Obtaining TSP etc with limited data

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
I was intrigued by sannax's wishing to try & design a horn http://www.diyaudio.com/forums/subwoofers/119854-hornresp-238.html for which only a limited amount of data is published. dirkwright uploaded specs for the B&C DE10 namely -

Throat Diameter 25 mm (1 in)
Nominal Impedance 8 ohm
Minimum Impedance 6.3 ohm
Nominal Power Handling (AES) 20 W
Sensitivity (1W/1m) 107 dB
Frequency Range 1.5 - 18 kHz
Recommended Crossover 2.5 kHz
Voice Coil Diameter 25 mm (1 in)
Inductance 0.1 mH
Flux Density 1.55 T

I downloaded the PDF & from looking at the fr/impedence graph i initially estimated fs @ around 1800 Hz to start working with.

I had an intuitive feeling it "might" be possible to determine more data by experimenting with one or more design programes ;) Well to my great surprise & joy, i was able to deduce a lot more info than i initially hoped for :) Using a combination of Horn Response & WinISD i came up with the following.

fs = 1907 Hz

Qes = 0.99

Qms = 0.1

Qts = 0.091

Vas = 0.05 L

Rms = 4.1

Sd = 13 cm2

Cms = 0.0002075 m/N

Mmd = 0.1 gm

BL = 1.6 Tm

Dd = 0.041 m

A few of the numbers were in slight disagreement between programs, but Not far too off to make much of a difference IMO.

Vd & Mms were elusive, but wouldn't prevent a design taking place. I "presume" these could be back engineered from some of the other data ?

From my above tests, it seems that this method of estabilsing data is not restricted to Bass/Mid drivers, but across the full range :)

Hopefully some of you might be encouraged to try it for yourselves with the same driver, and/or others, & post your findings :)
 
Corrections

Originally posted by weltersys

Sd = 13 cm2, from a 25mm voice coil?
If the diaphragm is a dome, it should be only around 4.9 cm2, how did you arrive at Sd = 13 cm2?

In my excitement, i slipped up = my bad :( Thanks for pointing it out :) I've now posted the corrected data.

Interesting that the voice coil diameter is the same as the horn exit.

That's what B&C state in their specs though !

Corrected DATA

fs = 1907 Hz

Qes = 0.99

Qms = 0.1

Qts = 0.091

Vas = 0.05 L

Rms = 0.57

Sd = 4.9 cm2

Cms = 0.0014680 m/N

Mmd = 0.1 gm

BL = 1.6 Tm

Dd = 25 mm
 
Is Mmd the moving weight of diaphragm, or diaphragm and coil?

How do you arrive at those weight without weighing them?

Why is the diaphragm weight still .1gm after it has shrunk from 13 cm2 to 4.9 cm2 ;^) ?

With no Xmax figures, how can you determine LF output capability?
 
usually the dome is 1.5"or 2" in case of a 1"driver
That is often the case for 1" exit drivers, but a 25mm (1inch) voice coil would have only a tiny 1 inch dome.

Alternatively, a 25mm voice coil could be connected to a ring radiator, with a diaphragm only a fraction of the area (and weight) of a 25mm dome shaped diaphragm.

From the posted specs, it is not possible to know the actual diaphragm Sd or weight as the diaphragm material is not even listed, phenolic, polyester, aluminum, titanium, or unobtanium all have different weights.

But I'm sure the TS parameters Zero D listed are "close enough for Rock & Roll" :D
 
Originally posted by weltersys

Is Mmd the moving weight of diaphragm, or diaphragm and coil?

Mmd = The mass of both

How do you arrive at those weight without weighing them?

I saw Mmd in Horn Response after inputting all the other specs, but i agree it now doesn't appear right.

I've posted some equations i pulled from a PDF "DERIVATION OF MOVING-COIL LOUDSPEAKER PARAMETERS USING PLANE WAVE TUBE TECHNIQUES by Brian E. Anderson" Please see my screenie. It "seems" as if at least Mms "might" be able to be arrived at by manipulating the data in there ?

Why is the diaphragm weight still .1gm after it has shrunk from 13 cm2 to 4.9 cm2 ;^) ?

Good question ?

With no Xmax figures, how can you determine LF output capability?

I havn't.

From the posted specs, it is not possible to know the actual diaphragm Sd

In post # 2 you said

If the diaphragm is a dome, it should be only around 4.9 cm2

So i went with that ;)

the diaphragm material is not even listed

It's listed as Polyester in the PDF & here http://www.diyaudio.com/forums/attachments/subwoofers/247414d1320170102-hornresp-b-c-de10-specs.jpg

But I'm sure the TS parameters Zero D listed are "close enough for Rock & Roll"

Cheeky :D

I fully accept that more work is required, but at least it's a start ;)

I'd be interested to see if anyone can produce realistic data from mimimal specs, in a similar fashion :)
 

Attachments

  • Mms.gif
    Mms.gif
    4.7 KB · Views: 89
AkaBak etc

I found that one of AkaBak's useful features is the ability to get Mms & Cms & Vas from just a few numbers, as shown in my screenie. Inputting fs = 1630 Hz instead of 1930 Hz & Re = 5.5 instead of the PDF quoted spec of 6.3 seemed to better correlate with some of the other data gained via the other 2 programs.

Looking at the impedance graph in the PDF shows the DCR/Re is 5.5, not 6.3 !

Work in progress, but getting there, hopefully ;)
 

Attachments

  • ak tsp 1.gif
    ak tsp 1.gif
    12 KB · Views: 82
  • 5.5.gif
    5.5.gif
    36.9 KB · Views: 76
I found that one of AkaBak's useful features is the ability to get Mms & Cms & Vas from just a few numbers, as shown in my screenie. Inputting fs = 1630 Hz instead of 1930 Hz & Re = 5.5 instead of the PDF quoted spec of 6.3 seemed to better correlate with some of the other data gained via the other 2 programs.

Looking at the impedance graph in the PDF shows the DCR/Re is 5.5, not 6.3 !

Work in progress, but getting there, hopefully ;)

The spec sheet PDF saying the minimum impedance is 6.3 ohms appears to be the minimum in the usable pass band, as you say the graph clearly shows the LF impedance as 5.5 ohms.
The sensitivity rating, frequency response, and impedance graph don’t indicate the horn used, oh well...

What are you trying to accomplish with the work in progress?

And why are we writing about tweeters in the Subwoofer forum:p?
 

Attachments

  • BC DE10.jpg
    BC DE10.jpg
    87.1 KB · Views: 78
Originally Posted by weltersys

What are you trying to accomplish with the work in progress?

To try & see if it's possible to get complete, or as near as, TSP & other data from driver specs with limited amount of published info. Not all manufacturers seem to be as helpful as others, or have gone out of business etc. Plus some people might find it useful to evaluate older drivers, without physically having them. Then it would be possible to back engineer them & plug the data into software for design comparisons etc.

And why are we writing about tweeters in the Subwoofer forum?

:D I just used that driver as an example of something with not much published data. & it's what initially aroused my interest in attempting to try & achieve this.

It might be enlightening to take any driver even with full specs & just use a few of them, & plug those into the named programs, and/or others, in various different combinations, & see how much extra useful data we can actually acrue, without the full info ;)

Personally i think it's interesting that "seemingly" so much extra data can be got from initially not a lot Also previously myself, & maybe others, would have bypased certain "possibly" good drivers due to the lack of extended data. If this can now be got via this method = :)
 
To try & see if it's possible to get complete, or as near as, TSP & other data from driver specs with limited amount of published info. Not all manufacturers seem to be as helpful as others, or have gone out of business etc. Plus some people might find it useful to evaluate older drivers, without physically having them. Then it would be possible to back engineer them & plug the data into software for design comparisons etc.

:D I just used that driver as an example of something with not much published data. & it's what initially aroused my interest in attempting to try & achieve this.

It might be enlightening to take any driver even with full specs & just use a few of them, & plug those into the named programs, and/or others, in various different combinations, & see how much extra useful data we can actually acrue, without the full info ;)

Personally i think it's interesting that "seemingly" so much extra data can be got from initially not a lot Also previously myself, & maybe others, would have bypased certain "possibly" good drivers due to the lack of extended data. If this can now be got via this method = :)
If you try taking only the parameters you had available for the HF driver, you will find the extrapolated results fairly useless.

As an example, the Electro Voice EVM 15L and 15B use the same motor, voice coil, cone diameter and both have an Fs of 43 Hz.

Your extrapolation would say they have the same TS parameters, yet Qes and Qms are quite different for both speakers.

The only difference between the two drivers is a slight difference in the weight and shape of the cones, and even though on paper their LF response looks quite similar in the same size BR box, they sound totally different, especially when pushed hard.
Put them in horn cabinets, and the difference is huge.
 
Originally Posted by sreten

Not a chance in hell the Qms number is correct, consequently also Qts.

Hi, yeah i now realise "something" is not quite right, but now we may have one of the reasons !

Originally Posted by David McBean Post # 2399 http://www.diyaudio.com/forums/subwoofers/119854-hornresp-160.html#post2769487

Input values in Hornresp are limited to eight characters. To display 0.0000768 would require nine characters - using scientific notation, only eight are required

Bingo, that "could" be why ? I'll retry & see how it pans out now, but due to the 8 limitation entry it'll have to be with a different driver. Of course i realise that it's not an HR limitation per se, as HR isn't designed to model drivers such as that, AFAIK.

Originally Posted by weltersys

If you try taking only the parameters you had available for the HF driver, you will find the extrapolated results fairly useless.

Your extrapolation would say they have the same TS parameters, yet Qes and Qms are quite different for both speakers.

It does appear that way now :( But based on the above, that "might" help to assist in some way/s :) We'll see ;)

Regards
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.