measuring resonant peak

[Peter Naglitsch]

What right answers do you need?

Got one meselfs. *Works. *Add in a couple of ~60MHz 10:1 probes and a BNC-to-clip-leads cable for a total under $250.

Since the network analysis display is dB vs. frequency, I drove the pickup through an equivalent DC resistance, e.g., *~9k P90 pickup is driven through a ~9k resistor, a got this -6dB plot:

Thanks Salvarsan. Nice to hear that you use this with success. My initial though was that I should use it as a combined signal generator and frequency analyzer (using a dummy coil to induce a frequency sweep into the pickup) and asked them if I could simultaneously use one channel as a send for the generator and the other as input for the analyzer. I should be able to find that out by myself as they have the manual available on line. However, reading on line manuals is something I hate and I might need to print out the manual to find out that myself as they have not responded for a few days (OK weekend right now). Then I found the network analysis mode and I thought that was exactly what was needed. When looking at your plot I assume that the network mode is great to find the resonant peak frequency, but not for plotting the frequency response curve. For that I need (?) the double function I have described above, or do I? I would be very happy if you could give me a hint here, Salvarsan.

[Possum]

....

Peter you do this for free. Go find Audacity's website here:
Audacity: Download

Record your sweep, select the waveforms from the timeline and go to the menu and have it analyze the frequency response, screenshot attached....

[Peter Naglitsch]

Thanks Possum, I'll have to look into that. I have used a software based generator/analyzer previously, but wasn't that satisfied with the result. That was why I was looking into HW based solutions.

[salvarsan]

Thanks Possum, I'll have to look into that. I have used a software based generator/analyzer previously, but wasn't that satisfied with the result. That was why I was looking into HW based solutions.

Distinctions: A digital O-scope is a general purpose tool that will do a lot more than look at pickups. *Audacity is primarily an audio editor with some incidental test modes added.

BTW, if you want to run a frequency sweep quickly, the CGR-101 has a noise generator. *Turn it on and run the CGR in sprectrum analysis mode for a quick response curve, conceptually the same as what you would do using Audacity.

-drh

[Possum]

..

If you're primarily interested in frequency response curves audacity or any free software will do that fine. If you want to see peak resonance Joe's method works fine. None of it replaces amp and ear testing, but it is educational to play with for awhile. In a day to day production world, the Extech is the only really useful tool I've found that gets used all the time.

[Jeff Callahan]

Couldn't resist getting one of these, but my inexperience with scopes has me pissed! Just got it in the mail, and can't seem to figure out how to connect it to a test pickup. Any help would be appreciated.

Jeff

[salvarsan]

Couldn't resist getting one of these, but my inexperience with scopes has me pissed! Just got it in the mail, and can't seem to figure out how to connect it to a test pickup.

Here is a quick schematic:



Connect a ~1.5k (1500 ohm) resistor to the pickup hot lead.

Signal Generator output goes across resistor+pickup and ground.
Probe A goes across resistor+pickup and ground.
Probe B goes across pickup and ground.

Start the CGR101 software.
Go to the network analyser screen.
Set the upper+lower frequencies to something rational like 50 and 15000 Hz.
Start the scan and go make coffee while it finishes.

Good luck and keep us updated on the problems.

[Mike Sulzer]

H

Connect a ~1.5k (1500 ohm) resistor to the pickup hot lead.

Have you tried a larger resistor? A pickup can be several hundred Kohm at resonance, and so there is a very small voltage drop across the 1.5 Kohm resistor. This makes it more difficult to measure the location and width of the peak accurately.

[salvarsan]

Have you tried a larger resistor? A pickup can be several hundred Kohm at resonance, and so there is a very small voltage drop across the 1.5 Kohm resistor. This makes it more difficult to measure the location and width of the peak accurately.

Sorry. I used 1.5K for testing woofers.
The P-90 tests used a resistor equal to the DCR, about 9k.

In network analyzer mode, the CircuitGear application also graphs phase angle that shows a zero-crossing at resonance, so the location is obvious enough.

[peskywinnets]

You should be on commision - on the strength of those screenshots, I've downloaded the manual & it looks ideal.

I've already a Chinese USB scope but the featureset is weak, the sw buggy (FFT no workeee), the manual sucks, the sw upodates have ceased the emails don't get answered.

So, tomorrow I will drop £180 ...there's only one dealer here in the UK (that's a bad consumer position to be in!) ....the only downside is that the CircuitGear scope GUI itself looks awful - white background?! (eek) ....is the display configurable?

[salvarsan]

..the only downside is that the CircuitGear scope GUI itself looks awful - white background?! (eek) ....is the display configurable?

The software was written in Tcl/Tk; source code comes with the package.

If you can program in C/C++, Java, or Python, you can handle Tcl.
If you can't, there are other folks here who'll take a whack at it.

[peskywinnets]

The software was written in Tcl/Tk; source code comes with the package.

If you can program in C/C++, Java, or Python, you can handle Tcl.
If you can't, there are other folks here who'll take a whack at it.

Thanks salvarsan, alas picbasic is about my limit!

Actually I think I've found a good (better?) contender - a velleman PCSGU250 ...it's a little bit cheaper (& the spec a little bit better), but what's really got me interested is the support - check out how quickly the developers reacted to probs & suggestion wrt the inbuilt bode plotter...

http://forum.velleman.eu/viewtopic.p...st=0&sk=t&sd=a

[salvarsan]

Actually I think I've found a good (better?) contender - a velleman PCSGU250 ...it's a little bit cheaper (& the spec a little bit better), but what's really got me interested is the support - check out how quickly the developers reacted to probs & suggestion wrt the inbuilt bode plotter...

Just so you understand that it's for Windows only.

Scopes are spec'd differently, too, since they must display phase differences at their rated limit.

A 2 MHz digital scope needs 20 Msamples/second.

In that respect, 25 Msamples/second isn't a great improvement.

Either one is plenty for audio work.

[peskywinnets]

Well, after having a dabble with Velleman's downloadable demo s/w here...

http://www.vellemanusa.com/downloads...00lt_v1_09.zip

...I was suitably impressed (btw: I ve only have 'puters running Windows)

For us Brits, the Velleman is working out about £50 cheaper vs the Circuitgear - it's £135inc VAT delivered from here http://www.esr.co.uk/velleman/pcsgu250.htm )

I've just pulled the trigger on one...it'll be with me tomorrow (wehey - 'phase central' testing over the weekend of a few self made circuits I've been curious about for a while! I'm curious to see where the resonant peak is for a few of my self made low impedance pickups too).

I presently have a Chinese DSO-2150 USB Scope - a much higher bandwidth, but its feature set isn't great. But alas, they appear to sell for peanuts on Ebay (£65 for a 60Mhz bandwidth storage USB scope anyone? http://cgi.ebay.co.uk/Hantek-DSO-215...item3a63c44837 ) , so'll I'll just hang onto it for when dabbling with high frequency digital (PICs)

It's weird, becuase all the higher spec'ed USB scopes appear to have far less features vs the CircuitGear/Velleman - they're all missing a trick IMHO...

[salvarsan]

(btw: I ve only have 'puters running Windows)

wimp.

For us Brits, the Velleman is working out about £50 cheaper vs the Circuitgear - it's £135 delivered from here Velleman USB PC Scope and Generator )

Money talks. VAT on the canadian CircuitGear doesn't look good.

I've just pulled the trigger on one...it'll be with me tomorrow (wehey - 'phase central' testing over the weekend of a few self made circuits I've been curious about for a while! I'm curious to see where the resonant peak is for a few of my self made low impedance pickups too).

Congrats. The Velleman's normalized Bode plot is a feature I'd like to have on the CircuitGear.

I presently have a Chinese DSO-2150 USB Scope - a much higher bandwidth, but its feature set isn't great. But alas, they appear to sell for peanuts on Ebay (£65 for a 60Mhz bandwidth storage USB scope anyone? Hantek DSO-2150 USB digital oscilloscope on eBay (end time 01-Mar-11 20:48:48 GMT) ) , so'll I'll just hang onto it for when dabbling with high frequency digital (PICs)

It's weird, becuase all the higher spec'ed USB scopes appear to have far less features vs the CircuitGear/Velleman - they're all missing a trick IMHO...

By the standard scope measure, that Hantek's 150 Msample/sec entitles them to claim 15MHz analog bandwidth. Since they don't open up the software interface (as do Velleman and CircuitGear), you may want to resell it.