How to measure frequency spectrum of a thump on power off?

[Horacio Lewinski]

Hello.
I'm running an active speaker with digital xo and have the tweeter with a protection cap connected to a single ended amp (SET). I was considering getting rid of the protection cap, but to make sure this was safe I first connected the SET directly to the midrange driver, turned on/off and listened. When I turn off the amp I hear a noise, a relatively high pitched thump. I measured the transient DC current with a voltmeter and I read 15 to 25mA upon turn-off, which doesn't seem much as it's a pro tweeter rated 80W (if played above 1100Hz). However low frequencies might damage it so I was thinking of a way to measure the spectrum of frequencies at power off.

How can I do this with REW?

 

[trobbins]

With a sampling scope to capture the time-domain of the transient.

With REW you may get a suitable RTA response by checking the 'peak' trace box and monitor the response at the amp output (assuming you have suitable voltage attenuation into your input device so as not to damage it).

When you say 'transient dc current', do you mean you put a current sense resistor in series with the mid-range driver and looked for a transient reading on a voltmeter across the current sense resistor?

When you say 'protection cap', do you mean a capacitor in series with the tweeter, with that circuit driven by a secondary speaker winding on a valve amp output transformer?

Does the SET amp have any inherent low frequency bandwidth limiting, or are you relying totally on the digital X/O ?

 

[Horacio Lewinski]

Thanks for the quick reply!
I didn't fully understand what you indicated, but let me answer your questions

With a sampling scope to capture the time-domain of the transient.

With REW you may get a suitable RTA response by checking the 'peak' trace box and monitor the response at the amp output (assuming you have suitable voltage attenuation into your input device so as not to damage it).

Can I save RTA response? In the past I used it to measure ambient noise, but wasn't able to save it. Transients are short so saving it would be important to be able to analyze it later.

Below I answer your questions:

When you say 'transient dc current', do you mean you put a current sense resistor in series with the mid-range driver and looked for a transient reading on a voltmeter across the current sense resistor?

The SET amp is directly connected to the midrange driver. I removed one connection into the driver and inserted the multimeter in the circuit in series with the driver, set to measure DC current. So the current the amp delivered passed through the multimeter in its way to the driver. I turned on the amp and the multimeter registered "peaks" about 8mA. Naturally the multimeter is not ideal for measuring the real transient, but I noted the reading it was able to register as directionally correct. Also measure DC current while turned on (it was zero), and also measured upon turn-off (15 to 25mA).

When you say 'protection cap', do you mean a capacitor in series with the tweeter, with that circuit driven by a secondary speaker winding on a valve amp output transformer?

Yes, the capacitor in series with the tweeter. The output of the amp connects on one side directly into the tweeter, and on the other side connects into the cap that connects in series to the other connector in the tweeter.

Does the SET amp have any inherent low frequency bandwidth limiting, or are you relying totally on the digital X/O ?

I'm relying totally on digital XO. I'm exposed to DAC failure and user mistake, but I am quite careful and haven't had any issues in the years I've been running this. But removing the protection makes me pause (part of being careful).

 

[trobbins]

Why do you want to remove the series capacitor?

 

[Horacio Lewinski]

reportedly it sounds better without caps, and I'm told it introduces phase distortion. I would like to try and listen, if it's safe for the tweeters. Alternatively, I might need to buy audiophile caps as I used cheap ones to get going. Maybe I try it and conclude I don't hear a difference and I put them back in for safety. But if it sounds better...would be tempting.

 

[Horacio Lewinski]

I measured this today.
Black is ambient noise. No noise in the house.
Red is trace registered when I turned off the amp connected to the midrange, so no protection caps. Interestingly, the trace shows this noise has content from 300Hz and up. The magnitude seems low enough...I think. About 15dB at 600Hz.

Opinions? Safe to use the tweeter without the protection cap?