Temperature Control In Speaker Measurements

[Jon AA]

That'll be a good test. If you can, try and track the differences over time (after 5 minutes of constant playing, 10 minutes, etc).

 

[scott wurcer]

I have two identical Yamaha woofers and they have the same impedance sweeps when at room temperature.

Just put one in the fridge (see pic). Let's see what happens in half an hour or so, I'll compare a room temperature one with a cold one.

View attachment 50760

Who else in the world has a woofer in their fridge, huh?

Fans of Peter Belt?

 

[restorer-john]

Not me, but you must admit the barrier to enter that exclusive club is not a very high hurdle.

Not as exclusive as you would think...

 

[restorer-john]

That'll be a good test. If you can, try and track the differences over time (after 5 minutes of constant playing, 10 minutes, etc).

OK, here you go:

The room temperature control woofer is the white sweep.
The other sweeps are the "cold" woofer coming up to room temperature (24.1 degrees at the time).
Yellow is 10 degrees C
Purple is 13 degrees C
Red is 17 degrees C
Green is 20 degrees C

As you can see, the resonant peak is lower in impedance and higher in frequency when cold, and increases in impedance and heads closer to the room temp woofer as it warms up.

Essentially as expected. The spider and surround become more pliable as they warm up. I cannot see how similar winter temperatures in a cold garage wouldn't affect low frequency tuning or bass output to some degree.

I've got tons of speakers and a spare fridge downstairs...

 

[scott wurcer]

OK, here you go:
View attachment 50764
The room temperature control woofer is the white sweep.
The other sweeps are the "cold" woofer coming up to room temperature (24.1 degrees at the time).
Yellow is 10 degrees C
Purple is 13 degrees C
Red is 17 degrees C
Green is 20 degrees C

As you can see, the resonant peak is lower in impedance and higher in frequency when cold, and increases in impedance and heads closer to the room temp woofer as it warms up.

Essentially as expected. The spider and surround become more pliable as they warm up. I cannot see how similar winter temperatures in a cold garage wouldn't affect low frequency tuning or bass output to some degree.

I've got tons of speakers and a spare fridge downstairs...

There's also the fact that copper has a ~3000ppm/degree TC. This might all end up being trivial but it is important to at least do due diligence. BTW this has absolutely nothing to do with burn in at all.

 

[hardisj]

Not really a surprise *that there is a difference*.

I think the real interesting thing to know would be: what happens after signal is applied to each for some period of time while in the cold environment. Run wire in the fridge to the speaker, play a test tone at resonant frequency for an hour at half-excursion and exercise the suspension. Then test again. You don't have to do it with varying temps. Just do it at the coldest one.

What I'm hoping/expecting is that it renders the impedance sweep moot and might add credence to a pre-test warm up with some sort of signal (pink noise, test tones, etc) so that there's at least an attempt at properly warming up the speakers so they're always loose, no matter the environmental conditions.


Aside from the spider, I do know that humidity and temp play a factor in frequency response in the pro-audio environment. I don't know this from experience. Just what I've read from a concert sound-rigger's point-of-view. There's a lot of info on Google about it. But the general theme is that it only matters in this environment when you're far... very far (100's of feet) away from the speaker.
https://www.sweetwater.com/insync/effects-of-temperature-humidity-live-sound/


Of course, Klippel themselves makes mention of the NFS being a better way to measure than standard methods because it's impervious to temperature issues as well:
https://www.klippel.de/products/rd-system/modules/nfs-near-field-scanner.html
http://warkwyn.com/wp-content/uploads/2015/01/NFS-ALMA-2015-PPT.pdf (page 7)

 

[restorer-john]

BTW this has absolutely nothing to do with burn in at all.

Who mentioned burn-in? I didn't.

Run wire in the fridge to the speaker, play a test tone at resonant frequency for an hour at half-excursion and exercise the suspension. Then test again. You don't have to do it with varying temps. Just do it at the coldest one.

You want me to play test tones to my meat and vegetables? People will think we are nuts.

 

[hardisj]

You want me to play test tones to my meat and vegetables? People will think we are nuts.

Dude! Don't talk about your nuts and meats and vegetables on a public forum!!!!

 

[scott wurcer]

Who mentioned burn-in? I didn't.

IIRC someone mentioned it, I thought confusing simple warm up with the concept of "burn in". If I was mistaken no biggie.

 

[Don Hills]

... You want me to play test tones to my meat and vegetables? People will think we are nuts.

I see what you did there.

 

[restorer-john]

Perhaps Amir should identically test the same speaker in winter and in summer and compare the results.

I am also wondering what happens to tweeters' parameters when they are physically cold or hot. I figured bass drivers would be the obvious ones for variation due to the reliance on compliant suspension components which are affected significantly by temperature.

edit: Just took a Yamaha silk dome tweeter and heated it from 24 degrees to 46 degrees C. both plots laid on top of each other- no significant difference.

 

[scott wurcer]

I am also wondering what happens to tweeters' parameters when they are physically cold or hot. I figured bass drivers would be the obvious ones for variation due to the reliance on compliant suspension components which are affected significantly by temperature.

I was also thinking old style paper cones could possibly be affected by large variance in RH.

 

[andreasmaaan]

OK, here you go:
View attachment 50764
The room temperature control woofer is the white sweep.
The other sweeps are the "cold" woofer coming up to room temperature (24.1 degrees at the time).
Yellow is 10 degrees C
Purple is 13 degrees C
Red is 17 degrees C
Green is 20 degrees C

As you can see, the resonant peak is lower in impedance and higher in frequency when cold, and increases in impedance and heads closer to the room temp woofer as it warms up.

Essentially as expected. The spider and surround become more pliable as they warm up. I cannot see how similar winter temperatures in a cold garage wouldn't affect low frequency tuning or bass output to some degree.

I've got tons of speakers and a spare fridge downstairs...

The effect can be modelled by altering the Qms and Rms of a driver. Here's an example. Top two traces are the SPL and impedance of a 6" woofer in a (roughly) QB3 ported box. The driver has a Qms of 4.33 and an Rms of 0.57.

Pulling the Qms down to 2 (which brings Rms up to 1.23), you get the lower two traces. Of course, for the purpose of illustrating the effects, this is a more extreme case than John's refrigerator or presumably Amir's garage:

 

[andreasmaaan]

^^Of course, it's not a perfect model as it doesn't take into account the effect on Fs.

EDIT: there it is with the Fs also raised 10%:

 

[amirm]

IIRC someone mentioned it, I thought confusing simple warm up with the concept of "burn in". If I was mistaken no biggie.

I said that. With burn-in tests, as with John's, it can easily be shown to make the driver measure differently. But put them in a box and variation in frequency response becomes negligible.

 

[scott wurcer]

But put them in a box and variation in frequency response becomes negligible.

Putting things in a box does not negate the environmental issues. Typical MIL spec was -55 to 125C I don't think an average speaker would have no variation in response if subjected to this. This might all be a tempest in a tea pot but it's easy to put it to bed with a few measurements.

 

[amirm]

This might all be a tempest in a tea pot but it's easy to put it to bed with a few measurements.

If only. I must have post the Harman break-in research one hundred times and people still believe it makes a difference. As I said no different than this discussion. What make sense to lay intuition wins over real results....

 

[amirm]

Typical MIL spec was -55 to 125C I don't think an average speaker would have no variation in response if subjected to this.

What does that have to do with anything? Speakers are used indoors so the reasonable temp range is what indoor temps would be, not milspec.

 

[RayDunzl]

My room is currently 78F and 46% relative humidity.

It will not range according to military requirements.

Ever.

By my estimation.

 

[Thomas savage]

13c is not cold , Amirm is just a wuss.

The speaker is active for a good amount of time during the test anyway and in near ' room temperature ' conditions that are maintained year round according to Amirm.

There's nothing to see here .

Now the lunar cycle, what's Klippel have to say about that ? Best to be safe and ' scientific ' and test all speakers at the same time in the cycle .

We need proof so Amirm you must take a picture of the moon and post it in the review .

This is a science forum after all.