Speaking of poor peoples Klippel
Always perform a simple test right after unboxing your investment. Stick a 1.5V / 4.5 / 6V/ 9V... battery (or your trusted cheap continuously variable DC power supply) to the terminals of a LF cone transducer to see the excursion. Then reverse polarity. The excursion away from the rest position has to be the same! In and out! If not, the "investment" has to be returned and refunded!
Otherwise there is no chance at all your project will perform ok.
The same applies to your well guarded treasure from the past.
Do not forget to unhook the DC after a very short time! Always turn down the DC before connecting the first time! @anatech ;-)
Always perform a simple test right after unboxing your investment. Stick a 1.5V / 4.5 / 6V/ 9V... battery (or your trusted cheap continuously variable DC power supply) to the terminals of a LF cone transducer to see the excursion. Then reverse polarity. The excursion away from the rest position has to be the same! In and out! If not, the "investment" has to be returned and refunded!
Otherwise there is no chance at all your project will perform ok.
The same applies to your well guarded treasure from the past.
Do not forget to unhook the DC after a very short time! Always turn down the DC before connecting the first time! @anatech ;-)
Last edited:
Or in the poormans way, Zmax-Re(V) and Fs(V), which are directly proportional to Bl(x) and Kms(x) 😉 🙂
When I look at OEM FR curves or third party review curves, I look for a couple of things.
I look for evidence of cone/dome breakup. This can be seen in the off-axis behavior. The example below is a 10" woofer from Eton. We see the response of the off-axis starts to fall off above 700 Hz compared to the on-axis. But then at ~ 1600 Hz the off axis becomes ragged, and seems unconnected from the on-axis. This is a clear sign of the first breakup mode.
This tells me the upper limit of the drivers range. It tells me if I will need some sort of notch filter with this driver. It helps me guess where the crossover frequency might be, and how steep the slope will be.
Another thing I look at, particularly with midranges and tweeters, is the overall smoothness and uniformity of the in-band response. I would much rather work with a driver that needs little eq rather than one that has peaks and valleys to deal with.
j.
Just look at the uneven harmonics, and you'll know immediately. 🙂
Sometimes this will also show up in the impedance graph as well.
Looking at off-axis graphs can be tricky, because sometimes diffraction effects can disturb this for example.
(obviously not when measurements were done on a AES baffle)
You can also estimate dome behaviour of soft domes with off axis FR. These domes don't move like a piston from pretty low frequencies (8-10kHz) and that shows in off axis behaviour.
Hi b_force,
I didn't miss anything.
What I said is that if you measure parameters differently than the manufacturer did (or anyone else), do not expect to get the same numbers. That is all I said. I will agree with you if you want to lament the fact that manufacturers don't use the same methods. That is an entirely different argument and has zero to do with what I posted.
Hi Hörnli,
There is a much easier and more accurate way of doing this. I have rejected many recones and new drivers by doing this. The same exact test as looking for resonance with a scope. This will show any non-linearities. You can vary the drive level, and it is far more sensitive than measuring displacement. Best of all, you can't hurt anything unless you get stupid with drive levels.
I didn't miss anything.
What I said is that if you measure parameters differently than the manufacturer did (or anyone else), do not expect to get the same numbers. That is all I said. I will agree with you if you want to lament the fact that manufacturers don't use the same methods. That is an entirely different argument and has zero to do with what I posted.
Hi Hörnli,
I wouldn't. This test can so easily show incorrect results. Your mechanical measurement of displacement has to be done extremely carefully.
There is a much easier and more accurate way of doing this. I have rejected many recones and new drivers by doing this. The same exact test as looking for resonance with a scope. This will show any non-linearities. You can vary the drive level, and it is far more sensitive than measuring displacement. Best of all, you can't hurt anything unless you get stupid with drive levels.
No problem, it was just a misreading or miscommunication. 👍😊
From your initial post I was reading it differently.
But yeah, some manufacturers are so bad that they use a different method per driver even 😕
You know, I completely agree, Given a chance, manufacturers who have products that don't perform as well will lie their azz off. This is true in all areas.
Should drivers be tested with a common procedure - heck yes! Governments need to clean this industry up - big time. We have proved we cannot police ourselves.
Should drivers be tested with a common procedure - heck yes! Governments need to clean this industry up - big time. We have proved we cannot police ourselves.
Suggestion, we create a thread with impedance measurements including the way done such as tools, testrig incl Rsense, temperature . To be expander with freq responses when available.
That is not a bad idea, although I think it would fit into this thread.
If you want to start a new thread, feel free. I won't be the one to do it, I am not a expert on this... I just use DATSv3...
If you want to start a new thread, feel free. I won't be the one to do it, I am not a expert on this... I just use DATSv3...
I was hoping to hear from someone who uses a USB mic, so they could explain the methods and techniques they use to get around the single-channel limitations. I am also hoping that if/when they do contribute their thoughts, they do not face a bunch of ridicule and condescending lectures regarding their choice of using a USB mic...
A USB mic has just a build in USB Interface.
Just use your trusted USB interface for the reference channel.
(I use two USB interfaces on one computer to perform potential free measurements. That is the same situation: Two channels, two USB-ports)
Best regards
Bernd
There are two problems with two separate interfaces.
- they don't share the same clock, which can lead to issues
- Windows doesn't always sync nice with different interfaces.
This is in general, your mileage may vary and in some occasions it will just work fine.
But it wouldn't be my default answer and solution for people.
Just get a simple 2-channel audio interface, they are dirt cheap these days.
You can even find them for almost nothing on the 2nd hand market.
(I recently got a EMU Tracker Pre for 20 bucks)
Upgrade path to a better microphone is also much better.
Btw, these days there are also inline XLR interfaces, that can make any microphone into a USB microphone.
They unfortunately have the same problem.
I am glad you said that... I was thinking that might be true, but was not certain...
yeah, although like I said, sometimes it works without any problems.
In my experience it's a bit a hit or miss, so just be aware would be my advice.
I personally don't like such "weird problems" for my measuring devices.
Especially for beginners it can make them instantly doubt they are doing everything wrong etc.
Oh my gosh. So, is the final story that we should still stick with analog (XLR) microphones for complete driver measurement? But it seems like a lot of people are satisfied with USB microphones, even for phase linearization and such.