Good way to measure transient distorsion ??

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
youyoung21147 said:
Yes, it's true that mixers compress dynamics a lot to make CDs more audible.

Do you think a modified Panasonic WM-61A is a good enough ECM to do some decent sound capture ?

I have no idea since i havent seen one...
One mic you could try is lineaudio CM3 , i havent tried it so extensively myself but it is dirtcheap. http://www.lineaudio.se/linemic.htm
However, you must remember that a mic doesnt posess any kind of intelligence at all, so micplacement is very important.
 
Svante : Let's suppose I have two completely unsmoothed frequency response graphs : how can you identify a driver which has very fast decay and another which has slower decay without any other graph ? I'm very interested in that because you often find frequency response measurements, but only that !


electroaudio : those mikes look nice ! 11dB noise floor and 73dB SNR is pretty good ! How much do they cost in dollars or euros ?
 
If the smoothing is very low (I don't know that you can make a graph with absolutely no smoothing), then a linear driver would have a very flat frequency response, without peaks or dips in the response. A driver with energy storage would have pronounced peaks and dips in the response- possibly quite narrow ones.

Here's an experiment you can do at home- take a small weight, and hang it from a rubber band. Hold the end of the rubber band in your hand and move your hand up and down, slowly at first. The position of the weight represents the output from the speaker. As you move slowly, the weight follows your hand's movement closely. As you approach a certain critical frequency (the resonant frequency), the weight will start to move more than your hand. The behavior above this frequency is a topic for another time.

If you made a graph of weight displacement versus hand movement, it would be quite easy to see where the resonant frequency was. Furthermore, a few points along that graph would tell you the quality of the resonant system, allowing you to see how long it would take the system's resonance to decay below a certain point given a known input (frequency and magnitude).

I'm sorry if what I've said is just a repetition of a physics class you've already been through. Cone vibration is a lot like the weight and rubberband, but more complex. The inputs aren't as simple as a moving hand (although AFAIK they don't often contribute to energy storage), and the resilient parts aren't a one-dimensional rubber band- which is why you see resonances at many different frequencies, and with many different Q-factors.

Don't trust a manufacturer's frequency response measurements, unless you specifically trust that manufacturer. Some companies have fairly accurate spec sheets- but many do not! Their job is to sell drivers, not to help you learn that someone makes better drivers than them.
 
Thanks for your answer joe !

What immediately came to my mind when you talk about resonant frequency are the Seas magnesium drivers, which are perfectly clean until you reach the breakup frequency.

Those are pretty easy to estimate because they behave like true pistons until breakup.

But paper cones... What a pain !! Some are damped in order not to have breakup bumps in FR, but their CSD sometimes rings in places you wouldn't expect.

BTW : do you think this driver is clean-sounding ?

http://www.supravox.fr/haut_parleurs/165_GMF.htm
http://www.supravox.fr/mesures/mes165GMF1.htm
http://www.supravox.fr/mesures/mes165GMF2.htm
 
It looks good, but nothing special. Do you see the bumps in the impedance phase curve? The one around 600 hz is fairly typical of a lot of drivers, but it's not desirable. The thing that concerns me is that there seems to be something else going on in the impedance curve just above and around 1khz.

It's a very sensitive driver, at 96 db. Part of the way they get this sensitivity is by using a very lightweight cone. The problem is that you can't really make a lightweight paper cone that stiff. It is definitely not moving pistonically. Their frequency response and CSD graphs really don't seem to be very informative. What would be informative is if you could see the impedance curve magnified and unsmoothed a bit more. I think you'd find it to be fairly rough.

It might sound "nice", but it won't be anywhere as clean as a well implemented Seas magnesium (or even aluminum) cone. The tradeoff is the more pistonic cone (in this case) is a good 10 db less efficient.

If the thermal power handling is OK, then these supravox drivers could do well in a pro sound environment. Their response isn't terrible, and they're quite efficient.
 
youyoung21147 said:
I wonder where I can find only this kind of ECM capsules, the rest doesn't interest me :devilr:

Well, this is a true condenser and i think the diagraphms are made inhouse just like any other professinal mics, but the electronics and the housing is also important for the final sound.


Originally posted by youyoung21147
Wow, that's pretty expensive compared to my DIY stereo mic, based on Panasonic ECM. It cost me roughly 30€ with the preamp and everything :D


Well, the mics i normally use in my work are more than 10 times as expensive...
This is my favourite for djembe/congas/bongos, and i dont have one...
http://www.frontendaudio.com/ADK_SL_F_E_T_Large_Diaphragm_Condenser_Microphone_p/1021.htm
And ofcourse my favourite vocalmic, i only have three...
http://www.mtlc.net/products/1343/LSR-3000+Condenser+Microphone/learnmore/

Hmm, maybe this is something in your pricerange... :devilr:
http://www.fullcompass.com/product/257890.html
http://www.mtlc.net/products/1350/DM1001+Digital+Condenser+Microphone/learnmore/
 
youyoung21147 said:
It must be really enojyable to work with such tools.

Both yes and no.
Sometimes it can be very close to heaven but it is also very troublesome to use studiomics on a stage.

youyoung21147 said:
BTW : what is the highest SNR a microphone can reach ? We build electronics that have 112dB SNR, speakers that have 0.2% distorsion, but what about microphones ??

Actually, i have no idea....
Since i do livesound so i havent had any problems with noise (Except the audiences selfnoise) but i suppose those who works with moviesound has a lot more problems with that.
However, distortion is less than 0,5% within their SPL range on most condensers i have seen, dynamics are (judged by their sound) a lot worse.
But to be honest, i have only cared about if i like the sound of the mic or not, so i havent studied any numbers, Sorry.

PS, this must be the best measurementmics aviable so see if you can find anything useful here: http://www.bksv.com/3026.asp
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.