New versions of old classics

[Snickers]

Hi guys!

I am new to this forum, so I hope I do things correctly.

One thing has been on my mind for a while. There are lots of classical and very popular drivers from JBL, Altec Lansing and others. Just to mention a few cone drivers from the top of my head:
JBL E145
JBK 2220
Altec Lansing 515
TAD 1601b

...and there are off course many others.

But what if one should make new drivers that would work as drop in replacement for any of these. What would be important? What really makes those drivers so popular? I assume the fact that they are obsolete gives them a kind of exclusivity, but things like high VAS, strong motors and low moving mass are things you do not see as often today as back in the days.

I am currently working on a 15 inch with parameters similar to the TAD 1601b, and some discussions are going on about what is most important. It is currently at 400 liters VAS and has a motor characteristic and moving parts that gives it the same low end response as the 1601 in bass reflex cabinets. However, it also differs in important areas. For example, instead of an alnico motor, it has a compact neodymium motor. It also has way lower Le, longer x-max, and a 100% symetrical differential drive motor. Off course it also has less progressive suspension due to the longer x-max.

What I want to discuss here is...:
- Why do we like the old classics?
- Which ones of the old classics are really interesting today?
- What would be important parameters to keep if one were to make a new drop in replacement?
- And which parameters should we allow ourselves to change?

I just attach an image to make the post look better

[Mr. Widget]

What I want to discuss here is...:
- Why do we like the old classics?
- Which ones of the old classics are really interesting today?
- What would be important parameters to keep if one were to make a new drop in replacement?
- And which parameters should we allow ourselves to change?

There is a review of the current Klipschorn in this month’s Stereophile. The review praises the speaker’s strengths but also acknowledges its shortcomings. My takeaway from the review is the honest assessment that every speaker is an assortment of compromises. The classic speakers that many of us admire are extremely dynamic... a feature rarely shared by most modern hi-fi speakers today. This is true of drivers and complete systems.


Widget

[Odd]

Welcome to Lansing Heritage Forums Jørn.

"Snickers" is a well-known figure on Norwegian forums and has considerable technical expertise.

He has also held courses on DSP and other HIFI related material. YouTube link

He has now started an ambitious project to create a new bass driver.
More info on new bass driver here. (Content in Norwegian)

[Mr. Widget]

Very cool! Thank you Odd, for the introduction.

Yes, welcome to the Forum, Snickers! Please tell us more.


Widget

[macaroonie]

Welcome Snickers from just across the water. M

[Snickers]

Thanks so much all of you for the warm welcome!

For those who want to follow the driver I am currently working on, I would like to recommend the DIY Audio-thread, as I guess English comes more naturally for most of you guys than Norwegian does. Here is the link:

https://www.diyaudio.com/forums/mult...ch-woofer.html

For those of you familiar with simulations and T/S-parameters, you will probably notice right away that it does not have the exact same parameters as the TAD TL1601b, but the response in the target enclosure (150 liter/24Hz) is as close as you could possibly get it.

If we look at the old JBL tech-stuff from around 1978 (when the crisis in Kongo started, making cobalt an extremely expensive material). They did some work on how to better take advantage of the properties of ferrite. The drivers originally available with both alnico and ferrite were almost always far better with alnico than with ferrite. However, when they seriously undercut the pole piece (getting it pretty much saturated at the bottom) the performance improved even past the performance of alnico drivers.

With modern simulation tools, such as FEMM/COMSOL etc, we know today that this is closely connected with the saturation level in the steel parts, and, even if it could mean being shot dead in a dark back yard to even mention it, alnico by itself is not necessarily better than ferrite if the saturation level in the steel parts is the same. We do not see drivers where this is identical between an alnico version and a ferrite version. Drivers available with both ferrite and alnico are pretty much allways far more saturated in the alnico version, so it should also sound better.

My point is not to discredit alnico in any way, and especially not to discredit any great drivers with alnico. But I would like to point out that, as JBL also found out in the 70's, there are ways to engineer a motor to improve performance significantly, regardless of magnet material. The overall geometry of the motor is really depending on what kind of magnet material you use. In the current driver, I get pretty small steel parts as the N40 magnets gets enough space inside the voice coil, so that I do not have to worry about reaching out to the edge of a 300mm ferrite ring. I do get around 2,2 Tesla in the pole tips. Everything above 1,95T should be regarded saturated for most types of steel, and above 2,1T the B/H-curve is pretty flat.

So with todays technology, we can achieve the alnico benefits, but with a more effective motor geometry.

There are, however, some of the old drivers that uses some interesting and rather odd solutions. Some of the vintage compression drivers do use silver shorting rings, which is quite interesting from a high frequency standpoint. Then you have one driver in the E-series, I do not remember which one, that uses an equal hung motor geometry. That is pretty much useless for a woofer, but as a mid bass horn driver that has probably less than 1mm excursion at max, it could be interesting. I do not suggest making an equal hung motor, but I think making a really short x-max driver (that has an extremely light coil and cone, and insane motor force) could be an interesting thing. The E145 is also popular. But as JBL stated back in the days, they are not for music reproduction, they are for music production. And then, if I got my facts right, they went on to use it in the Everest... Probably due to its lively midrange.


So I'll try to rephrase a bit. Which drivers would you like to see new versions of?

That could potentially form a base for which drivers to discuss. Then we could look more into what makes those drivers special. We could also discuss what should be changed or added. A good friend of mine just replaced his 2226 (i think) with 2220. He is a bit worried about loosing the shorting rings, but really liked the sound of the 2220s. This is just an example of what I mean with "things that should be changed or added". I've got to invite him to this thread. I think he has read all documents in the world that says JBL, Altec Lansing or GPA somewhere on it.

At the moment we are, in a way, discussing all drivers at once. That was probably not the best idea.

[oks81]

Hello!

I have tried to read some JBL yes..but som of the guys in this forum probably do know more than me.
First the equal hight voicecoil/gap 15" driver you mention is the E130.
This is an instrument driver, but can`t see anything wrong with using it for horn and midrange in hifi setup.
As it is equal hight it is very efficient for it`s application, but off course very short stroke.
The specs says Xmax 2,5mm, the VC is then out of the gap so I don`t know how the linearity of this driver is at 2,5mm.

In my setup I hva tested the 2226, 2227 and 2220.
Tried to compare the 2227 and the 2220 in the range of 150-600hz.
The 2220 is maybe the one I prefer. It is a little more efficient, underhang motor, curvelinear cone and quite high Vas.
My 2220 is the H model with ferrite motor. Could be interesting if this driver had some of the features from the 2227 motor, as copper shorting rings.
The 2227 sounds good at the same application, and that tells me that the SVG motor has something to it.
Both the 2227 and 2220 is quite a bit bether than 2226 in the midrange.


Your driver Snickers, is more in the range of Altec/GPA 515, JBL E145, TAD 1601.
Both 515 and E145 has underhang motor.
The TAD probably overhang since higher Xmax.
All three quite high Vas and not so high mms.
Classic drivers used to play all the way from bass and up to a horn in typical two way setup.

The GPA 515 quite expensive these days, E145 is not made anymore, TAD 1601 quite expensive, is it only NOS that is sold at proaudiodesign?
Think your driver should be welcome in the application for bassreflex to play all the way up to 6-800hz.

[oks81]

In the specs of the E145 it has Vas of 275 litres.
This I have read that is wrong, and should be 428 litres.
Had to test this in VituixCad and seems like 428 litres is correct.



Mms is quite low on this driver with 55 grams.
Never heard it but can imagine it sounds nice

[Mr. Widget]

So I'll try to rephrase a bit. Which drivers would you like to see new versions of?

Now that you put it that way...

My favorite classic woofers have been the light coned vintage JBLs. 2220A, K130, K145. The fact that all three are alnico may or may not have been all that important. None are great at deep bass, but they have a “liveliness” that I find very desirable. The first two have curvilinear cones and the K145 is a straight sided cone and all three have very different motor structures.

I haven't really studied the differences, but look forward to your thoughts and discoveries.

Now, that’s woofers... what about HF drivers? My favorites there are all 4” diaphragmed compression drivers with either 2” or 1.5” exits. The models I have owned and liked in no particular order, JBL 2440, 2441, 476Be, and from TAD the TD-4001 and TD-4003.


Widget

[oks81]

Offcourse!
The K-series is the alnico and the E-series is ferrite.
Never stop learning!

[RMC]

Hi oks 81,

RE: post # 16, "I do look at interest at the E145 with its low mms, high vas and lower fs than the 2226, maybe E145 would do a bether job than 2226...?"

Personnaly, I'd be more interested in E145 lower Qts VS 2226, instead of E145 lower Fs than 2226 for what you seem to be looking for. My understanding, right or wrong, is that you may be looking for a more dynamic driver in E145. Dynamic here not equal to absolute driver output but rather one having better transient response. e.g. there's some of these in older JBL drivers, the 2220 you have could be one of them.

The 5 hz difference in Fs between E145/2226 isn't a big gain and possibly not really what would put you ahead in your quest. In my view the E145 lower Qts than 2226 might be more rewarding dynamics wise than the Fs aspect.

As indicated by Bullock, low Qts drivers tend to have better transient response, same for flat enclosure alignments. For the latter, Eargle showing Small's work on this, for sealed and vented, is more than clear. Could be why I really like the sound of my good old 2205H drivers, they're fast/dynamic, even though not up to more recent standards in terms of low-frequency capability and spl output.

BTW since you appear to have an interest in LF driver BL and other TS, plus me having an interest in LF driver Qts and other TS, well here's the relation between these two as mentioned by Eargle (Loudspeaker Handbook, P. 61):

" Figure 4-3 shows a family of curves in which the value of Qts is the only variable. This is roughly equivalent to varying the BL product of the driver. Reducing BL (increasing Qts) diminishes the piston band sensitivity of the system, while allowing the response at system resonance to peak progressively higher, relative to the piston band value." Then he mentions that design trade-offs are made on the basis of T/S analysis. So, the above suggests BL and Qts evolve in opposite directions (roughly), plus incidental effect on sensitivity and system response.

Best regards,

Richard

[Snickers]

Now that you put it that way...

My favorite classic woofers have been the light coned vintage JBLs. 2220A, K130, K145. The fact that all three are alnico may or may not have been all that important. None are great at deep bass, but they have a “liveliness” that I find very desirable. The first two have curvilinear cones and the K145 is a straight sided cone and all three have very different motor structures.

I haven't really studied the differences, but look forward to your thoughts and discoveries.

I do not have all that in depth knowledge about all of these, but I assume the parameters are not too far from the E-series counterparts.

The E130 and K130 are absolutely useless for any kind of bass application. It has to be considered a pure midrange. It should also have relatively low inductance and insanely low mass combined with high efficiency. The high efficiency comes from botn a 100% coil/gap-coverate (100% of the coil covers 100% of the gap) that gives a really strong force factor, and off course, the low moving mass.

The sound of these drivers comes partially from distortion caused by motor induced distortion, but for applications where the excursion is extremely short, it will be heavily influenced by the fact that this driver creates a massive sound pressure with almost no load on the magnetic circuit. It also has a short and lightweight coil, which makes life easier for the cone. Las, but not least, a driver like this can use light suspension.

It is not hard to make a replacement driver for it, but I am not all too sure people would really want it exactly as the original.

The K145 is, IMO, more interesting. It can be used as a woofer, but at the same time, it represents something a bit special for midrange. It is popular for midbass horns too. It too does have a rather limited x-max. Making a new version of it could be interesting, but I would probably rather give it a good 8-10mm x-max than the 5mm of the original. This can be done without compromising on other parameters.

The 2220 is not that different from the K145, but has a slightly stronger motor. I guess a "New version" could cover both of them for most applications. The E145 and the K145 are more different than the K145 and the 2220A, so they are not very far from each other any of these.

Now, that’s woofers... what about HF drivers? My favorites there are all 4” diaphragmed compression drivers with either 2” or 1.5” exits. The models I have owned and liked in no particular order, JBL 2440, 2441, 476Be, and from TAD the TD-4001 and TD-4003.

I am on this, but at the moment, I am concentrating on cone drivers.

The things I have been looking at is a large format 3 inch with 2 inch throat (a bit like the Altec 288, but with a beryllium option) and give it more extension in both ends than the typical 4 inch drivers out there.

I have also been asked for a small driver for super tweeter, like the TAD 703 and others. Maybe as a driver with 0,75 inch throat.

[Robh3606]

The K145 is, IMO, more interesting. It can be used as a woofer, but at the same time, it represents something a bit special for midrange. It is popular for midbass horns too. It too does have a rather limited x-max. Making a new version of it could be interesting, but I would probably rather give it a good 8-10mm x-max than the 5mm of the original.

The K145 and E145 are short coil log gap so the coil is 100% in the gap even at extremes. The E has significantly longer xmax about a 3rd more. You are going to end up with a very thick top plate and end up needing subs anyway if you keep everything else the same.

Rob

[Snickers]

The K145 and E145 are short coil log gap so the coil is 100% in the gap even at extremes. The E has significantly longer xmax about a 3rd more. You are going to end up with a very thick top plate and end up needing subs anyway if you keep everything else the same.

Rob

Yes, they are under hung. I really loved the idea of under hung for some time, but after a while I discovered that the flux is extremely unstable in a long gap. One could make it stable, but for a 20mm tall gap, that requires a lot more magnet than you find in the K or E 145. The problem can be seen two ways. One is that the flux tends to move from the part of the steel with load, to a part of the steel without load. So even if the coil has 100% coverage, the gap has not. Another way to see it, which is kind of different, but at the same time represents the same thing seen from a different perspective, is that having a narrow gap covering the entire coil makes the inductance go fairly high for a coil this size. A lot of steel is available at all times, and it all contributes to slow down current in the coil and convert it to magnetic flux in the steel itself.

We also have a huge benefit using a king of geometry like this, and that is the fact that the coil is light, and you can lower the total moving mass.

However, making a motor with a light coil can be done in several different ways. Some good, and some not so good. Among the not so good solutions are the use of an extremely short gap together with a short coil. This only gives a super low excursion, not very linear motor. Another not so good way is to use a pole piece and/or top plate with a split, and have the coil half way into both gaps. The problem here is that when the coil is situated in one gap, all the flux tend to move to the other gap. One way that do work is to make two magnetic circuits, one for each gap, and position the coil as described above. That way a 10mm winding height can result in an x-max of 6-7mm. One can also do a more complex version with two tiny coils, each measuring just 6mm, and both of them being half way into a 3mm gap with a 3mm space between This gives an incredible 10+mm x-max. One could also make an all magnet motor, but that is really far ahead of the vintage JBL drivers.

I think a double coil version with two separate circuits and the coil half way into both is a perfect solution for this type of driver.

[RMC]

Hi OKS 81,

RE post # 7: "First the equal hight voicecoil/gap 15" driver you mention is the E130.
This is an instrument driver, but can`t see anything wrong with using it for horn and midrange in hifi setup.
As it is equal hight it is very efficient for it`s application, but off course very short stroke.
The specs says Xmax 2,5mm, the VC is then out of the gap so I don`t know how the linearity of this driver is at 2,5mm."

The following, discussing The Magnetic Motor Structure, should clarify the above matter. But I'm not sure someone would really want this in a Hi-Fi type setup:

"The form shown in figure 2-3b [RMC: voice coil and top plate of equal length] concentrates all of the flux in the coil at its rest position. It is evident that even moderate excursions of the voice coil will result in some loss of total flux engaging the voice coil, thus producing distortion. This design is common in very-high-efficiency drivers used for musical instrument amplification, where some degree of distortion may indeed be sonically beneficial." [John Eargle (JBL), Loudspeaker Handbook, P. 23-24]

Best Regards,

Richard