New DIY Mid High (90deg) - AKA PM90

[David Chrapla]

Re: New DIY Mid High

Ah ok. Didnt know where his design came from. But they are very deep. For my build I am looking at this design by Peter, or a variation with my waveguide or a bigger design with cheaper components.

 

[Peter Morris]

Re: New DIY Mid High

A couple of points I should mention again;

· The whole idea of this project was to put as much power and quality on a stick as I could.
· It was initially designed to work with our existing racks, all of which include Lake processing and Powersoft K6 & K10 amplifiers. However a simpler version with a passive HF+VHF and IIR processing will still work quite well.
· I would have liked to get down to 80 Hz at full power, but horn loading the 12" drivers down to that frequency made the box too big.
· If I was prepared to compromise at 100 Hz, it all worked. It does go lower but the output is is limited.
· To minimize the box size and weight, there had to be virtually no unused space within the box.
· To minimize the box size I also had to fold the LF horn. It’s about 2 feet long which load the horn down to about 130Hz - 150Hz, below that it’s assisted by the port. If you don’t fold the horn and use an approach like Danley’s synergy horns, the box ends up being about 3 or more feet deep.
· The LF horn will only work up to about 800Hz. If you want to go higher the volume between the speaker cone and the mouth of the horn needs to be reduced. I suspect a phase plug like Fulcrum uses will be needed. In addition a driver with a lighter cone would be needed. It may even be necessary to use 10” driver. In either case the LF performance will be compromised.
· For the best pattern control the box needed to be roughly twice as high as it is wide.
· If the crossover frequency was around 700Hz and the width of the box kept down to about 15 inches, then the sides of the horn could be parallel and not cause horizontal directivity problems.
· If the width of the HF horn was similar to the width of the box, the pattern of the LF and HF would integrate well (in this case only).
· The maximum height of the HF horn was limited to about 1 ft.; more than that would cause vertical pattern and cancellation issues.

The things I don’t like about the design;

· At 34 Kgs it’s a little difficult to lift onto the “stick”
· Having the horn located in the middle of the box solves a lot of acoustic problems but you need to lift the box a little higher.

The things I like;

· It sounds great; it’s clean, clear, smooth with no harshness.
· The separation between the instruments in the mix is exceptional.
· It goes loud effortlessly and fills the whole room evenly with sound.
· Although it’s not rider spec the initial reaction to the box locally has been so positive I starting to win work because of it.

 

[Peter Morris]

Re: New DIY Mid High

Show Report;

As I said in an earlier post, Darren would be doing a mix on the double 12” DIY last night.

The show could not have gone any better, it sounded fantastic, the band was great and Darren did an awesome mix … just perfect mate!

There was applause for every song and three standing ovations.

There was also a flown double EV Delta Max + subs in-house system. We had time and did a comparison. It was of special interest to me because of all the discussion on this thread about the DH1A compression driver which is used the Delta Max. It wasn't really possible to push the system and find out where the limits were, but it certainly seemed to have more get up and go than the double Delta Max and sounded much better.

The DIY seems to work as well as I hoped ... I just need to build 4 of the 60 degree versions now.

 

[David Chrapla]

Re: New DIY Mid High

Congrats on the praise for your speakers. What in house subs did they use?

Wish I could have made it tonight but I had a few family things to commit to and a sick child. Hope to try and see if your up for a visit very soon.

 

[Peter Morris]

Re: New DIY Mid High

Congrats on the praise for your speakers. What in house subs did they use?

Wish I could have made it tonight but I had a few family things to commit to and a sick child. Hope to try and see if your up for a visit very soon.

EV DML 2181A - 1 sub and 2 x DML 1152 tops per side.

 

[Riku Ylinen]

Re: New DIY Mid High

Peter,

It would really be easier to compare your 12" horns to other designs if we had a little more data about the true sensitivity. Do you have gear to measure calibrated frequency responses?

Or could you do some raw measurements with comparable reference measurement included in the plot? The reference could be your BMS/RCF horn. Or you could also take some other cabinets into your measurements - the moore the better- as long as measurement conditions are equal among them. I'd prefer same voltage-distance-position-no crossovers, only raw measurements. Scaling 8/4/2 ohms etc. can be done afterwards.

Simulations are very helpful but they never beat a proper measurement!

Thank you if you have time for this in the near future!

-Riku

 

[Peter Morris]

Re: New DIY Mid High

Peter,

It would really be easier to compare your 12" horns to other designs if we had a little more data about the true sensitivity. Do you have gear to measure calibrated frequency responses?

Or could you do some raw measurements with comparable reference measurement included in the plot? The reference could be your BMS/RCF horn. Or you could also take some other cabinets into your measurements - the moore the better- as long as measurement conditions are equal among them. I'd prefer same voltage-distance-position-no crossovers, only raw measurements. Scaling 8/4/2 ohms etc. can be done afterwards.

Simulations are very helpful but they never beat a proper measurement!

Thank you if you have time for this in the near future!

-Riku

Its not easy as a DIY-er to produce accurate sensitivity figures.

I did some comparisons with different drivers and horns and came to the conclusion that the DIY HF section was running at about 112 dB watt per meter average, peaking at about 114 dB. This is consistent with the manufactures specifications. It is about 2 dB more efficient in the mid band compared to an 18sound driver I measured - once again this was consistent with the manufactures specifications - see plots below.

I have shown what the Lake processing curves look like assuming that the LF and HF had the same impedance by raising the LF gain 6dB - The gain is then on average 7 dB more on the LF than the HF. Based on the HF sensitivity of 112 dB, the 12" LF has an average sensitivity of 105 dB and a peak of 110 dB.

That my best guess ... and its as per my original specification in my first post. The only change since then is that the crossover is now 700Hz.

I have also modelled everything else I could find and compared them. Its probably reasonable to assume that there is some relativity between the results even if the absolute values are not perfect. In that respect, for its size and weight the DIY has more output than anything I could find. That's one of the reasons I decided to build it.

I hope this helps - Peter

EDIT - PS in the current world of exaggerated BS specifications - LF maximum SPL = 147.5 dB, HF Maximum SPL = 144 dB 8O~8-O~:shock:8O~8-O~:shock::roll: (maximum/peak efficiency x peak watts)

 

[David Chrapla]

Re: New DIY Mid High

Do we really need a sensitivity spec to compare. Once your above about 100db or much higher in this case everything else is easier amp wise. But power is not that expensive unless your restricted to a outdoor solar system. BUT then maybe some people need to use a speaker to is maximum SPL.

And what are we comparing? Raw measurements of just the drivers?

If designing a crossover I can understand this but wondering why we are getting so picky on technical specifications. Seems like a 10m 28.3v FR would be nice. BUT thats what I am used to seeing. BUT I am not telling anyone they need to do this. I dont want to create anymore work for anyone. Just my thoughts.

 

[Hiep Nguyen]

Re: New DIY Mid High

Peter,
Fantastic DIY, will the crossover and iir filter be released anytime soon? The dbx dr260 be enough? The Lake would be expensive. This will be a step up from sls920 with the weight of <70lbs? Keep the good work.
Hiep

 

[Darren Scaresbrook]

Re: New DIY Mid High

It's been a busy week, so I'm only now getting a chance to post with a report on the gig using Peters 2x12"s.

I can say the boxes performed beautifully! The setup was 1 2x12 over 1 2x18 per side. This was in place of the venues Deltamax system of 2 1151 over 1 2x18 per side.
We had enough time at soundcheck to do a direct comparison of the two systems and the double 12 was far superior in every way. The double 12 had much more definition in the top end and had strong, but controlled output, in the low mids. This is something I found lacking in the Danley system I listened to recently. The Danley has a big hole from 100-200hz, whereas Peters box is solid all the way down and transitions smoothly into the sub. This is important as the fundamental frequencies of so many things are found in this area.

The boxes gave excellent coverage o the room. The tonality of the boxes stayed the same everywhere within the coverage of the box indicating it has good pattern control over the entire operating range of the box. I noted, in particular, the low distortion. Even when the gig got louder the sound didn't change. I put this down to all the transducers operating within their pistonic range and the lack of cone/diaphragm break up modes which can cause harshness.

It's hard to comment on maximum output level as I did not reach it. One double 12 could comfortably outrun the 2 Deltamax per side, and then some.

There was really nothing I felt I wanted to change about the box, I'm looking forward to using them again.

Darren

 

[David Chrapla]

Re: New DIY Mid High

Great to hear the results. I will have to try and come to an event where these dual 12's are at to really see them stretch their legs.

 

[Peter Morris]

Re: New DIY Mid High

Peter,
Fantastic DIY, will the crossover and iir filter be released anytime soon? The dbx dr260 be enough? The Lake would be expensive. This will be a step up from sls920 with the weight of <70lbs? Keep the good work.
Hiep

I have some rough IIR settings ... hopefully in a week or two I will have time to do some better settings.

If my SIM's are correct ... about 8 dB or more louder than the SLS 920 ... I posted the calculated comparison above somewhere.

I have never heard the SLS920, however I believe they are still a very good box. I would however expect the FIR version of the DIY to sound much better. I have never run them in an IIR formate other than a quick test, but I suspect they would still sound better that the SLS920.

 

[Rick Powell]

Re: New DIY Mid High

I have never heard the SLS920, however I believe they are still a very good box. I would however expect the FIR version of the DIY to sound much better. I have never run them in an IIR formate other than a quick test, but I suspect they would still sound better that the SLS920.

One thing is for sure; they would definitely have more low-mid end than a SLS920 or even the SLS 960. The 920 starts rolling off around 250 Hz.

 

[Ossian Ott]

Re: New DIY Mid High

admiring build and documentation, wish I had more $$$ for something like this.

I got a bit interested in the Lake processing, here in Sweden it's about 6 times the price of the regular IIR processor(like DRPA+) which could be considered doing the same thing(3 way, two input), which is the one I'm using as we speak. Would that price justify the difference in quality you get from the Lake? I know about phase issues, but I don't see them as that much of a problem as long as frequencies from different sources are not overlapping in a negative way, but I might be totally wrong on this. What magic is going on here, I mean, it doesn't increase the sensitivity of the speaker with 10 dB which I would consider the most important part when paying 6 times as much for a product.

Thanks for a great thread, I'm learning all the time.

 

[Helge A. Bentsen]

Re: New DIY Mid High

Would that price justify the difference in quality you get from the Lake?

Yes.

 

[Peter Morris]

Re: New DIY Mid High

admiring build and documentation, wish I had more $$$ for something like this.

I got a bit interested in the Lake processing, here in Sweden it's about 6 times the price of the regular IIR processor(like DRPA+) which could be considered doing the same thing(3 way, two input), which is the one I'm using as we speak. Would that price justify the difference in quality you get from the Lake? I know about phase issues, but I don't see them as that much of a problem as long as frequencies from different sources are not overlapping in a negative way, but I might be totally wrong on this. What magic is going on here, I mean, it doesn't increase the sensitivity of the speaker with 10 dB which I would consider the most important part when paying 6 times as much for a product.

Thanks for a great thread, I'm learning all the time.

I think the best thing for me to do when I have some time is to produce some IIR settings and publish the frequency, phase and impulse responses results, and of course the settings.

I used the Lake and ran the box 4 - way partly because I already had amp racks configured that way to run my Turbosound Flex Array. I also had a lot of nice medium sized boxes and there was no point building a box unless it was special.

I don’t think the sound difference between the FIR processed box and an IIR processed box will be huge. The phase response will not be flat, the impulse response will not be as good and it will not be able to reproduce a square wave. It will sound nice but not as real as the FIR, but as with so many things, its the last few percent that takes all the effort, and in this case $$$.

If you run the HF and VHF with a passive crossover it will not be possible to use a separate limiter on both sections of the driver. If you drive the system hard your protection on the BMS 4594 will not be quite as good.

I hope that helps.

Peter

 

[Mark Wilkinson]

Re: New DIY Mid High

I think the best thing for me to do when I have some time is to produce some IIR settings and publish the frequency, phase and impulse responses results, and of course the settings.

I used the Lake and ran the box 4 - way partly because I already had amp racks configured that way to run my Turbosound Flex Array. I also had a lot of nice medium sized boxes and there was no point building a box unless it was special.

I don’t think the sound difference between the FIR processed box and an IIR processed box will be huge. The phase response will not be flat, the impulse response will not be as good and it will not be able to reproduce a square wave. It will sound nice but not as real as the FIR, but as with so many things, its the last few percent that takes all the effort, and in this case $$$.

If you run the HF and VHF with a passive crossover it will not be possible to use a separate limiter on both sections of the driver. If you drive the system hard your protection on the BMS 4594 will not be quite as good.

I hope that helps.

Peter

Peter, that would be great!

I am ready to give a pair of the passive xover / IIR version a try. Maybe later I can switch to FIR to keep on learning....

I also think your 90x40 version makes a lot of sense for my intended usage, so no need to wait and see if a 60x40 develops.

Hopefully, plans as specific as possible will continue to emerge...... I'm an ok woodworker....labhorns weren't too difficult...but i really need plans as exact as possible since I have no clue about speaker design.

Also hopefully, a group driver purchase will form, but I won't let the lack of one stop a build here...

Ready to make sawdust.

Many Thanks, Mark

 

[David Chrapla]

Re: New DIY Mid High

I will be interested in seeing what others end up using myself. I will have to slowly gather the parts for this build myself. The BMS being the most expensive part and of coarse the electronics. BUT I was always going to have to use something else other than the Lake products. They are quite nice though.

 

[Peter Morris]

Re: New DIY Mid High

Peter, that would be great!

I am ready to give a pair of the passive xover / IIR version a try. Maybe later I can switch to FIR to keep on learning....

I also think your 90x40 version makes a lot of sense for my intended usage, so no need to wait and see if a 60x40 develops.

Hopefully, plans as specific as possible will continue to emerge...... I'm an ok woodworker....labhorns weren't too difficult...but I really need plans as exact as possible since I have no clue about speaker design.

Also hopefully, a group driver purchase will form, but I won't let the lack of one stop a build here...

Ready to make sawdust.

Many Thanks, Mark

As I have mentioned earlier this box was built as a prototype. While I’m quite competent with most woodwork and metal work projects I don’t have a serious workshop or any drawing software.

What I did was sketch up a full scale drawing of halve the box and checked it fitted the horn response design criteria.

I then worked out a way to construct the box with the tools I had. For example, many of the angles I just hand planed until correct.

At this stage the best I can do is measure what I have done and post it. The shape of the curved mouth of the horn is not that critical. The frequency response if you model that section as conical or exponential is similar. By making curved (exponential) it increases the volume enclosing the two12’s to about the correct amount.

One critical dimension relates to the clearance between the horn driver and the spokes of the 12”. It all fits but there is not much room for error. Check that what you are going to do all fits. The location of the 4 ports is also not that critical – against the side of the box and a few inches from the front of the box.

The next thing I was planning to was build some 60 degree versions …. Now that I have tested the concept and it works I will take some more time and do some better plans ...and now that March is almost over I should have some more time.

In the meantime if anybody has some drawing software and sketches it up I could compare it to what I have made and suggest any changes that maybe necessary.

Peter

Box dimensions: 930mm - H x 430mm - D x 385mm - W
Construction: 17mm ply (light weight pine braced)

 

[Mark Wilkinson]

Re: New DIY Mid High

Thanks for the further info!
I think I'll start trying to put together a cut list & layout.
Best, Mark