's the chart that I nicked from the Bartolini site (in millivolts): http://www.blueguitar.org/new/misc/pickup_mv_chart.pdf While these measurements might have been made with a scope, I will use a DMM and whack the heck out of my guitar to get a maximum reading. I'll also do the same meaasurement at my speaker output, and usually the bass of the neck pickup will blow the bridge pickup out of the water... although subjectively I think that our ears are more sensitive to the treble sounds... HTH Steve Ahola

4/27/2006 12:39 AM

Steve A.

Here's the chart that I nicked from the Bartolini site (in millivolts):

http://www.blueguitar.org/new/misc/pickup_mv_chart.pdf

While these measurements might have been made with a scope, I will use a DMM and whack the heck out of my guitar to get a maximum reading. I'll also do the same meaasurement at my speaker output, and usually the bass of the neck pickup will blow the bridge pickup out of the water... although subjectively I think that our ears are more sensitive to the treble sounds...

HTH

Steve Ahola

4/27/2006 9:27 AM
moocow	The Feltcher-Munson curves demonstrate that human hearing is not more sensitive to high frequencies than to low. Your ears aren't lying to you because guitar pickups really do put out more voltage at higher frequencies. If you read up on electrical generators, voltage is dependent upon the number of 'lines of magnetism' that are cut per second. Cutting more lines gives more voltage. A string that vibrates at a higher frequency cuts more lines of force than a string vibrating at a lower frequency so the output is higher. This means that the 'frequency response' of the voltage generated within the pickup is a ramp, proportional to frequency. But the voltage generated by the pickup is not the voltage that reaches the speakers, or the amplifier for that matter. The pickup inductance, cable capacitance, and associated guitar electronics (wiring, pots, etc.) work together to create an EQ for the pickup voltage. At the resonant frequency (pickup inductance and cable capacitance), the pickp voltage gets choked off and the voltage at the amp begins to decrease. If we could somehow hear the pickup voltage, without this EQ-ing, our guitar pickups would sound absolutely awful. Years ago, I attended a Fender guitar/amp clinic at a local music store. The Fender representative was showing us the relatively new Lace sensors and told us how when the first pickups were demonstrated for Fender, they sounded terrible because the frequency response was too wide. Lace had to go back and redesign the pickup to make it sound more acceptable. I think the first design had almost no inductance so the voltage generated by the pickup actually did get to the amp and caused the awful sound. Also, I have some familiarity with a high-speed turbine that used magnetic pickups to measure RPM: http://www.smith-systems-inc.com/products_and_services/application_specific_design/custom_aerospace/ I used to be resonsible for the one on the bottom right corner, with the white wire hanging out of it. The sensor has a magnet at the back end in contact with a long iron rod. The rod is wrapped with wire, just like a guitar pickup. The end of the assembly is placed near the turbine shaft, which has a smooth groove machined on both sides. As the shaft spins, the groove passes the end of the sensor and causes a voltage to be developed, very similar to the way a vibrating string creates a voltage within a guitar pickup. As you ca imagine, the frequency response and output voltage for the speed sensor was very well defined. Since the shaft was always the same distance from the sensor, all we had to do was rotate the shaft at different RPM's, take the voltage readings, plot them, and we would have our pickup frequency response. The response plot looked just as I have described, a straight ramp up to about 15kHz. At this frequency, the inductance began to interact with the associated wiring capacitances so the voltage began to drop off. But we ran our turbine at about 5kHz (72,000 RPM) so the 15kHz frequency response was more than enough. With all of this in mind, it seems to me that this Bartolini number is not very well-defined. It doesn't specify a frequency or a cable capacitance, both of which will affect the output. Pot values especially will affect the peak voltage at the amplifier input. If Bartolini does not specify the test conditions, their spec is worthless. It may be that Bartolini tested the pickup with no volume/tone pots at all, and with a very short cable. This will greatly increase the voltage reading, but it will be a misleading number if you happen to have volume and tone controls on your guitar. This would also be a misleading test because it would favor pickups with low inductance/high resonant frequency. But once the pickup was placed in a guitar circuit, the resonant frequency would drop as would the peak output voltage. Smith Systems uses a standard test for their magnetic pickups. They rotate a toothed wheel near the sensor, which disturbes the magnetic field at a known frequency. The output voltage is measured as the 'frequency' changes and this gives the frequency response of the sensor. A similar method could be used for guitar pickups, but the test conditions must be specified. You could measure the voltage out-of-circuit, but this information would be somewhat misleading. An in-circuit test would be more realistic as it would include the EQ effects of the entire signal path, but you would need to specify the pot values, cable capacitance, etc. that you used in the test if it is to have any meaning.

4/27/2006 9:27 AM

moocow

The Feltcher-Munson curves demonstrate that human hearing is not more sensitive to high frequencies than to low. Your ears aren't lying to you because guitar pickups really do put out more voltage at higher frequencies.

If you read up on electrical generators, voltage is dependent upon the number of 'lines of magnetism' that are cut per second. Cutting more lines gives more voltage. A string that vibrates at a higher frequency cuts more lines of force than a string vibrating at a lower frequency so the output is higher. This means that the 'frequency response' of the voltage generated within the pickup is a ramp, proportional to frequency.

But the voltage generated by the pickup is not the voltage that reaches the speakers, or the amplifier for that matter. The pickup inductance, cable capacitance, and associated guitar electronics (wiring, pots, etc.) work together to create an EQ for the pickup voltage. At the resonant frequency (pickup inductance and cable capacitance), the pickp voltage gets choked off and the voltage at the amp begins to decrease.

If we could somehow hear the pickup voltage, without this EQ-ing, our guitar pickups would sound absolutely awful. Years ago, I attended a Fender guitar/amp clinic at a local music store. The Fender representative was showing us the relatively new Lace sensors and told us how when the first pickups were demonstrated for Fender, they sounded terrible because the frequency response was too wide. Lace had to go back and redesign the pickup to make it sound more acceptable. I think the first design had almost no inductance so the voltage generated by the pickup actually did get to the amp and caused the awful sound.

Also, I have some familiarity with a high-speed turbine that used magnetic pickups to measure RPM:

http://www.smith-systems-inc.com/products_and_services/application_specific_design/custom_aerospace/

I used to be resonsible for the one on the bottom right corner, with the white wire hanging out of it. The sensor has a magnet at the back end in contact with a long iron rod. The rod is wrapped with wire, just like a guitar pickup. The end of the assembly is placed near the turbine shaft, which has a smooth groove machined on both sides. As the shaft spins, the groove passes the end of the sensor and causes a voltage to be developed, very similar to the way a vibrating string creates a voltage within a guitar pickup. As you ca imagine, the frequency response and output voltage for the speed sensor was very well defined. Since the shaft was always the same distance from the sensor, all we had to do was rotate the shaft at different RPM's, take the voltage readings, plot them, and we would have our pickup frequency response. The response plot looked just as I have described, a straight ramp up to about 15kHz. At this frequency, the inductance began to interact with the associated wiring capacitances so the voltage began to drop off. But we ran our turbine at about 5kHz (72,000 RPM) so the 15kHz frequency response was more than enough.

With all of this in mind, it seems to me that this Bartolini number is not very well-defined. It doesn't specify a frequency or a cable capacitance, both of which will affect the output. Pot values especially will affect the peak voltage at the amplifier input. If Bartolini does not specify the test conditions, their spec is worthless. It may be that Bartolini tested the pickup with no volume/tone pots at all, and with a very short cable. This will greatly increase the voltage reading, but it will be a misleading number if you happen to have volume and tone controls on your guitar. This would also be a misleading test because it would favor pickups with low inductance/high resonant frequency. But once the pickup was placed in a guitar circuit, the resonant frequency would drop as would the peak output voltage.

Smith Systems uses a standard test for their magnetic pickups. They rotate a toothed wheel near the sensor, which disturbes the magnetic field at a known frequency. The output voltage is measured as the 'frequency' changes and this gives the frequency response of the sensor. A similar method could be used for guitar pickups, but the test conditions must be specified. You could measure the voltage out-of-circuit, but this information would be somewhat misleading. An in-circuit test would be more realistic as it would include the EQ effects of the entire signal path, but you would need to specify the pot values, cable capacitance, etc. that you used in the test if it is to have any meaning.

4/27/2006 12:19 PM
Dr. Strangelove	RFQ: Standard pickup load for single coils Here's a simple suggestion for a standard load that trivially replicates the guitar, cable, and amp passive loads. <http://www.salvarsan.org/amps/pickupload.html> -drh --

4/27/2006 3:19 PM
moocow	That's pretty good! I've run tests where I run a cable from the guitar to the amp, then measure the voltage at the amp itself. My circuit simulations use a similar circuit, but I model the volume and tone pots as post, that way, I can run multiple simulations and see the frequency response for various volume or tone control settings. I also include the parallel 68K resistors at the input of the amp, but they really don't have much of an effect on the simulation results. I used to include the input capacitance of the first stage vacuum tube but it is only 2pF or so, not worth modeling since the cable capacitance is so much larger. Going back to the Smith Systems pickup, the previous version had a much larger inductance. The spec for that pickup indicated a simple resistive load of 10K ohms, but during tests, a very long cable was attached to the sensor. It turns out that the cable capacitance formed a resonant circuit with the sensor inductance, and the resonant frequency just happened to be twice the frequency that the turbine was running. The resonance was creating a 'double-pulse' at the input to the speed controller, which sometimes made the controller think the turbine was spinning way too fast. This would cause shutdowns of the turbine during test. Fortunately, this only happened in the test cell and not when the turbine was installed and in use. Eventaually, these types of sensors began to fail because of thermal cycling, so a new sensor was developed by Smith Systems. The new sensor has a very, very low inductance and the resonance was very well damped and did not cause any problems during ground testing. Anyway, this is sort of a long story, but it shows how cable capacitance affects the behavior of magnetic pickups. I ran simulations nearly ten years ago that showed where the double-pulses came from in the first place, but I didn't use the same methods to study guitar pickups until maybe two years ago. I know guitar players are probably skeptical about this simulation data of guitar pickups, but it was good enough for our customer.

4/27/2006 3:19 PM

moocow

That's pretty good! I've run tests where I run a cable from the guitar to the amp, then measure the voltage at the amp itself. My circuit simulations use a similar circuit, but I model the volume and tone pots as post, that way, I can run multiple simulations and see the frequency response for various volume or tone control settings. I also include the parallel 68K resistors at the input of the amp, but they really don't have much of an effect on the simulation results. I used to include the input capacitance of the first stage vacuum tube but it is only 2pF or so, not worth modeling since the cable capacitance is so much larger.

Going back to the Smith Systems pickup, the previous version had a much larger inductance. The spec for that pickup indicated a simple resistive load of 10K ohms, but during tests, a very long cable was attached to the sensor. It turns out that the cable capacitance formed a resonant circuit with the sensor inductance, and the resonant frequency just happened to be twice the frequency that the turbine was running. The resonance was creating a 'double-pulse' at the input to the speed controller, which sometimes made the controller think the turbine was spinning way too fast. This would cause shutdowns of the turbine during test. Fortunately, this only happened in the test cell and not when the turbine was installed and in use. Eventaually, these types of sensors began to fail because of thermal cycling, so a new sensor was developed by Smith Systems. The new sensor has a very, very low inductance and the resonance was very well damped and did not cause any problems during ground testing.

Anyway, this is sort of a long story, but it shows how cable capacitance affects the behavior of magnetic pickups. I ran simulations nearly ten years ago that showed where the double-pulses came from in the first place, but I didn't use the same methods to study guitar pickups until maybe two years ago. I know guitar players are probably skeptical about this simulation data of guitar pickups, but it was good enough for our customer.

Page 1 of 5

Next>

Last Page>>