FE127E Fs measured at 93 Hz. All T/S parameters high. Are these results feasible?

[arjscott]

Hi all,

I'm learning my way around Speaker Workshop at the moment. I have a suitable jig, have been through the calibration process a few times, and can measure passive components quite accurately.

The attached picture shows the calculated T/S parameters and impedance curve for one of my FE127E drivers. My other FE127E measures almost the same as this. The drivers are about 1 year old - very well broken in.

The Qms, Qes, and Qts are much higher than the FE127E datasheet, and the Fs appears to be 93.1 Hz. I've read that individual drivers are likely to vary from the specifications a bit, but these values seem waaay too high.

Using the measured parameters, it's just impossible to simulate a descent looking bass reflex box. The box recommended by Fostex looks horrible.

I've measured two other drivers (not FE127E), and the results are much closer to the specifications. So, I don't think there's anything wrong with my setup.

I have also tested the impedance using Arta with the same results.

The T/S parameters were calculated using the frequency range from 30 Hz to 20K Hz, so the invalid readings at the lower end did not affect the results.

I'd really appreciate thoughts on this.

Andrew.

[Svein_B]

It seems that others have similar results for of the 4" Fostex.

Here are some measurements of the FE126E: http://www.troelsgravesen.dk/FE126E.htm

SveinB

[arjscott]

Hi SveinB,

Thanks for the link. It's an interesting project, and I guess I can have some confidence in my measurement results.

Can anyone offer some advice from this point? Perhaps I can:

Quit annoying people with measurements, just go and build a Fonken / MLTL.

Get some design ideas from more experienced members based on my measurements.

Get some ideas as to why the parameters are so different from Fostex's spec.

???


Troels Gravesen was able to create suitable bass reflex box for the FE126E based on his measurement results, but I can't simulate anything that looks good for my FE127E.

I know all the Planet-10 hosted FE127E designs are well regarded. But the purpose of my exercise is to design something from scratch based on what I've measured.

Andrew.

[chrisb]

FWIW, I know that DaveD has measured and treated several batch lots of more than 20pairs of FE126 & FE127 supplied by the US distributor, and has been quite impressed by the tightness of QC. He may even have graphics of some of these measurements handy,

There must be something strangely wrong with your pair.

As for building to your measurements, as a huge fan of the Fonken design, I'd suggest if you have the time, building a pair of them and your own design, then publishing your measured results and listening impressions of each

[gmilitano]

I measured the FE127E a while back and the Fs was in the mid 80s. At the time, I figured that I must have been doing something wrong since Fostex quotes about 70Hz.

However, it seems that more and more you see measurements which show numbers in the 85-95Hz range.

It would be interesting to see more measurements.

[arjscott]

Hi guys,

Thanks for the comments. Yes, it would be a good idea to build a Fonken with these strangely-spec'd drivers. To be honest the complexity of that box scares me a little, but maybe I can do it.

It would be really good if DaveD or anyone else who was extremely familiar with these drivers could share their measurement results.

Planet10? I respectfully request your assitance

[Svein_B]

I just made a quick measurement of two FE126e.
The results were approximately Fs=83, Qts=3.2. This is a little closer to Fostex values than some other measurements.

I do not know if the measurements are to be trusted however since it is my first attempt to measure speaker parameters with ARTA/Limp, and not using my best PC either.

The strange thing however is that one driver is new, and the other is well broken in and has the dustcap removed and phaseplug installed. They both measured about the same !

SveinB

[vinylkid58]

Originally posted by arjscott
Yes, it would be a good idea to build a Fonken with these strangely-spec'd drivers. To be honest the complexity of that box scares me a little, but maybe I can do it.

Don't be intimidated by the Fonken box, It's well worth the effort IMHO.

Jeff

[MarkMcK]

Anyone know the Chinese factory that makes Fostex? I don't. I do know of another factory in China where the tolerance for Fs is plus or minus 12 Hz. I also know that that factory rarely produces smaller diameter transducers that meet that rather large tolerance. Fs is almost always higher than claimed.

Regardless, I do maintain a "library" of transducers that include a 126E and 127E. I have attached an image showing imped and basic T/S specs. The technique uses a slow sweep sine wave and is a good check of the LMS test that started this thread. I shall leave the interpretation to the reader.

[planet10]

I would suggest that there is no problem with the measurements.

T/S parameters are not single numbers, they are a curve that is a function of voltage drive. With a computer based T/S measing kit, you should expect higher Qt & Fs than if you manually measured them with setup capable of, say, 1 V of drive (that takes quite a robust amp to maintain 1 V at low frequencies).

This also illustrates that it is important to design a box that is tolerant of the changing T/S parameters as you play the speakers louder or softer. Most BR boxes need not apply, as they will work really well at a narrow range of levels (you will find discussion of this on the Basslist and Dan Wiggins made some enlightening posts here on the varability of T/S parameters).

The FE126/FE127s i have tested have had pretty tight tolerences, With drive from my G4 mirror door, i get an Fs of 96 Hz +/- 2.8% for the FE127 & 96 Hz +/- 3.7% for the FE126 (fresh out of the box). (N=20). If i take the same driver and test manually numbers are closer to the Fostex data

I have another batch of each, and they have 100 hrs on them before testing. I'll let you know how they turn out.

dave