Inconsistencies in driver specifications

[Gerrit Boers]

I have this crazy idea of building my own fullrange driver. To get a handle on the things involved in driver design I'm studying what's out there. I looked at the specifications for the Lowther drivers and there is something I do not understand:

In 'Testing Loudspeakers' Joe D'Appolito describes two methods for calculating the efficiency of a driver. The first uses fs, Vas and Qes to calculate the efficiency:

N0= 9.6 x 10^-10 ((fs^3 x Vas) / Qes) (equation 2.35)

If you look at the specs for the Lowther DX2, DX3 and DX4 you'll see that the specified fs, Vas and Qes are the same for all three drivers, yet the stated efficiency is not the same.
How can this be?

Next I looked at the Fertin 20EX fieldcoil driver. The stated efficiency is 101dBSPL/W/m (max current in the fieldcoil), the Bl product is 8.7.
Sd=0.02m^2
Re=6.4?
Mms=18g

Using the following equation:

S0= 20log((Bl x Sd)/(Re x Mms))+88.5 (equ. 2.38)

I get 92 dBSPL/ 2.83V / m, quite a difference.

Am I doing something wrong? Or is there something else going on here?

[MJK]

The Lowther specs on the manufacturer's site are not accurate. I have measured a bunch of Lowther drivers and have posted the results on my site. I believe you will have better luck using my results.

[Gerrit Boers]

Thanks for the link Martin.

'Not accurate' seems to be quite an understatement. Lowther specifies an Fs of 36Hz for the DX2, DX3 and DX4. Your results are 56Hz, 60Hz and 59Hz for these three models.

Based on your results I also question the validity of the specified flux density, which was the part that interested me javascript:smilie('')

Is it really that difficult for a manufacturer to provide an accurate and comprehensive set of specifications? If a car's specifications would be this inaccurate all hell would break loose.

[MJK]

Originally posted by Gerrit Boers
Is it really that difficult for a manufacturer to provide an accurate and comprehensive set of specifications? If a car's specifications would be this inaccurate all hell would break loose.

For a car, anybody driving it would know immediately that someting was wrong.

But most DIY speaker builders, and in particular the single driver users, do not have the capability to measure the drivers. I don't mean to pick on single driver users but this subset of speaker builders tend to be more into the esoteric subjective aspects of audio then the multi-way builders who tend to measure and engineer everything about their speakers. So my guess is that most Lowther users do not know, and probably do not care, what the T/S parameters are for their drivers.

Lowthers have been around for a long time but if you search the Internet there are very few (actually close to none) measured T/S parameters or SPL response plots available. Compare that to the latest popular SEAS or Vifa midrange drivers where you can probably find all the measured data you want.

I think the biggest problem Lowther has is the variability of the foam suspensions used in their drivers. It must be extremely difficult to get a repeatable foam spider or surround stiffness from a thin sheet of foam. The moving mass, BL, Re, are reasonably repeatable but the Cms and therefore Vas are diffcult to get matching between two units. But in the end they are great drivers and I enjoy the ones I have very much. I cannot speculate on why Lowther continues to show specs that are very suspect on their web site.

[richie00boy]

n0 is the reference efficiency, which is not the same as the dB/watt efficiency published by manufacturers.

[Scottmoose]

Originally posted by MJK
But in the end they are great drivers and I enjoy the ones I have very much. I cannot speculate on why Lowther continues to show specs that are very suspect on their web site.

Probably because, as you mentioned earlier Martin, many (most?) Lowther owners simply don't care about the T/S parameters of the drivers, apart from possibly Fs. And in my turn, I suspect the 'optimistic' claimed Fs of their drivers is partly responsible for a lot of Lowther cabinet designs appearting to be so mediocre.

[Gerrit Boers]

n0 is the reference efficiency, which is not the same as the dB/watt efficiency published by manufacturers.

I know, but if Fs, Vas and Qes are the same for two drivers, the efficiency must also be the same. BTW, interesting that the efficiency increases with the cube! of Fs.

And in my turn, I suspect the 'optimistic' claimed Fs of their drivers is partly responsible for a lot of Lowther cabinet designs appearting to be so mediocre

Exactly my point, how are you supposed to design a decent enclosure when the stated Fs is so far off?

I guess I was 'lucky' with the Fostex208S, the measured T/S parameters are within 5% of the specs.

[MJK]

I guess I was 'lucky' with the Fostex208S, the measured T/S parameters are within 5% of the specs.

I have found that in general the Fostex manufacturer's stated T/S parameters are reasonably accurate. Lowther and AER are not very accurate. I always run one of my T/S consistency check worksheets to assess a manufacturer's data sheets when considering a driver and it is surprising how many of the premium expensive full range drivers do not have consistent properties. Then if I actually get my hands on a pair of these premium drivers, reality is something totally different which is not such a big surprise. If you design an enclosure using real numbers the odds of coming up with a good design are increased significantly. The down side is that sometimes you have to buy on faith that it will work out in the end, that can get expensive.

[ronc]

think the biggest problem Lowther has is the variability of the foam suspensions used in their drivers. It must be extremely difficult to get a repeatable foam spider or surround stiffness from a thin sheet of foam. The moving mass, BL, Re, are reasonably repeatable but the Cms and therefore Vas are diffcult to get matching between two units.



It must be extremely difficult to get a repeatable foam spider or surround stiffness from a thin sheet of foam

The major problem in manfacturing is repeatability. I have been fighting this for over 40 years. Sometimes there are so many variables that its hard to concieve.
The answer is that consistant base materials have to be established to produce the final product. Manfacturing defects have to be be brought to a minimum to arrive at a consistant product. Control of the manfacturing process has to be established from all aspects. Without control we are little more than any China produced products.

ron

[planet10]

Another issue is that T/S parameters aren't really a single number, but a curve that is a function of voltage drive. The measured parameters of same set of drivers by 2 different people can be quite different. (these numbers are also affected by temperature, barometric pressue and humidity). In every case they should be consistent within the set thou.

The measures i get for instance of the Fostex FE126 & 127 i get are fairly different from the factory but show good consistency.

the sims run for the Fonken (using factory specs) are quite consistent with the measures of the axtual boxes.

Martin, have you measured the FE126s you got from me? It would be interesting to compare your numbers with the ones i have pf the same drivers.

dave