Open Source Monkey Box

...
You can clearly see that funny things happen at 250 Hz with the front/slot port open (red curve). No such resonance happens with the rear/round port, which is quite a bit shorter than the front/slot port. This indicates that the 250 Hz issue happens within the long slot port, not in the enclosure.

No, it does not.

What it indicates is: That the slot port throat is placed somewhere a 250hz resonance triggers! And I have warned about this since way back.
I will concede to almost any points in regards of filters, tweeters and midrange drivers. But not on the point of BR/equivalent type boxes and internal modes.
It is not the port itself, it is the port in relation to internal surfaces, and driver placement.
A mode does not trigger uniformly in a room, or enclosure.

Edit:
Everything has a resonance! But it depends on a number of various things like the structure, the volume, the temperature, the size, other structures in contact etc etc.
What you seem to be implying is that resonant behaviour in one point will be the exact same in a different point, on the same structure? *Does not compute.*


Edit2:
If you think my point is not valid, then you should reconsider your effort on baffle diffraction. It is almost the same thing happening, reflections from surfaces, distances, voids, influencing the way the sound is spread, only difference is: it is happening inside the box.

Edit3:
Have you tried putting a tubed port near center on the front?
 
Last edited:
What it indicates is: That the slot port throat is placed somewhere a 250hz resonance triggers!

You are right, my previous measurements cannot exclude this. I therefore placed the microphone inside the box at the entry to the front/slot port (with the rear/round port closed). Then I first measured the SPL with the front/slot port closed, and then again with the port open (see attachment). If the resonance arises within the main enclosure volume, the SPL around 250 Hz should be about the same in both measurements. That's not what I see in the measurements. The SPL is much higher with the port open. This looks to me like something is happening in the port. What do you guys think?
 

Attachments

  • SPL_port_entry_open_closed.png
    SPL_port_entry_open_closed.png
    40.6 KB · Views: 307
Last edited:
out of curiosity, how do you do the off axis measurements: do you use rotating table or do you relocate the mic (stand)? i am about to build largish speaker and was wondering how to actually do the measurements. thanks :)

Moving the mic stand is a PITA, so I just I rotated the speaker. You don't need a rotating table if you don't do this stuff on a daily basis.
 
For a tuntable you need two pieces of coated particle board "kalustelevy", a drill and a wooden pin or bolt. With some felt pads in-between rotation gets easier. If you have a jigsaw, cut the upper board to half-circle. The plates must be big enough so that a speakers front baffle can be positioned on rotating axle, my plates are 70x70cm.

Here I put and extra board on grass under the turntable.
 

Attachments

  • turntable.jpg
    turntable.jpg
    60 KB · Views: 307
Thanks! was reading on the subject and many references point that all objects including floor should be as far as possible/practical while measuring. this led me wondering how to actually elevate large speaker and rotate it = large rotating table with some large pedestal for the speaker i guess. Google finds lots of images of small speakers being measured with such setup, but not single large :) I guess y axis measurements are done the speaker laying sideways on the rotating table? Horn at the other end of the meter high baffle the pedestal needs to be huge to get the horn to the rotating center.
 
Last edited:
You are right, my previous measurements cannot exclude this. I therefore placed the microphone inside the box at the entry to the front/slot port (with the rear/round port closed). Then I first measured the SPL with the front/slot port closed, and then again with the port open (see attachment). If the resonance arises within the main enclosure volume, the SPL around 250 Hz should be about the same in both measurements. That's not what I see in the measurements. The SPL is much higher with the port open. This looks to me like something is happening in the port. What do you guys think?

I do not dispute that it does not happen, or that the port does not have some influence the resonance. But it is because the port is placed where it is. It is not caused by the port itself.

Have you tried tube port on the front vs the back yet?

I think it's very interesting to see what happens when you measure stuff like this, thank you for investigating the issue.
 
Last edited:
Thanks! was reading on the subject and many references point that all objects including floor should be as far as possible/practical while measuring. this led me wondering how to actually elevate large speaker and rotate it = large rotating table with some large pedestal for the speaker i guess. Google finds lots of images of small speakers being measured with such setup, but not single large :) I guess y axis measurements are done the speaker laying sideways on the rotating table? Horn at the other end of the meter high baffle the pedestal needs to be huge to get the horn to the rotating center.

You got it! I use 1-2 kitchen ladders with the turntable for small speakers.Turntable on top of the first ladder, speaker on top of second ladder. Kitchen ladder is a very good stand, put steps on the backside of the speaker to minimize reflections.

Vertical is trickier because distance varies by rotation easily. To get data for simulation, measure each driver at the time and put that on rotation axis. Simulator eg. Vituixcad takes care of rest. If you want to check how a finished multiway works/interferes, you must determine the location of listening axis (=0) and set that at rotation axis.

I use only dsp, no simulators. I took verticals of the finished speaker by changing the height of the microphone. Obviously you must use your creativity to find a solution!
 
Last edited:
Nope. Not enough space on the front. Installing the tubes I have (or other ports of acceptable size) would mean I'd have to destroy the tweeter or the midrange driver.;)

... What is this? "My port is bigger than your port" kind of thing? :D
I do not mind using a tube port if that is one of the compromises the design calls for. And moving the slot port, or the 12", is apparently not a popular thing to do.

... But to have the port exit on the back?
I would at least like you to try front and back, just to compare.
Never heard a back ported speaker I could "fall in love with", but then again, one time has to be the first...
 
So far I have no preference between round tube or slot. I also do not (yet) have a preference with respect to placing the port on the front or the rear. I will need to do some listening tests.

The only constraints I have at the moment is that I don't want to place the port right next to the mid or tweeter for diffraction reasons. Also, the space available on the front panel for a port with a largish cross section is quite limited (see drawing in post 346).
 
I do have a preference for slot ports, but if it's not "done right" then there's no point.
It becomes a wasteful sacrifice on the altar of compromises.
A tubed port's only redeeming factor is that it is more flexible in terms of placement IMO. Make a hole and attach port vs. carefully plan the design in detail and build it exactly to spec.

What dimensions of tube port are you contemplating? Maybe the design calls for a slimmer, shorter tube?
Could be placed close to the mid and woofer, how much space could be used?
Yet another sacrifice, but it can still turn out nice.

Edit:
Another option could be to use something in the neighbourhood of the Beyma 12BR70, and make a closed box instead. Might be more sensible if a port cannot be used. I really like the 12BR70, sounds very nice, cheap, cast alu basket, fits really well in a closed box, what's not to like?
 
Last edited:
mbrennwa; I was reading about MLTL enclosures, which at first seems to be a tall ported box but is actually a quarterwave resonator of sorts. Here is a quick read for you:
Designing a Mass Loaded Transmission Line speaker.

Might get you on the track regarding the weird resonance if you read some MLTL info / threads. Sorry, I don't know much about them since have been reading half an hour.
"Driver offset -- In a straight TL, moving the driver down approximately one third of the line will eliminate half the peaks in the higher harmonics. This is a very useful, old trick to tame the frequency response. The cost is slightly less bass." -http://www.t-linespeakers.org/design/MJK-for-dummies/index.html

There is somewhere a MLTL designer where you get to input box, driver and port placement and it'll show you the resonances but haven't found it yet :)
 
Last edited:
I do have a preference for slot ports, but if it's not "done right" then there's no point.

I was busy in the workshop all day, working on non-loudspeaker stuff. The Monkey Coffin was playing music all the time, and I switched between the front/slot and rear/round ports a few times. It was great to have such great sound in the workshop all day long!

Some very quick-n-dirty impedance measurements suggest that the new rear port results in a tiny little bit higher tuning frequency than the front port (I will do some proper measurements later). My conclusion from a day of listening to both bass-reflex versions is that the rear port sounds better to my ears. The bass is a bit stronger, but still nice and "dry", very much like with the front port. No traces of the typical "bass-reflex boom". I guess the difference I am hearing between the two ports is mainly the slightly different bass-reflex tuning, not necessarily the shape of the port (slot vs. round). Also, the colorations of the upper bass / lower midrange that I got with the front port are gone with the rear port, which is in line with my earlier measurements related to the "250 Hz issue". The speaker can go pretty loud before the bass-reflex port starts "locking up".

What dimensions of tube port are you contemplating? Maybe the design calls for a slimmer, shorter tube?

I am aiming for a large cross section to avoid the port from "locking up" when it gets loud. With a large cross section, the velocity of the air in the port is lower, which helps to keep (non-linear) turbulent losses low. The front/slot port was pretty nice in this regard. The cross section of the Monacor BR-100HP I used for the rear/round port is quite a bit smaller, but still okay. I don't want to make the cross section any smaller though.

Another option could be to use something in the neighbourhood of the Beyma 12BR70, and make a closed box instead.

We discussed the woofer choices a long time ago, and the Faital 12PR320 came out as the best candidate for the Monkey Coffin. There is no need to go back and start from scratch, as the bass-reflex system does work very well. It's just a matter of tweaking the details of the bass-reflex port.
 
mbrennwa; I was reading about MLTL enclosures, which at first seems to be a tall ported box but is actually a quarterwave resonator of sorts. Here is a quick read for you:
Designing a Mass Loaded Transmission Line speaker.

Might get you on the track regarding the weird resonance...

Yup, the work of Martin J King is the foundation of some of the more recent box modeling software tools that are based the "acoustic impedance" approach (like Hornresp or AJ Horn). Adopting this view, the Monkey Coffin could maybe be considered as a woofer in a large pressure chamber (the main box volume), which is coupled to a transmission line (the bass reflex port). This might explain why the longer port triggers the "250 Hz issue", whereas the shorter port does not. But that's just a wild guess...
 
I think this is what Kaffiman was thinking, see MJK quote below. The monkey box might be tall enough to have such pressure nodes inside the enclosure that driver and port location would matter.

Carry on :)

"
There is a difference between a ML TL (or ML TQWT) and a classic bass reflex box.

For a bass reflex box typically the internal dimensions of the enclosure are reasonably close in value, the longest dimension is probably much less then twice the length of the shortest dimension. In this case the air in the box is compressed uniformily to act as the spring while the air in the port moves as a slug to become the mass. The spring and the mass of the bass relfex box determine the tuning frequency.

In a ML TL box the internal dimensions are very different. Usually the height of the enclosure is multiple times the other two dimensions. In this case the air in the box is not compressed uniformily and the pressure varies like a quarter wavelength (a quarter cosine wave) with the max pressure being generated at the top and the pressure decreasing as you move towards the port at the other end. The air in the port again acts like a slug to become most of the mass at the tuning frequency. I say most of the mass because there is a portion of the moving mass in the standing wave that forms the spring. So the tuning frequency of a mass loaded transmission line is lower than what would be calculated using the simplifed lumped parameter model of a bass reflex enclosure.

There is no black and white answer defining what geometry behaves as a bass reflex and what behaves as a ML TL, as you stretch one dimension a transitions starts to occur toward ML TL behavior as one dimension of the box gets longer and longer. If you run my ML TL MathCad model, starting with a volume and port from a bass reflex design as generated by a lumped parameter model, and stretch the height from a geometry that is short and fat maintaining the same volume as it becomes tall and skinny you will see the low frequency cut-off dropping a few Hz.

There is a difference and a designer needs to recognize when to move from a lumped parameter model to a more sophisticated model to avoid measured results that do not agree with predictions."

by Martin @ ML-TQWT vs Bass Reflex -

Techtalk Speaker Building, Audio, Video Discussion Forum
 
Now we're getting there, at least in terms of mutual understanding.

At any rate, I do not think it is good to compare a poor implementation of a slot port vs a good implementation of a tubed port.
The result gives itself even before the test has begun...

And I wish that if you where comparing tubed port placements, that there where equal ports on the front and back, to give a comparable frame of reference.
 
Remember we are not aiming for a full blown study of how to design a bass reflex system in general. We are looking for the best option for the Monkey Coffin.

So far, the long slot/front/bottom port created a problem which the shorter round/rear/center port did not have. A front port next to the mid/tweeter is out for reasons of space and diffraction.

What other design options do you suggest? Side port?