Go Back   Home > Forums > >
Home Forums Rules Articles diyAudio Store Blogs Gallery Wiki Register Donations FAQ Calendar Search Today's Posts Mark Forums Read

Planars & Exotics ESL's, planars, and alternative technologies

Help me pick the best drivers for my hybrid esl?
Help me pick the best drivers for my hybrid esl?
Please consider donating to help us continue to serve you.

Ads on/off / Custom Title / More PMs / More album space / Advanced printing & mass image saving
Thread Tools Search this Thread
Old 19th February 2012, 07:39 PM   #11
geraldfryjr is offline geraldfryjr  United States
diyAudio Member
Join Date: Feb 2008
Location: Jackson,michigan
Check these out I didn't write these as I had found most of them here in DIYAudio.
These should be of great use to you as they are to me.

Sorry it took awhile as I had been up for nearly 30 hours and fell out. he,he,he

enjoy !!!

Attached Files
File Type: zip dipole xls's.zip (816.1 KB, 19 views)
  Reply With Quote
Old 19th February 2012, 08:47 PM   #12
geraldfryjr is offline geraldfryjr  United States
diyAudio Member
Join Date: Feb 2008
Location: Jackson,michigan
As I start to remember correctly I believe the Qts of the driver becomes the Qts (Q of the system)of a open baffle, like the Total Qts is in a closed box design.

So, yes, you may want to be looking for a driver that has a Qts in the .5 to .9 range for the flattest low end response curve.
But being as a dipole you have to add a filter any how to get a flat response so this is where driver Q's can get to be very subjective in preference.

The Qms is mainly related to its Fs and the higher this is the less power it needs at this frequency, thus the reason they run out of Xmax at such a low amount of power at in the range of Fs.

You will see in the spreadsheets the relationships in driver displacement vs frequency and Spl.
Keep a note on these when you try different drivers in Winlsd.
You will find the the Spl is soley dependent on driver displacement.

The shape of the response curve and power curve (requirement) are relevant to the drivers parameters in relation to the box parameters.

Another words any driver with X diameter will produce Z amount of Spl with Y amount of Xmax as any other driver with exactly the same diameter or surface area.

It is its parameters (in conjunction with the boxes parameters) that determines how much power (its effects and/of curves) it takes to get there(Spl), not to mention Xmax of course.

In the dipole simulation spl_max1 it takes 8 X 8" drivers at 8mm Xmax to produce about 94db at 35Hz,103db at 50Hz,112db at 71Hz and 121db at 100Hz.
This is with an effective baffle distance (width, or front to back distance) of 400mm.
Remember this is at a full 8mm of Xmax for all of the frequency's.

But at 2mm Xmax it is still a respectable 82,91,100 and 109 for the same frequency's.
Double the amount of drivers to 16 (8 per side) and add 6db to those figures.

This is where displacement rules when it comes to bass.
Put those things in a box and then you don't have bass cancellation issues.

I am pretty sure that dipoles will give the lowest THD figures though.
I have found this out from some quick tests and past experience with my 6 X 6.5" speaker as this is why they are so well found of.

I am not sure but I think the smaller the box also raises the THD as well.
I had started to study this and that was when I found out how badly my .67 Cu. ft. boxes profusely vibrated and messed up my data.
But it did show that the open speaker had the least amount of THD because there wasn't in thing restricting its movement.

This is somewhat well known of course.

  Reply With Quote
Old 23rd February 2012, 08:02 AM   #13
geraldfryjr is offline geraldfryjr  United States
diyAudio Member
Join Date: Feb 2008
Location: Jackson,michigan
I was messin' around with Winlsd today and decided to compare it to some real world test of my RS 8" subs.

First I did some tests on the boxes that I had made for them and as expected I had terrible results.
So I won't bother posting those curves until maybe at a later time just to show the effects of the rise of THD from using a much smaller box.
The THD was quite high and the boxes rattled very badly as they did the last time I tried to test them.
But they did very well for there intended application and that was behind the backseat of my 84 Camaro where there was lots of cabin gain back in 2004.

In these tests I choose to use a open free air method.
These where done in my listening room and if the weather was better I would have done these outside to eliminate any effects from the room.
But they weren't loud enough to have any serious results in the data anyhow.

I used HolmImpulse and REW V5 for these tests with a Dayton measurement Microphone.

The HolmImpulse program shows the THD and frequency response curves at two distances.
One flush to the driver and one at a distance of about .5 meters ( I think as I forgot to note this and the charts show that it may have been closer).

It is sometimes hard to juggle the calibration between the the two programs as they are different but the data usually comes out the same between them.

The REW V5 plots were calibrated to my SPL meter and four measurements were taken with one flush to the driver, and, the rest about 12", .5 meter and 1 meter away from the driver.

Once I had gotten the graph limits set properly I had printed them all out to compare. I was quite amazed to find that when I superimposed one on top the the Winlsd simulation I had found that it was very spot on!
With very little differences in the curve.
The 20Hz point and the 60hz point nearly exactly lined up within only a few percent's of difference.

The only discrepancy in comparing the data was that with the simulation, as you decrease the box size past where the curve doesn't change any more the Spl starts to drop.
So, I readjusted the signal level to 30watts in the simulation to compensate for this and to match the REW V5 plot at 100db spl.

In the real world test I had measured 1.9Vrms and for the 4 ohm driver this is .9 watts rms.
As I had mentioned that the db scale in the REW V5 plots were calibrated so they show the true real world Spl of the tests.

Thd curves show to be pretty decent and If had done this outside like I wanted to that measurement might have come out a lot better.
Also I didn't have a scope hooked up to check for any clipping or deformities of the waveform at the lowest frequency's.
I will do this at a later time.

Generally I can't hear much of difference of the 10% THD that low.
But, It is good to be able to measure it though any how!
What I can hear is the huffing of the air as it moves through the vent holes and this can be annoying at times but generally not heard when playing music as it is only noticeable at about 30Hz to 20Hz and below.

enjoy !!!

Attached Images
File Type: jpg rs sub open air only.jpg (157.2 KB, 136 views)
File Type: jpg rs sub open air.jpg (72.0 KB, 136 views)
File Type: jpg rs open air distance.jpg (72.0 KB, 135 views)
File Type: jpg rs open air distance rewv5 hr.jpg (74.9 KB, 134 views)
File Type: jpg rs sub open air only holmipulse.jpg (95.6 KB, 134 views)

Last edited by geraldfryjr; 23rd February 2012 at 08:17 AM.
  Reply With Quote
Old 23rd February 2012, 09:16 AM   #14
geraldfryjr is offline geraldfryjr  United States
diyAudio Member
Join Date: Feb 2008
Location: Jackson,michigan
So now that I have convinced my self of the accuracy of Winlsd I did a few more simulations using the MCM 8" sub compared to my RS 8" sub.

Even though these were meant to be for a dipole configuration Winlsd still depicts them as a closed box.

The results were very impressive besides.

The first plot was each type of single driver at one watt of signal input.

The second plot is the same but for a maximum Xmax setting at 20hz about 22 watts of signal input.

The third is for 6 drivers with xmax of 8.5mm at 20hz at 140 watts input.

And, The fourth is for 8 drivers at an xmax of 9.7mm at 20hz (8mm at 30hz) with 250 watts rms input.

This shows why I like these drivers for the money.
But, I have not tried one yet so I would get a few of them to try out first before getting a bunch of them.
But so far the do look very promising especially for a closed box type system if they turn out to be to noisey for a dipole setup.

Remember to added another 6db to this for stereo and they still should only be -6db shy from the maximum in these plots and that would be about 110db Spl.
Not to mention that room gain may make up the difference as well.

Knowing what my past experience was with my 6 X 6.5" open back cabinet, 12 to 16 of these drivers will fill a room quite nicely even in a dipole setup or open back line array.

I did a little calculation using surface area alone using my system, and my system is about the equivelant to 15 X 8" drivers and my Spl's seem to coincide with the data in the simulations.
My 10" and 12" drivers don't have anywhere the xmax that these things have.

When I got my drivers I almost got 12 to 16 of them for the soul purpose of trying a dipole system.
But I ended up with 8 and I burned one up by accident last year trying to test it, and, another one I dropped the day a brought them home trying to carry the stack in.
It had bent the frame and the voice coil is rubbing on the magnet,it sucks but things happen.
There is a chance that I could fix it but I am now down one driver so I now have six.

After todays measurements I am seriously think of just sticking with a dipole setup with my remaining six 8"er's and use them nearfield in front of my mixer just because of the lower THD and sound quality.
As my main system is for completely rockin out and it does that job very well indeed !!
At least they will hold me over until I can get some newer and higher quality drivers.

I will do some checking in the dipole spread sheets as well as comparing to the peerless driver.

Peerless drivers are very good quality and so are the Daytons so I will take a closer look at those too.

More later !!!


P.S. I did not use any Linkwitz Transform's on any of these simulations as it would only be useful at moderate listening levels, and, If it was to be used I would make it an adjustable boost and cut type in order to be able to taylor the system to the room and the type of material that is being listened to at the time,IMHO.
Attached Images
File Type: jpg MCM vs RS.jpg (95.9 KB, 23 views)
File Type: jpg MCM vs RS SPL 22WATTS.jpg (99.3 KB, 15 views)
File Type: jpg MCM vs RS X6 SPL 140WATTS XMAX 8.5mm.jpg (96.0 KB, 13 views)
File Type: jpg MCM vs RS X8 SPL 250WATTS XMAX 9.7mm.jpg (98.3 KB, 13 views)

Last edited by geraldfryjr; 23rd February 2012 at 09:29 AM.
  Reply With Quote
Old 24th February 2012, 05:26 AM   #15
markusA is offline markusA  Sweden
diyAudio Member
markusA's Avatar
Join Date: Dec 2009
Location: Gothenburg
I must have started writing a reply something like half a dozen times but I get interupted or there's not enough time.
This time I'll just write something short to tell you that I do read and appreciate what you write. I just haven't had the time to write a reply...

At the moment I'm leaning towards 6―" drivers, 8 if possible but you never know...
Ok, time caught up with me again and I have to run. I'll be back and write some more later...
  Reply With Quote
Old 24th February 2012, 06:13 AM   #16
geraldfryjr is offline geraldfryjr  United States
diyAudio Member
Join Date: Feb 2008
Location: Jackson,michigan
Thats cool, markusA,

I had been doing some more more sims today and I had some very interesting results.
I won't go into great detail,but here is what I have.

My Optimus driver has nearly the exact same specs as the peerless driver only has a 4 ohm voice coil and 2mm more of xmax.
And it has nearly the same curve as well with the difference in level showing due to it being a 4 ohm drive instead of 8 ohms.

The daytons look very good as well.
In all of these sims I used a 550 cu.ft. box and set the power levels for a 8 mm xmax at 20hz for 6 drivers.

The all look very good But to me the MCM driver seemed to be the best one for the money and had the highest output as well as a smoother low end roll off.

One of the dayton's only has an xmax of 6 mm but I set them all to 8mm's in the sim.

Then I got to thinking how the peerless even looks the same as the RS sub except for the gold spring loaded push connectors that are on the RS sub.

So, I did a free air THD test at three levels.
I didn't note any voltage levels but the highest one was near its xmax if not more.
Now they may look kind high but in reality the aren't bad at all.
The curves are the bottom of the HOLMimpulse chart are the THD curves and are in db's so here is a link to convert them,

THD to dB - convert percent to decibels dB and dB decibels to percentage percent voltage % vs per cent converter THD+N distortions calculation signal distortion factor attenuation in dB to distortion factor k in percent decibel damping - sengpielaudi

The sad thing is that you can never get those figures any where unless somebody tests them and posts them and this is where the risk of buying drivers comes in.
Zaph Audio has done this for a few of the well known drivers and a quite few from dayton as well.

If the manufacture's stated their THD's in there data sheets they probably wouldn't sell any drivers at all. he,he,he.

If the peerless is any thing like the RS driver then I would say that it is a little pricey.
These are the highest xmax drivers for an 8" that I could find.
The next possible step are the Bazooka 10"'s that parts express has for a very nice price.

And then there is the 12's but the displacement per cost factor now starts to sky rocket as I am assuming that you want to go with a dipole bass system.

I have not yet tried a TL type enclosure yet, but I plan to do so this summer as well with my 8's just to see if it is really all that people say that they are.

Anyway here are todays charts and I will start scoping out some 6.5"er's.

Attached Images
File Type: jpg EXCURSION.jpg (113.1 KB, 13 views)
File Type: jpg GROUP DELAY.jpg (110.6 KB, 7 views)
File Type: jpg IMPEDANCE.jpg (116.4 KB, 6 views)
File Type: jpg TRANSFER FUNCTION MAGNITUDE.jpg (117.0 KB, 6 views)
File Type: jpg SPL.jpg (119.3 KB, 7 views)
File Type: jpg RS SUB THD 3 LEVELS.jpg (116.4 KB, 14 views)
  Reply With Quote
Old 24th February 2012, 06:05 PM   #17
Calvin is offline Calvin  Germany
diyAudio Member
Calvin's Avatar
Join Date: Nov 2004
Location: close to Basel

Not to mention that room gain may make up the difference as well.
Thereīs no room gain for dipoles. Instead- in theory- they donīt have any output at all below the lowest room mode. Fortunately a real dipole in a room doesnīt completely behave as a theoretical dipole.

  Reply With Quote
Old 24th February 2012, 09:34 PM   #18
geraldfryjr is offline geraldfryjr  United States
diyAudio Member
Join Date: Feb 2008
Location: Jackson,michigan
As I was going over the Linkwitz web site again I was starting to wonder about this.

The lowest lowest room mode frequency of my room is 30hz as it is 18.5 feet long and this make sense looking over the plots of my system.

I don't have a dipole system setup yet but I am thinking that it will take alot more drivers to get the kind of performance that I am now getting with 8 in semi ported boxes.

I say semi ported boxes because they are run of the mill JVC ,Technics and some other brand that I had reglued the cabinets and stuffed them generously with of fiberglass insulation.
As well as the port tubes removed to reduce the boomyness but the holes are still there on four of them and it makes only a little difference if a close them.
I may try this at a later date but I am very happy the way they are now.
They are stuffed in the corners and what I do get is a nice tight non boomy low THD sound that is flat to 25Hz at about 109db 14 feet away .

So a dipole system is going to be some thing very new to figure out.

I when I did have my 6 X 6.5" open back cabinet it was in a room that was 23.5 feet long and sat 4 feet up off of the floor.
This helped couple into the room very well and it sounded like crap when it was sitting on the floor as expected.

However that cabinet was in a 3 X 2 array and I am not sure that 6.5" drivers would create enough displacement (partly due to the width of the structure) to preform well above 105db ,or so , at very low frequency's.
They Just don't have enough Xmax to do it or displacement in general.

Even in a closed box it takes a minimum of 8 of the $8 6.5" drivers to get to 105db while reaching their Xmax at 30hz and that is for one channel.

So maybe I am wrong, But, you would be just barely getting 100db for a total 16 drivers in stereo from 30Hz on up, may be even less.

In the same situation it takes only 4 of the MCM 8" drivers in a 4 cu.ft. box to match the performance of 8 X 6.5" drivers in a 11 cu.ft box.

I know we are talking about dipoles here, So ,using those spreadsheets that I posted shows you how much displacement is needed to reach your goals.

But, I did find another killer deal!
Parts express has some cheaper 10" woofers for $5 and with 6 of them in a 6.5Cu.ft. box Winlsd shows it to produce 109.8 max flat from-3db at 20Hz using a Linkwitz filter and the Xmax set a 4.5mm at 30Hz and only 8 watts!

They will do even more if you don't use the filter for bass extension and may even be a great candidate for a large line source dipole system,I know the price is just right for sure !!

It is here,

10" Ribbed Paper Cone Woofer Speaker 299-284

I just noticed this one too,

10" Ribbed Treated Paper Cone Woofer Speaker 299-283

The Qts looks rather high on this one but I will model it and see how it comes out.

This is very impressive and talk about some Spl's on the cheap !!!
This is the best deal I have ever seen!

  Reply With Quote
Old 25th February 2012, 07:26 AM   #19
Calvin is offline Calvin  Germany
diyAudio Member
Calvin's Avatar
Join Date: Nov 2004
Location: close to Basel

if the parameters of the second driver ī10" Ribbed Treated Paper Cone Woofer Speaker 299-283ī are just halfways correct, itīd be useless crap.
It should be forbidden to waste mother earths precious resources by producing something like that.

The problem of the dipole is twofold.
- No roomgain, as such no output below lowest roommode. In practise, due to non-theoretical behaviour thereīs still some output, but considerably less than with a monopole.
- Lower bandwidth limit defined by room dimensions. The larger the room the lower f-3dB. On the other hand require big rooms very large membrane areas.

Iīd recommend a pure dipole bass just for small rooms. In larger rooms Iīd always add a dedicated subwoofer, restricted to very low frequencies, preferrably below the lowest room mode. If restricted to a F-range below the lowest room mode, the sub pressurizes the room, working very efficiently, taking full advantage of room gain, even more so when positioned in a corner.
The sub may then be built quite compact. Because the Sub doesnīt excite room modes the performance is free from booming.
If the dipole is used above say 50Hz, it works for most in a frequency range where itīs efficiency is higher than that of a monopole and the dynamic range is at least on par. The requirements regarding membrane area and excursion reduce to a very praticable level here. The dipole -and even more so a dipole line array- than can fully profit from their superior behaviour in the bass and kickbass range, where they perform highly precise, boom-free and dynamic.


Last edited by Calvin; 25th February 2012 at 07:29 AM.
  Reply With Quote
Old 25th February 2012, 08:07 AM   #20
markusA is offline markusA  Sweden
diyAudio Member
markusA's Avatar
Join Date: Dec 2009
Location: Gothenburg
That's what I'm going for.
Stand alone sub for the lowest octave or two. Dipole line array for the bass/mid bass and the panel takes over somewhere in the 250-300Hz area.

The panel is roughly 9" x 51".
I have perf metal for both a curved and a flat set. (And a couple of small extras for initial experiments)
Work table/stretch jig to be completed next week.

I have realized something a little disturbing.
I want to build the line array with 8 drivers but doing so will make it 63" tall, not a elegant match for the 51" ESL panel. :P
6 drivers will fit nicely in a 51" or so cabinet.
Maybe I'll order 16 drivers, build with 12 and keep the 4 extra ones for a later build if I want to experiment with a 63" panel?
Maybe I'm just being stupid and should just suck it up and realize 6 drivers per side will be plenty?

Last edited by markusA; 25th February 2012 at 08:23 AM.
  Reply With Quote


Help me pick the best drivers for my hybrid esl?Hide this!Advertise here!
Thread Tools Search this Thread
Search this Thread:

Advanced Search

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off

Forum Jump

Similar Threads
Thread Thread Starter Forum Replies Last Post
Final 0.3 Hybrid ESL TimA Planars & Exotics 36 7th November 2012 01:26 PM
Help with esl hybrid electros Planars & Exotics 8 5th December 2010 11:02 AM
New hybrid ESL Build Freddy G Planars & Exotics 36 30th August 2010 06:25 PM
Geert - Hybrid ESL GeSo Planars & Exotics 21 24th September 2008 08:08 AM

New To Site? Need Help?

All times are GMT. The time now is 07:38 PM.

Search Engine Optimisation provided by DragonByte SEO (Pro) - vBulletin Mods & Addons Copyright © 2018 DragonByte Technologies Ltd.
Resources saved on this page: MySQL 15.00%
vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2018 DragonByte Technologies Ltd.
Copyright ©1999-2018 diyAudio