Project M2 DIY Thread

Hi Scott,

Thank you for chiming in and your PM.

With your 2451 driver are you able to post a graph of the overall DSP corrected curve with a flat HF response. Also which particular diaphragm you have un your 2451? A screen dump is fine.

I will then look at the passive network. Ideally a Clio measurement of the 2451 impedance and a response measurement would allow me to import the data into Leap and simulate the passive response. A measurement with REW would also be of assistance.

I will look at a 2447 driver l have and assess it with a simulation.
Ian

Quote:

---

Originally Posted by Ian Mackenzie

Hi Scott,

Thank you for chiming in and your PM.

With your 2451 driver are you able to post a graph of the overall DSP corrected curve with a flat HF response. Also which particular diaphragm you have un your 2451? A screen dump is fine.

I will then look at the passive network. Ideally a Clio measurement of the 2451 impedance and a response measurement would allow me to import the data into Leap and simulate the passive response. A measurement with REW would also be of assistance.

I will look at a 2447 driver l have and assess it with a simulation.
Ian

---

The 2450,51,52,53 SL series are all pretty good. Much cleaner up top. I used the same network configuration optimized for an 8 ohm 2453 SL and it dialed right in. The 2451 SL, except for the bolt pattern, is about as close to the 476MG as you are likely to get. So it should work just fine. You might loose one of the parallel notches. Curious to see how it plays out!

Rob :)

Quote:

---

Originally Posted by Ian Mackenzie

Hi Scott,

Thank you for chiming in and your PM.

With your 2451 driver are you able to post a graph of the overall DSP corrected curve with a flat HF response. Also which particular diaphragm you have un your 2451? A screen dump is fine.

I will then look at the passive network. Ideally a Clio measurement of the 2451 impedance and a response measurement would allow me to import the data into Leap and simulate the passive response. A measurement with REW would also be of assistance.

I will look at a 2447 driver l have and assess it with a simulation.
Ian

---

The current setup has the 2451 with a D8R2450SL. The weekend is shot for doing anything audio, but will try early next week.

Thanks,
Scott

Quote:

---

Originally Posted by Robh3606

The 2450,51,52,53 SL series are all pretty good. Much cleaner up top. I used the same network configuration optimized for an 8 ohm 2453 SL and it dialed right in. The 2451 SL, except for the bolt pattern, is about as close to the 476MG as you are likely to get. So it should work just fine. You might loose one of the parallel notches. Curious to see how it plays out!

Rob :)

---

Hi Rob,

Thanks.

I will go though your passive monitor thread over the weekend and study your curves on the passive EQ compensation. Btw what is the static impedance of the 476 above the M2 horn effects? 12 ohms?

Ian

As a curious fellow I stumbled across Leap FIR synthesis tonight and deployed the design Wizard.

This is what popped out. Don't ask me how (because they said don't ask - its landed from another world where hifi is obsolete)

The schematic looks like a Terminator Chip!

Robert's got the same software but of course your talking about two people with somewhat different attributes.......Lol. It all happened while you were asleep. Take the piss mate...........

Edit I dont know why the program has not EQ'd the low pass below 300 Hertz, Further reading of the large manuals required.

Ian

Hi Scott,

I've purchased a pair of 2451 drivers and the SL diaphragms to add my inventory for test and measure. They will arrive in a couple of weeks.

This afternoon as advised I have reviewed the known measurements comparing the 2430K and the 2451h/J on several web forums and referenced them to RoB Hamel's voltage drivers here on LHS for his 476 Mg drivers. The consensus is the 2451 is the go to driver with the SL dusted diagrams which I have in my inventory. I am going to deliberate over the rationalisation of proposed passive crossover and passive options in the near future. The behaviour of the 2451 is likely similar to my 2447 1.5 inch driver in the crossover region and I will refer on that with testing next week.

With known measurement data I can then simulate and live test the 2451/2447 driver in the crossover region with reasonable confidence and arrive at the full passive crossover /EQ hopefully within the next ten days subject to prevailing weather. I will then order the crossover parts and do a live test and measure. If you are interested I will make you an offer on the beta test crossover at wholesale cost once the testing is complete. I prefer not to hoard project crossovers once they reaches live test and measure stage.

I like to keep this type of activity moving through my operating rhythm and close the loop as soon as practical.

I will apprise you of the actual voltage drives, response curves and relevant adjustable response shaping once I get a bit closer to live testing.

If you are aware of any who maybe interested in pairing a specific driver to the M2 horn with a passive crossover send them here.

Please let me know if you have any questions.

As you can see I have touched on an FIR simulation above and will pursue that as part of my M2 program in parallel with my M2 passive and analogue active audio amateur solutions.

Ian

Quote:

---

Originally Posted by Ian Mackenzie

Hi Rob,

Thanks.

I will go though your passive monitor thread over the weekend and study your curves on the passive EQ compensation. Btw what is the static impedance of the 476 above the M2 horn effects? 12 ohms?

Ian

---

Yes that's correct! Going into the digital realm in LEAP never tried are those boxes individual EQ filters?? Do you have to enter a digital processing limit or is it open ended??

Rob :)

Hello Robert,

Thanks for the impedance confirmation.

I now also have Scott’s Dat impedance file and will investigate.

The FIR post was just that. I have yet to compare the voltage drives with those on Pos M2 crossover document. It’s works the same as the passive crossover using Wizard. If your curious have a look at the help menus. Or refer to the hard copy manuals if you have them. It’s a real shame the developer passed. I think he spent some time working with Ed May. There’s nothing particularly different to running an FIR simulation with the Wizard to running a passive crossover simulation.

Have you used Wizard before? It’s relatively easy. My impression is you may only be a casual user of Leap. Once you master how to apply optimisation it’s a very efficient tool. Manually updating a schematic and fitting a curve is tedious but it’s great therapy like mind craft.

Top open the Wizard simply click on File and the drop down menu will show Wizard.
When you click on Wizard the dialogue box opens with options for passive, analogue active and options for digital. Put the your guide curves as normal prior to running Wizard. Once the dialogue selection boxes for crossover frequency and the driver guide curves are populated click commence.
Wizard then enables one of Leap’s optimisation engines and it does the rest. It takes about 5 seconds to create the simulation.

Kinda makes setting up a Crown amp look painful. I personally find the back end of anything digital tedious and time consuming. I spent years analysing, designing and improving customer journeys for a large bank and the systems side of it is often a mishmash of overly complicated processes by program code writers over in Mumbai who knew nothing whatsoever about what the solution was for and what it was supposed to do. It’s the same with any industrial applications. But that’s the way this stuff is done these days. It’s never end user friendly and be careful what you wish for.

To do this manually in Leap would be rather involved as the user needs to drive the program as a first time user. There is a learning curve. I have not explored this further. I only work from 9.00 - 5.00. Once l have the passive M2 diy crossover wrapped up for Scott l will look at it. l need to delve into the manuals to investigate the whys and wherefores when time permits. By all means explore it yourself and see what you come up with ifyou have the time. I’m actually quite busy with several projects at the moment and a renovation next week.

My brief view of the “Alien” the controls on the Dashboard indicates there are a number of approaches to building an FIR crossover with Leap. On the face of it though it does open the door for FIR dsp simulations which is really cool.

Edit. For those less familiar with me l pop in here occasionally and generally on request. I assist loudspeaker builders with troubleshooting or road blocks to completion of their projects and resolving a range of technical questions.

I’m generally too pre occupied for an on going social yabba fest but l will say hello.

In loudspeakers and audio rarely does anything come together first time smoothly and often you will have un answered questions or a problem that needs fixing or a solution. So if you get stuck or need clarification on something just knock on my door with a pm. My pm box is almost full so bare with my while l clean it out.

If you need a start up on a project from scratch or a re design for alternative drivers which happens a lot or ongoing technical support l can assist with a consultation. Depending on the extent of the work l may charge for my time. No real advice is free and its a small premium on your overall investment in your project. For the most part if you scout around long enough you will find what you are looking for. But there is no certainty of the accuracy of the information or if the context specifically answered your particular scenario. In diy audio there is no expectation for you to know anything or know everything. Sometimes misinformation, or an incorrect understanding can lead to confusion. It happens. So it’s a big leg up if you can go to a single, trusted and reliable source of information and get it all right first time. It can save a stressful or frustrating moment out in your shed and a lot of time.

I refer to this as the C factor. The 3 C’s.

Certainty of your technical facts on drivers, buyer advocacy, schematics, parts lists, assembly ect to your particular scenario.

Certainty on completion of your project. Troubleshooting.

Certainty on obtaining the desired results, testing and set up.

I enjoy helping people get their project to completion and making it perform to their expectations.

Ian

This is Dash.

Dash is Director of Acoustics DC to Dog Whistle and has the final word on what’s subjectively acceptable. Go Dash.

Hi Robert

We all have our own journey with this sort of learning curve type tech software.

It was love hate for me but l have had to persevere with Leap as it’s R& D and client solutions now. All my active crossover work is in Leap and Tina TI Filter Pro for all the systems. We load up the actual voltage drive using actual driver loads then optimise to the voltage drive with the woofer reactance as that makes a difference. The modelling in never 100% so we match by measurement with LMS. Greg has truly incredible insights on how drivers behave. They are motors after all. It’s a big step up on plugging in an active crossover and hoping for the best. In diy you cab relax the required specification but commercially it has a to work correctly not just theoretically.

If you measure the voltage drive of a passive system at the driver terminals the low pass filter is an eye opener? What really happens is quite interesting. Then look at large signals. Things like passive crossover gain and bandpass filter responses go to hell when you bi amp. So you have to get right into the deep end to work it out.

Btw Loudsoft in the EU offer more user friendly Apps that l think Peter Larson developed with big players like Harman but is expensive. The automated optimisation is fantastic. Just put the curser on the crossover point and scroll with Mouse wheel to move the crossover till you find your optimum system response. The impedance, phase and everything is auto on the fly. Harbeth use it apparently.

Leap can be weird until you get your head around the whole process. Once you break the back of it then it all makes sense. Leap has a strong maths backbone so it’s very robust. It’s about speed to get the job done with multiple deadlines.

There is a component sensitivity function to see what component can be deleted. From scratch the Wizard will always create insane impedance circuits that aren’t viable, do l do that manually

ON a new design what I do is start with some pre determined control like the driver impedance. It can do a calculation or do it manually for the woofer reactive component above the R minimum on the curve than make it flat with an RC network. I find the passive synthesis schematic values very accurate if the impedance is pre-conditioned. I did this with the 6th order low pass filter on the M2 diy crossover. The compression drivers are largely resistive so l damp the peaks with a shunt resistance sometimes. The synthesis high passive components are then more stable on the optimisation object curve. In optimisation it’s then less likely to freak out.

Where optimisation is really valuable is global where you need a room curve or you have a rising woofer response and you need to tilt the woofers acoustic response. You have to tell Leap the objective. I did that with the M2 diy passive crossover by calculating the voltage drive on the M2 dsp curve at 100 hz and at 800 hertz. That is the ballpark objective in optimisation less the -6db crossover point. I then read up on the comments around the Harman dsp file to get the net sensitivity which is 92db. Without that your shooting in the dark trying to optimise. It was then simple to reduce the Flat Objective Line from 95db to 92db. Not hard but needed to find that value. I had a simulation of the 4367 low pass network loaded do it was easy to let the optimisation engine crunch the values for the 6th order network. The voltage drives matched the 4367 perfectly but extended to the 6th order 800 hertz crossover. Did it beautifully. Impossible to to manually with so many values. I got the whole thing done in two hours.

Unfortunately l can’t afford a whole day or into the wee hours with competing demands on my time. Communicating what your doing is also important when working within a team on a project. This stuff does my head in at the best of times so l prefer to get in get it done once and move onto the next job. In summary l research the project, get the key facts then l define the challenge or problem for Leap. So it’s fairly focused. A tool is only as good as the operator. It all starts with reliable measurement data of course.

On your last comment its fine to copy but that does not mean one approach is always fit for the purpose. Your only speaking for yourself BTW. Hardly anyone without Leap has the vaguest notion of how those complicated networks function. My advice is try and be more self aware and less a set in your ways.

I doubt Greg would agree with you. Those system schematics are for one end use need only. They are not diy networks. It depends on the level of complexity that is appropriate and flexibility or adaptability if needed. For example where the design needs to address more than one driver for a user in a diy situation. Most users don't have the luxury of Leap or anything like it to dial in. When you post you dialled it in what does that mean to anyone reading it? It reads like only something you understand. Pretty selfish.

So thought needs to be given to as topology that is flexible in some situations. There are a number of ways of arranging passive EQ. Then consider the cost and complexity for construction. Odd inductor and capacitor values and minimum impedances.

I am not sure what to make of this, but Speaker Exchange appears to offering a substantial discount on M2 woofers (2216nd).
Possibly a 25-30% discount off their usual price.

The reason I am hedging on this is the website shows a "sales" price of $399. However it also shows it as being "out of stock" although you can join the "waitlist".

In the back of mind I am wondering if this showing any indication that JBL is dropping this woofer from their line (just my paranoid speculation and I have no other information).

It would be fun to DIY a M2-like cabinet with a second "helper woofer". Food for thought ......

Quote:

---

Originally Posted by withTarragon

I am not sure what to make of this, but Speaker Exchange appears to offering a substantial discount on M2 woofers (2216nd).
Possibly a 25-30% discount off their usual price.

The reason I am hedging on this is the website shows a "sales" price of $399. However it also shows it as being "out of stock" although you can join the "waitlist".

In the back of mind I am wondering if this showing any indication that JBL is dropping this woofer from their line (just my paranoid speculation and I have no other information).

It would be fun to DIY a M2-like cabinet with a second "helper woofer". Food for thought ......

---

Give them a call they may have some insight on availability. I wouldn't think they would cancel with the M2 going strong but who knows! For the very least would need spares for warranty repairs. I purchased a pair right before they went out of stock for that price. I have purchased from them several times and service was always good.

Rob :)

Quote:

---

Originally Posted by Robh3606

Give them a call they may have some insight on availability. I wouldn't think they would cancel with the M2 going strong but who knows! For the very least would need spares for warranty repairs. I purchased a pair right before they went out of stock for that price. I have purchased from them several times and service was always good.

Rob :)

---

I checked with Pro Sound Solutions. They have them comparably priced (actually cheaper since they don't charge for tax and shipping - or at least they did not a year and half ago).

As far as maintaining spares, JBL has not been very good about that in the last few years. This woofer is only used in two cabinets i believe, and neither seem to be getting much promotion.

Perhaps after a glass of wine this evening, I might convince myself to run a couple of models this week for a M2 that is about 40% smaller with cabinet/vent tuned close to an octave higher and call it a center channel. Fun to think about.

Quote:

---

Originally Posted by withTarragon

I checked with Pro Sound Solutions. They have them comparably priced (actually cheaper since they don't charge for tax and shipping - or at least they did not a year and half ago).

As far as maintaining spares, JBL has not been very good about that in the last few years. This woofer is only used in two cabinets i believe, and neither seem to be getting much promotion.

Perhaps after a glass of wine this evening, I might convince myself to run a couple of models this week for a M2 that is about 40% smaller with cabinet/vent tuned close to an octave higher and call it a center channel. Fun to think about.

---

I agree with your view of JBL legacy inventory.

My perception is JBL has a minimum production run on drivers such as the 2216nd to for economies of scale. However the production runs on high value finished products like the M2 are actually quite small. Perhaps 50-100 units to fill the global supply chain. Having working capital tied up in stock makes Harman’s anally retentive bean counters look bad to so they shovel off the surplus 2216nd production through a wholesale margin leaving only a handful of 2216nds for the odd warrant failure or when someone blows one up. People blew up the Everest 1500ALs so they added a 1/4 inch winding depth to the VC…apparently.

If you did manage to contact Reconers l’d be interested to know the pricing story.

Harman’s mentality on distribution and pricing has had a lot of industry on lookers scratching their heads. A lot of companies are failing post pandemic so you might best grab your wish list now. They are predicting a global shortage of liquidators and administrators .

Quote:

---

Originally Posted by Ian Mackenzie

I agree with your view of JBL legacy inventory.

My perception is JBL has a minimum production run on drivers such as the 2216nd to for economies of scale. However the production runs on high value finished products like the M2 are actually quite small. Perhaps 50-100 units to fill the global supply chain. Having working capital tied up in stock makes Harman’s anally retentive bean counters look bad to so they shovel off the surplus 2216nd production through a wholesale margin leaving only a handful of 2216nds for the odd warrant failure or when someone blows one up. People blew up the Everest 1500ALs so they added a 1/4 inch winding depth to the VC…apparently.

If you did manage to contact Reconers l’d be interested to know the pricing story.

Harman’s mentality on distribution and pricing has had a lot of industry on lookers scratching their heads. A lot of companies are failing post pandemic so you might best grab your wish list now. They are predicting a global shortage of liquidators and administrators .

---

2216ND PRO, Harman Service Pro has a delivery date scheduled for June 23, 2022, for 23 units. 9 more coming in July. 21 more coming in Sept.

I can order them for you if you want them. Prepaid, non-refundable special orders.

Inquire at [email protected].

Yes...I am Harman Authorized Service with an active account.