Editors Note:
Deep in the thread for the Open Source Monkey Box:
Open Source Monkey Box
were some posts regarding mounting the speaker flush in the wall. This is often desired for studio monitors and people who just happen to have an adjacent closet where speakers could be mounted. Since in most cases a in-wall flush mount speaker is going to have to be a good old (monkey) box type, the excellent and efficient Monkey Boxes are a logical choice. Of course the port would have to be moved to the front or converted to infinite baffle, and the crossover redesigned, so I’ve corralled the relevant posts here on a new thread hoping that it will take off with some good design work!
Variac,
DiyAudio Editor
—————————————————————————————
In post 835 Open Source Monkey Box
You mentioned a "user option" to decide whether to keep the port on the back, as is specified in the documentation, or move it to the front as either a slot or round port.
For flush mounting, it would be preferable to move the port to the front.
I know you originally built a box with a front slot port. But you also mentioned that the back round port seemed to improve low freq response a bit. Would a front round port do the same?
Either style of front port seems to require enlarging the box height a bit. Any thoughts on the best way to accommodate a front port? Slot, round, position, how much to expand box?
thanks, gabo
By the way FYI, I'm back to using the passive crossover. Mostly due to protection issues. Some very expensive tweeters in there that I don't want to destroy due to a power blip or other issue that might cause an amp to throw out a spike. Protecting the tweeters with either a cap or bulb introduces undesired effects compared to just using the full XO.
Deep in the thread for the Open Source Monkey Box:
Open Source Monkey Box
were some posts regarding mounting the speaker flush in the wall. This is often desired for studio monitors and people who just happen to have an adjacent closet where speakers could be mounted. Since in most cases a in-wall flush mount speaker is going to have to be a good old (monkey) box type, the excellent and efficient Monkey Boxes are a logical choice. Of course the port would have to be moved to the front or converted to infinite baffle, and the crossover redesigned, so I’ve corralled the relevant posts here on a new thread hoping that it will take off with some good design work!
Variac,
DiyAudio Editor
—————————————————————————————
In post 835 Open Source Monkey Box
You mentioned a "user option" to decide whether to keep the port on the back, as is specified in the documentation, or move it to the front as either a slot or round port.
For flush mounting, it would be preferable to move the port to the front.
I know you originally built a box with a front slot port. But you also mentioned that the back round port seemed to improve low freq response a bit. Would a front round port do the same?
Either style of front port seems to require enlarging the box height a bit. Any thoughts on the best way to accommodate a front port? Slot, round, position, how much to expand box?
thanks, gabo
By the way FYI, I'm back to using the passive crossover. Mostly due to protection issues. Some very expensive tweeters in there that I don't want to destroy due to a power blip or other issue that might cause an amp to throw out a spike. Protecting the tweeters with either a cap or bulb introduces undesired effects compared to just using the full XO.
Last edited by a moderator:
There are several questions in there, and we had some heated debates about these questions. Anyway, here is what I think:
- Port on the rear or front? The advantage of a rear port is that any midrange garbage that comes out of the port is directed away from the listener. This is why my OSMCs have rear ports. If you want to put the box flush to a wall, a rear port is not a good choice, though.
- Vertical position of the port? A port right at the top or at the bottom samples the pressure node of the standing wave that develops along the vertical axis of the box. The port likes to couple to this pressure node, and will therefore "release" the standing wave resonance to the outside of the box, which is not desired. A port near the center of the box sees the velocity node of the standing wave, which is better.
- Shape of the port (round vs. slot)? I honestly don't think a round vs rectangular (slot) port makes a big difference.
The issue with the original idea of the slot port at the bottom of the box was that it developed a 1/4 wave resonance. This was a bit an awkward observation, but the many tests showed that this really was the case. The inside "end" of the port was terminated by the rear wall of the box, with an upwards opening of the port toward the interior of the box. Since the respective "rear wall area" and the port opening area are about the same, the inside end of the port was a bit like a 50% closed pipe and a 50% open port (I hope you get what I mean). In any case, I don't recommend this particular port geometry as it was in this first prototype. A possible way out would be to make the port a bit shorter (with a smaller diameter), to avoid the 50%-50% termination at the inside.
You might also consider a port on the side of the box. This would allow you to move it to the vertical center of the box, thus avoiding the pressure node of the standing wave.
Thanks Mathias. This is for the studio monitor project I'm working on, so the monitors will be flush mounted in the wall, which means we need to find the best solution to a port on the front. Back and side options are not the best.
Obviously there are also going to be other issues with flush mounting. They will probably have about a 5+ db rise in low frequency due to mostly getting rid of the baffle step and also a few other baffle changes due to widening the effective front of the box.
I'll probably just EQ those out with DSP, which unless there is something really nasty going on, could also work for the port situation. The increased output for the low end just means we get better effective efficiency because EQ'ing that out means lowering the power in the area that uses the most power. Furthermore, there will be a subwoofer added, so most everything below about 80hz will be transferred to there anyway.
Do you think the slot port would work better on the top? I need to go back through and find your dimensions of that port and the box dimension changes when you had it in.
I will also play around with hornresp and vituixcad to refresh my memory of what various port options do.
gabo
Obviously there are also going to be other issues with flush mounting. They will probably have about a 5+ db rise in low frequency due to mostly getting rid of the baffle step and also a few other baffle changes due to widening the effective front of the box.
I'll probably just EQ those out with DSP, which unless there is something really nasty going on, could also work for the port situation. The increased output for the low end just means we get better effective efficiency because EQ'ing that out means lowering the power in the area that uses the most power. Furthermore, there will be a subwoofer added, so most everything below about 80hz will be transferred to there anyway.
Do you think the slot port would work better on the top? I need to go back through and find your dimensions of that port and the box dimension changes when you had it in.
I will also play around with hornresp and vituixcad to refresh my memory of what various port options do.
gabo
Since you are going to flush mount the whole speaker into the wall, you can just make a box for the woofer, without the tweeter + midrange. Then you are way more flexible about the port position.
The x-over will have to be modified for your infinite baffle geometry. I would recommend to think about the DSP approach again. This is not going to be a trivial task.
The x-over will have to be modified for your infinite baffle geometry. I would recommend to think about the DSP approach again. This is not going to be a trivial task.
Gabo,
Think about it. I can make a digital xo design for a Monkey Box speaker version with the transducers mounted on infinite baffle (flush mounting of the cabinet with a wall). For example, miniDSP can be chosen as the digital platform, I already have practical experience with it.
I don't think it is a good idea to mount the current version of the Monkey Box flush with a wall and do some EQ at low frequencies. A new design is much more accurate.
It is possible to fit the SPL of the new speaker on the SPL of the last version of the Monkey Box design (for the upper frequency part).
A new passive filter for the infinite baffle version is also possible.
Adding some sub with a digital filter to realize a 4-way can be an option too.
From your side you must be able to measure some SPL and impedance curves/parameters to make a design that fits on your own transducers.
Last edited:
Thanks Paul. I have no problem doing any of that. I have an earthworks mic, ARTA, and everything I need to do measurements. I also have the mid/tweet drivers here, but not the woofer just yet. I was still floating all this round in my head.
My first thoughts were to build the tower version, in my listening space that would be preferred over putting them on stands. To do that right requires essentially a complete redesign, which is not totally out of the question. But I also already have a killer set of towers in that room anyway.
But then, the recording studio where I do a lot of session work, is moving and building a new studio. Which I am doing a lot of the design for, and the owner asked about a new monitoring environment. The monitors there will be flush mounted into the wall. So we talked about using these speakers for that purpose. They haven't even broke ground yet for the studio, so it's going to be a while before we need them.
All this is background just to say that regardless of the build, even if I build the original speakers, I will do a lot of measurements and tweaking to get my specific speakers the way I want them. So I have no problems with that.
However, none of it will happen quickly. Since our whole country is quarantined for the virus, would be a great time to do some of these measurements. However, even all the places to buy a piece of plywood are shut down, so that's not going to be possible either. I might be able to conjure up enough stuff in my shop to build some sort of not-completely-infinite baffle, so maybe I can do a few measurements for that. I can scratch around and see what I can come up with.
Actually now that I think through it, I might be able to cobble together a baffle that is a reasonable mock up of the one we might use in the studio. So that might actually work. But for now without woofers available.
Also, if we want to take this to another thread, or even offline, that's cool too. And I have no problem doing it all by myself either, so only work on this if you have interest in it. It's always great to have help and input, and I welcome it, but I also have no problems doing things myself.
gabo
Thank you Paul, I will do this, but I'm going to wait until I'm ready to work on it. I have several big projects going on right now. I'll let you know when I start it.
gabo
Hello,
To make an infinite baffle design of the Monkey Box, the horizontal on and off axis SPL response measurements of the midrange and the tweeter on infinite baffle are needed, as these drivers are horn loaded.
In the simulation I cannot make good transducer infinite baffle models if they are horn loaded. So I need the measurements to make a correct design.
Deriving the infinite baffle responses out of the Monkey Box measurements is too complex and probably not accurate enough. I already looked to that.
So some measurements of the drivers on a large IEC baffle or of a prototype are needed.
Paul
To make an infinite baffle design of the Monkey Box, the horizontal on and off axis SPL response measurements of the midrange and the tweeter on infinite baffle are needed, as these drivers are horn loaded.
In the simulation I cannot make good transducer infinite baffle models if they are horn loaded. So I need the measurements to make a correct design.
Deriving the infinite baffle responses out of the Monkey Box measurements is too complex and probably not accurate enough. I already looked to that.
So some measurements of the drivers on a large IEC baffle or of a prototype are needed.
Paul
Last edited: