PDA

View Full Version : 30-50 ohm resister in series with JBL 2446



David Ketley
02-15-2015, 11:54 AM
I'm sure I've read somewhere that a 30-50 ohm resister in series with the JBL 2446 takes the distortion at 7000 higher can anyone enlighten me on this?

Dave

ivica
02-15-2015, 12:27 PM
I'm sure I've read somewhere that a 30-50 ohm resister in series with the JBL 2446 takes the distortion at 7000 higher can anyone enlighten me on this?

Dave

Hi Dave,

I think that so large resistor "in-series" with 2446 would not be the best solution if You want to reduce its output level.
Usually one "in-series' and one "in-parallel" would be better solution. Values depends on the 'reduction level' You want to get.
In the first quest You can use that the 2446 impedance is about 12~16 Ohms....

reagrds
ivica

martin2395
02-16-2015, 01:06 PM
Also, I've been told not to use a single resistor but at least 2 or 3 paralelled and then put these in series.

pos
02-17-2015, 03:26 AM
When putting such a high value resistors in series you are basically leaning toward current drive mode.
This can reduce distortion, especially with woofers, but will also change the frequency response of the driver, so beware...

David Ketley
02-28-2015, 12:46 PM
I'm sure I've read somewhere that a 30-50 ohm resister in series with the JBL 2446 takes the distortion at 7000 higher can anyone enlighten me on this?

Dave

I see its already been well covered here http://www.audioheritage.org/vbulletin/archive/index.php/t-7027.html

Dave

pos
02-28-2015, 12:49 PM
So you meant a parallel resistor, not a series one.

David Ketley
02-28-2015, 01:55 PM
So you meant a parallel resistor, not a series one.

I had only a very brief note of a series resistor being used but this discussion clears up a lot of my questions.

Dave

Ian Mackenzie
03-01-2015, 11:54 AM
Pos is right.

Nelson Pass has written an article on this topic for the Amplifiers that have this property..

The modification to the response will be more significant than a reduction in distortion as an audible change.

At the low frequency end it will effect the crossover characteristic where you are using a passive crossover network.


What type of diaphragm are you using?

Since that very old thread there have been advances in both diaphragm and horn design.

If you have access to a RTA it will help validate any modifications.

Back to the point I am aware of resistive shunting of drivers.

Basically a driver is an electro mechanical motor that is not resistive but can be also inductive. By swamping the impedance with a low value resister the driver becomes largely a resister by characteristic. This is good for the driver and good for the network looking at the driver. The overall impedance would still need to be a suitable load for the amplifier

Assuming a loss in sensitivity is not a problem there are advantages.

I would also have a read of the crossover design paper on the PI loudspeaker site

David Ketley
03-01-2015, 02:22 PM
Pos is right.

Nelson Pass has written an article on this topic for the Amplifiers that have this property..

The modification to the response will be more significant than a reduction in distortion as an audible change.

At the low frequency end it will effect the crossover characteristic where you are using a passive crossover network.


What type of diaphragm are you using?

Since that very old thread there have been advances in both diaphragm and horn design.

If you have access to a RTA it will help validate any modifications.

Back to the point I am aware of resistive shunting of drivers.

Basically a driver is an electro mechanical motor that is not resistive but can be also inductive. By swamping the impedance with a low value resister the driver becomes largely a resister by characteristic. This is good for the driver and good for the network looking at the driver. The overall impedance would still need to be a suitable load for the amplifier

Assuming a loss in sensitivity is not a problem there are advantages.

I would also have a read of the crossover design paper on the PI loudspeaker site
Thanks Ian,
Im using the Jbl 2446 with the JBL diaphragms not Beryllium yet that is a cost I will have to defer for a while. Its the fact that in the region I want to cross to the JBL 2405 8000hz I gather there are issues with the response there. Im using three way active with a Marchand crossover then a cap or high pass filter to the 2405s, but wondered what resisters in series or parallel would do to the upper response of the 2446?
Does a L pad have the same sort of affect on the frequency response and distortion after all these are series parallel resisters? I run Musical Fidelity amplifiers.

David Ketley
03-05-2015, 12:14 PM
Thanks Ian,
Im using the Jbl 2446 with the JBL diaphragms not Beryllium yet that is a cost I will have to defer for a while. Its the fact that in the region I want to cross to the JBL 2405 8000hz I gather there are issues with the response there. Im using three way active with a Marchand crossover then a cap or high pass filter to the 2405s, but wondered what resisters in series or parallel would do to the upper response of the 2446?
Does a L pad have the same sort of affect on the frequency response and distortion after all these are series parallel resisters? I run Musical Fidelity amplifiers.


Thanks Ian the Pi site made things very clear.

Dave

David Ketley
03-05-2015, 01:22 PM
Thanks Ian the Pi site made things very clear.

Dave


Ian I run an active system using a Marchand XM44-3AA crossover. Would this set up gain with experimenting with some passive items installed at the speaker end?

2245 2123 2246 2405.

Thank you.

.:hb:.
03-05-2015, 04:31 PM
What's the point of using an active filter if you want to add "some passive items" in the end?
If you are not happy with your sound and want to fiddle with your system's frequency response, get an equalizer. Or, better, use an active filter with bulit in equalizing possibilties - which will most likely be a DSP speaker management system.
Playing around with passive components doesn't get you anywhere, if you haven't the chance to actually measure what you are doing and, much more important, interprete your measurements.

Ian Mackenzie
03-06-2015, 05:50 AM
I would put a 20 ohm resister in parrellel with the 2446 and 10 ohm in parrellel with the 2405.

Typically a JBL compression driver is about 12 ohms impedance with an ac signal in the pass and.

In the absence of an impedance measurement that will make the drivers look more like 8 ohms.

You can then use an 8ohm L pad to passively pad back both drivers to bring down the s/n ratio.

If you use a large enough capacitor before the L pad on e 2446 it will look like low impedance in series with the driver in the passband. Bypass the large capacitor with 0.01 uf high quality capacitor. If you read up on the titanium innovation JBL used the self resonance to help extend the HF response. But it's a bit like bacon in a fry pan compared to a piston range driver like Be.

If you have an impedance curve we can make an lcr shunt to smooth the impedance at the low end of the horn and this damp the diaphragm.

The 2405 apparently works better with a low impedance souce so if you have a suitable capacitor at the input of the amp that will block any doc as long as the amp does not have a turn on thump.

What horn are you using?

If you are using a cd horn that hold up the horizontal dispersion to 10-13khetz I would try just try a soft high pass filter on the 2405 at above 15k.

If that works see if you can get the 2446 diaphragm dusted with aquaplas as that will damp the self resonance in the titanium and reduce the glare and lower the noise floor of the driver.

JBL did this with the early K series systems. It's a economical compromise to Be.

I use it in my 4344mk11 (diy)

David Ketley
03-06-2015, 09:11 AM
I would put a 20 ohm resister in parrellel with the 2446 and 10 ohm in parrellel with the 2405.

Typically a JBL compression driver is about 12 ohms impedance with an ac signal in the pass and.

In the absence of an impedance measurement that will make the drivers look more like 8 ohms.

You can then use an 8ohm L pad to passively pad back both drivers to bring down the s/n ratio.

If you use a large enough capacitor before the L pad on e 2446 it will look like low impedance in series with the driver in the passband. Bypass the large capacitor with 0.01 uf high quality capacitor. If you read up on the titanium innovation JBL used the self resonance to help extend the HF response. But it's a bit like bacon in a fry pan compared to a piston range driver like Be.

If you have an impedance curve we can make an lcr shunt to smooth the impedance at the low end of the horn and this damp the diaphragm.

The 2405 apparently works better with a low impedance souce so if you have a suitable capacitor at the input of the amp that will block any doc as long as the amp does not have a turn on thump.

What horn are you using?

If you are using a cd horn that hold up the horizontal dispersion to 10-13khetz I would try just try a soft high pass filter on the 2405 at above 15k.

If that works see if you can get the 2446 diaphragm dusted with aquaplas as that will damp the self resonance in the titanium and reduce the glare and lower the noise floor of the driver.

JBL did this with the early K series systems. It's a economical compromise to Be.

I use it in my 4344mk11 (diy)

Thank you for the information Ian. I realise that often someone has spent many hours and extended knowledge researching and coming to the right conclusions regarding specific audio components and crossovers and am extremely grateful for the advice.
Im on a tight budget running the 2446 with a 2" to 1.4" reducer on a 18 sound XT1464 horn crossed at 1600 hz. As the JBL 2446 seems to need correction above 8000Hz that's my crossover to the 2405 with either a capacitor or I might try a high pass and see how it sounds. Looking at the response of the 2446 I'm surprised you suggest running them up to 13000Hz? I do intend to upgrade to the Iwata 300 horn as soon as funds allow and then some Be diaphragms. I'm also looking for a simple way to try and time align the 2405, of course one can do this physically but it puts it so far back from the mouth of the horn that it creates its own problems. I'm looking to try a cheap and simple solution but am not happy about adding to the signal path, Behringer Shark DSP 110.

Ian Mackenzie
03-06-2015, 12:54 PM
Okay

Most of these compression drivers when used in CD horns like the 18 Sound XT1464 either a passive eq or active eq to flatten the response.

You may have mentioned earlier but do you have any form of RTA or measurement kit?

There are some reasonable diy options at Parts Express.

Attempting to optimise CD horn passively is difficult without an crossover software package with an optimiser and take some skill.

For diy home audio all CD horns require an optimised eq contour as they tend to have constant directivity that means in the non eq state the on axis response looks like a banana.

The iwata is more diy friendly but will still require some intervention.

Many years ago I had a JBL 2397 horn and I took it to a friend who was running a college loudspeaker builder course and he measured the horn for me and built the crossover for the 2405.

Depending on where you are that can be a blessing if you can reach out and find some assistance.

If that is not an option you can self help and look at what the 18 Sound people do with this particular horn to make it function correctly.

A starting point is the 18 sound web site enclosure design tab. If you have a look at the 15 inch 2 way it's quite intuitive process. If you are keen as a starting point just use the eq part of the 18 Sound crossover and place a 1.5 uf capacitor in parrellel with a 27 ohm resister. This will be approximate but will be in the ball park as to normalising the CD characteristic of the XT1464 horn.

The effect of this is a response shaping network. It woks by attenuating the response from a selected high frequency in the form of a high pass filter . At the same time the lower end of the response has some sensitivity shave off

Most compression driver work on the same principle so you are not doing anything wrong by trying this.

Still use the large series capacitor before the above eq and crossover with the Marchand crossover at 1500 Hz. I would also suggest a fuse in series to protect the driver in the event of an accident.

The Lpad will not be required.

Measure the response if you can or listen to the horn with Pink noise (or an off station FM radio) and compare to the non EQ version of the crossover. Do not connect the 2405 at this point.

The horn should sound smoother and more extended.

You can then add the 2405 back and compare.

The horn will extend further but will tend to sound soft in the upper hf area. This is normal and I would listen carefully on some know program content .

Adding the 2405 will add spatial resolution with a crisp crystal quality.

Let me know how you go

ivica
03-06-2015, 12:58 PM
Thank you for the information Ian. I realise that often someone has spent many hours and extended knowledge researching and coming to the right conclusions regarding specific audio components and crossovers and am extremely grateful for the advice.
Im on a tight budget running the 2446 with a 2" to 1.4" reducer on a 18 sound XT1464 horn crossed at 1600 hz. As the JBL 2446 seems to need correction above 8000Hz that's my crossover to the 2405 with either a capacitor or I might try a high pass and see how it sounds. Looking at the response of the 2446 I'm surprised you suggest running them up to 13000Hz? I do intend to upgrade to the Iwata 300 horn as soon as funds allow and then some Be diaphragms. I'm also looking for a simple way to try and time align the 2405, of course one can do this physically but it puts it so far back from the mouth of the horn that it creates its own problems. I'm looking to try a cheap and simple solution but am not happy about adding to the signal path, Behringer Shark DSP 110.

Hi David Ketley,

I would suggest you to use XR2064 -18sound 2inch horn if you prefer to get the most out of such driver as 2446. converting from 2inch to 1.4inch would not be the best solution. From my experience 2446 would work over 10kHz without any effort, if original not over powered diaphragm is used.

regards ivica

David Ketley
03-06-2015, 01:52 PM
Okay

Most of these compression drivers when used in CD horns like the 18 Sound XT1464 either a passive eq or active eq to flatten the response.

You may have mentioned earlier but do you have any form of RTA or measurement kit?

There are some reasonable diy options at Parts Express.

Attempting to optimise CD horn passively is difficult without an crossover software package with an optimiser and take some skill.

For diy home audio all CD horns require an optimised eq contour as they tend to have constant directivity that means in the non eq state the on axis response looks like a banana.

The iwata is more diy friendly but will still require some intervention.

Many years ago I had a JBL 2397 horn and I took it to a friend who was running a college loudspeaker builder course and he measured the horn for me and built the crossover for the 2405.

Depending on where you are that can be a blessing if you can reach out and find some assistance.

If that is not an option you can self help and look at what the 18 Sound people do with this particular horn to make it function correctly.

A starting point is the 18 sound web site enclosure design tab. If you have a look at the 15 inch 2 way it's quite intuitive process. If you are keen as a starting point just use the eq part of the 18 Sound crossover and place a 1.5 uf capacitor in parrellel with a 27 ohm resister. This will be approximate but will be in the ball park as to normalising the CD characteristic of the XT1464 horn.

The effect of this is a response shaping network. It woks by attenuating the response from a selected high frequency in the form of a high pass filter . At the same time the lower end of the response has some sensitivity shave off

Most compression driver work on the same principle so you are not doing anything wrong by trying this.

Still use the large series capacitor before the above eq and crossover with the Marchand crossover at 1500 Hz. I would also suggest a fuse in series to protect the driver in the event of an accident.

The Lpad will not be required.

Measure the response if you can or listen to the horn with Pink noise (or an off station FM radio) and compare to the non EQ version of the crossover. Do not connect the 2405 at this point.

The horn should sound smoother and more extended.

You can then add the 2405 back and compare.

The horn will extend further but will tend to sound soft in the upper hf area. This is normal and I would listen carefully on some know program content .

Adding the 2405 will add spatial resolution with a crisp crystal quality.

I see http://www.eighteensound.com/Portals/0/EnclosuresKits/18sound_15%202ways.pdf

Let me know how you go

I see http://www.eighteensound.com/Portals/0/EnclosuresKits/18sound_15%202ways.pdf

I,m sitting in our caravan on the North Coast of Spain waiting to pick up a ferry home to the UK on Monday from Santander rafter 3 months away.
No I do not have any measuring equipment, I must admit to being reluctant as I know that although its important to get the basics right the endless pursuit of perfection can become an obsession leaving little time for actually listening, hence my trying to get a simple system that sounds good to me. The Marchand does have the facility to add correction cards for the drivers but as I intend to put in Be diaphragms and upgrade to the Iwata 300 later I will try and get the best out of the existing drivers following your suggestions. I did find that mounting my last UF unit a Heil P5 on a alloy plate in line with the top of the horn gave a nice response and it was correctly time aligned to the listening position. Will update on my return.
Thanks

Ian Mackenzie
03-06-2015, 02:05 PM
Hi David,

Enjoy Spain and we can pursue this when you get home!

There are a number of great Lansing forum people in the UK and Europe who I am sure would offer to assist and from diy audio.com

Inrecall some off us met at gathering to hear Richards 4345 in SE London back in 2007.

Most UK audio buffs would be more familiar with small book shelve or narrow tower loudspeaker.

I wonder what they would think of the JBL movement?