Had the bass drivers out yesterday so that I could experiment with some more bracing. Basically the insides of the cabinets now look like a Kerplunk tower ready for the balls to be dropped in. Into each cabinet I fitted 2 more side-to-side braces and 6 more front to back braces. Significant effect on sound. The midrange dropped away. So much so that I found it best to completely eliminate the L-pads on the midrange drivers and further trim back the tweeters a little. Now the cabinets are sounding tighter and more focussed. Stereo image is improved. Sounds more distinct; purer. With biggest improvements to upper bass and midrange. Fascinating to see how the cabinet and drivers interplay with each to produce an overall sound with the the room.
I've been idly toying with the notion of getting my loudspeaker cabinets rebuilt by someone with access to CNC facilities, is that something you might get done? I doubt it is a cheap thing to do, but I really think it could benefit designs with strangled potential. Better materials, proper bracing, curved sides etc.
For magisters I have long thought of side firing woofers at floor level (a la audio physic) with a slim front baffle, perhaps with a decent cone midrange and tweeter upgrade (ATC or Ruark). The whole thing would need calculating for correct internal volume as the are infinite baffle, and would likely stand a metre and a half tall. In gloss piano white. Crossover design by our Col it even RFC, and I bet they would image nicely!
I've been idly toying with the notion of getting my loudspeaker cabinets rebuilt by someone with access to CNC facilities, is that something you might get done? I doubt it is a cheap thing to do, but I really think it could benefit designs with strangled potential. Better materials, proper bracing, curved sides etc.
Thanks Alan. :-) I have learned from my recent, ad-hoc, experiment with bracing, how cabinets can effect sound. So I would be interested to learn more about different cabinet designs and constructions. As you suggest, things like CNC are expensive. My going about this from-scratch DIY build has been contingent on it being affordable to me, especially given that the speakers may have turned out to be rubbish (thankfully they haven't). Also, I'd have to learn how to do it. But, if I make another pair I may experiment with different materials and perhaps get a carpenter friend to help with more substantial bracing at the outset.
For magisters I have long thought of side firing woofers at floor level (a la audio physic) with a slim front baffle, perhaps with a decent cone midrange and tweeter upgrade (ATC or Ruark). The whole thing would need calculating for correct internal volume as the are infinite baffle, and would likely stand a metre and a half tall. In gloss piano white. Crossover design by our Col it even RFC, and I bet they would image nicely!
I do like white speakers. Piano white would be classy. :-) You mention side firing woofers. Yes, this is something I have considered to allow for a narrower front baffle. In fact, I have thought about trying it with my DIY speakers (simply rotating them 90 degrees and drilling holes in the sides for the mid and tweeter. But I think I'll leave it for a future build. Don't want to spoil the existing cabinets).
I agree about replacing the midrange on the Magister. The standard one is not the best (not sure if it's just its age or it whether even a new one would be poor). I replaced them with a sealed back cheapo Pyle driver and the improvement was massive. There's no internal enclosure for the midrange on the Magister, so unless one fits an internal enclosure (which would effect the volume of the woofer enclosure) it's got to be a sealed back midrange unit to avoid the pressure changes from the woofer effecting the midrange cone. I think I saw recently that Monacor now produce a sealed back unit. May look at that.
Regarding internal volume, I agree it's important to get this right. For my DIY build I found on-line an old manual for Goodmans drivers that included information on the 18P that I used. It gave a clear figure for internal volume, which I used as one of the starting points for the cabinet design. IIRC the manual also stated that it was the volume that mattered, not so much the specific dimensions that determined it. If one wanted to rehouse a 15 inch Magister woofer, I suppose it would be fairly easy to calculate the internal volume of a standard Magister cabinet and then work out how slim the front baffle of a replacement cabinet would need to be if a metre and a half tall and 15 inches deep.
I'd be interested to read any speaker build you do undertake Alan. I know we have similar aural tastes, so am always keen to read of your experiences.
Following the bracing of the cabinets I decided to introduce some sound dampening to the inside of the cabinet walls of the DIY 3-ways. At specific frequency (chesty male voice in movie dialogue Hz) there was a slight resonance from the cabinets.
In the past I've used this sort of thing inside speakers (some old Wharfdales I was mucking about with). I think I remember seeing it inside some Dynaudios I once owned, so reckoned it was probably OK. BUT, this time I wanted to try something different. I decided to go for some car sound proofing. It was cheaper than the audio stuff. (Surprise surprise.) Thinner than the foam egg-boxes. Self adhesive and with a silver foil face. (Possibly thermal and sonic...?) Previous buyers reported good sound-deadening results (when fitted to the doors and sills of their cars :-D ). I ordered a couple of packs. When they arrived i found the foam used in these is higher density than the audio stuff above.
Of course I underestimated the amount and the 2 packs didn't cover all the inner the surfaces. It had an effect though. The resonance around male voices had gone. Replaced by an even less frequent (but still dissatisfying) resonance from the female actors' voices. I put up with this for a while. But last week decided to buy a third pack of the above to complete the job. Pleasingly this seems to have done the trick.
At the same time I also properly glued the speaker gasket that I had previously just rested in place. (The Goodmans Audiom 18P were intended to mounted behind the fascia and so have a built-in foam gasket on their fronts. But mounting behind looks a bit shit, so I had to buy new one to go behind the drivers, front mounted on the fascias.)
The whole sound is (even) clearer. Imaging is better and bass is (even) more defined. As I did both jobs at the same time, I'm not sure whether the clearer sound is down to the soundproofing or the better sealed bass driver. Or both. Best thing is that this is the first time I've tweaked the speakers and I haven't needed to adjust the L-Pads on the mid and tweeter to compensate for some subjective sonic change to frequency levels. They just sounds better. So that's a bonus! :-)
When I first built the speakers I gave little thought to the cabinet. "It's just a box!" How naive. Probably the steepest learning curve for me has been how influential on sound even a simple cabinet is. Retrospectively bracing and dampening the cabinet has been a bit of a pain (fortunately the hole for the 18 inch driver makes for a decent access point). If I do a build again I will certainly invest more time and energy on making the cabinet as rigid and inert from design-day 1. On a more self-congratulatory note, when I first conceived of these I did not imagine getting such pleasing results. The build was just a bit of hobbyist fun. Now I absolutely love them and the revamped Goodmans Goodwoods and Magisters don't get a look-in in the main system. Think I've spent about £800-900 (and a bloody lot of hours) on them altogether. #chuffed
Loudspeaker cabinet construction, and vibration control in general is very interesting. I've been reading some thoughts on the subject recently by a very experienced engineer, and it's way beyond my grasp!
Thanks guys. I have tried to further educate myself about the technical details, but yeah, beyond me too. I understand the principle of reducing cabinet vibration as far as possible, but that’s about it.
What Colin says is interesting, and not surprisingly fits in with all the good material I've read. You need two main things, one of which you have: a barrier to transmitted sound, the lead, or sound deadening sheet. This will also function as dampening material to the sidewalls and is especially effective if sandwiched between two layers of cabinet material. You also need absorption, to take sound power away (sheep skin - with fleece).
I was thinking about screwing additional mdf sheets around the cabinets. They would be enormously heavy! Would be interesting to sandwich sone insulation in between. (So not merely wool, but sheep flesh too...?)
I was thinking about screwing additional mdf sheets around the cabinets. They would be enormously heavy! Would be interesting to sandwich sone insulation in between. (So not merely wool, but sheep flesh too...?)
No use CD,s and LP,s of Ugly Music and spare the world of despair and pain. Rescue us Doc.
Well I'd suggest that a second material ought to have different properties to the first. So if you used MDF for the cabinet, go for HDF or birch ply. Sandwich dampening material between, apparently this should be soft. Birds custard should do.
Then fleece or foam inside.
I bought fleece on Colin's advice when I fettled the Triangles, but misunderstood. I bought a bag of fleece rather than a hide, which is what he meant. I had to stitch bags together out of mesh to hold it.
I'll go off and look up what I was reading, and link to it.
You want to pay attention to an experienced engineer named Frank Dernie. He posts in both threads. Bonus points if you work out who he is and what his engineering pedigree is. 😉
Well I'd suggest that a second material ought to have different properties to the first. So if you used MDF for the cabinet, go for HDF or birch ply. Sandwich dampening material between, apparently this should be soft. Birds custard should do.
Then fleece or foam inside.
I bought fleece on Colin's advice when I fettled the Triangles, but misunderstood. I bought a bag of fleece rather than a hide, which is what he meant. I had to stitch bags together out of mesh to hold it.
I'll go off and look up what I was reading, and link to it.
I think what’s called for is a new build where I can consider all of these points from the outset. For a while I’ve been thinking about possibilities for a build in the summer when I have more time.
I like the cavity wall approach and would like to try. How is the cavity best spaced and fixed do you think? And to what depth?
Just noticed that the revamped goodwoods that I put in the shed are getting a bit of cabinet water damage, so I might use the drivers and new crossovers from those for a new super-cabinet build.
Coming from a zero knowledge pov, with a cavity wall construction you'd use a material "that matters" for the inner and could use pretty much anything for the outer. Leaving the way open to get all creative on looks.
Not forgetting you'll use less heating in your new drive units home.
Not cavity as such, rather sandwich (mmm). Two different materials, with a gooey constrained damping 'filling'.
Take a look, especially at the first four or five pages, of the linked thread relating to Harbeth. I think that's where the best info was.
I certainly learned a lot about cabinet behaviour I wasn't aware of.
I’ve been through the first 10 pages. So far Alan Shaw has been quoted as saying thin (9mm) walls are likely to create less resonance issues than 18mm ones. And further to this, that lining 9mm cabinet walls with bitumen (or similar) will also help. interesting. Also a link to Toole’s thoughts, that I will try to make time to pursue later. I haven’t read anything about sandwiching yet.
Here is a very excellent Tannoy build thread on pfm. I think the builder has likely read many of the same resources I referred to, and the too-and-fro about fixing tightness is rather interesting as well. Just four pages with pics.
Excellent tip to use PU18. I don't know why I didn't think of this, I've used it for other things in the past. All in all, a very fine DIY implementation of constrained damping and minimal bracing in a loudspeaker cabinet that should be easy to copy.
Thanks Al. That was useful. And not overwhelming for me.
I think I’ve got a good idea for how to go about things when the time comes. Also I can use the 3D workshop at college, so will have access to better tools than I have at home.
Comments
Basically the insides of the cabinets now look like a Kerplunk tower ready for the balls to be dropped in. Into each cabinet I fitted 2 more side-to-side braces and 6 more front to back braces.
Significant effect on sound. The midrange dropped away. So much so that I found it best to completely eliminate the L-pads on the midrange drivers and further trim back the tweeters a little.
Now the cabinets are sounding tighter and more focussed. Stereo image is improved. Sounds more distinct; purer. With biggest improvements to upper bass and midrange.
Fascinating to see how the cabinet and drivers interplay with each to produce an overall sound with the the room.
But it does keep me off the streets.
I've been idly toying with the notion of getting my loudspeaker cabinets rebuilt by someone with access to CNC facilities, is that something you might get done? I doubt it is a cheap thing to do, but I really think it could benefit designs with strangled potential. Better materials, proper bracing, curved sides etc.
For magisters I have long thought of side firing woofers at floor level (a la audio physic) with a slim front baffle, perhaps with a decent cone midrange and tweeter upgrade (ATC or Ruark). The whole thing would need calculating for correct internal volume as the are infinite baffle, and would likely stand a metre and a half tall. In gloss piano white. Crossover design by our Col it even RFC, and I bet they would image nicely!
I have learned from my recent, ad-hoc, experiment with bracing, how cabinets can effect sound. So I would be interested to learn more about different cabinet designs and constructions. As you suggest, things like CNC are expensive. My going about this from-scratch DIY build has been contingent on it being affordable to me, especially given that the speakers may have turned out to be rubbish (thankfully they haven't). Also, I'd have to learn how to do it.
But, if I make another pair I may experiment with different materials and perhaps get a carpenter friend to help with more substantial bracing at the outset.
I do like white speakers. Piano white would be classy. :-)
You mention side firing woofers. Yes, this is something I have considered to allow for a narrower front baffle. In fact, I have thought about trying it with my DIY speakers (simply rotating them 90 degrees and drilling holes in the sides for the mid and tweeter. But I think I'll leave it for a future build. Don't want to spoil the existing cabinets).
I agree about replacing the midrange on the Magister. The standard one is not the best (not sure if it's just its age or it whether even a new one would be poor). I replaced them with a sealed back cheapo Pyle driver and the improvement was massive. There's no internal enclosure for the midrange on the Magister, so unless one fits an internal enclosure (which would effect the volume of the woofer enclosure) it's got to be a sealed back midrange unit to avoid the pressure changes from the woofer effecting the midrange cone. I think I saw recently that Monacor now produce a sealed back unit. May look at that.
Regarding internal volume, I agree it's important to get this right. For my DIY build I found on-line an old manual for Goodmans drivers that included information on the 18P that I used. It gave a clear figure for internal volume, which I used as one of the starting points for the cabinet design. IIRC the manual also stated that it was the volume that mattered, not so much the specific dimensions that determined it. If one wanted to rehouse a 15 inch Magister woofer, I suppose it would be fairly easy to calculate the internal volume of a standard Magister cabinet and then work out how slim the front baffle of a replacement cabinet would need to be if a metre and a half tall and 15 inches deep.
I'd be interested to read any speaker build you do undertake Alan. I know we have similar aural tastes, so am always keen to read of your experiences.
At specific frequency (chesty male voice in movie dialogue Hz) there was a slight resonance from the cabinets.
In the past I've used this sort of thing inside speakers (some old Wharfdales I was mucking about with). I think I remember seeing it inside some Dynaudios I once owned, so reckoned it was probably OK.
BUT, this time I wanted to try something different.
I decided to go for some car sound proofing. It was cheaper than the audio stuff. (Surprise surprise.)
Thinner than the foam egg-boxes. Self adhesive and with a silver foil face. (Possibly thermal and sonic...?) Previous buyers reported good sound-deadening results (when fitted to the doors and sills of their cars :-D ).
I ordered a couple of packs. When they arrived i found the foam used in these is higher density than the audio stuff above.
Of course I underestimated the amount and the 2 packs didn't cover all the inner the surfaces.
It had an effect though. The resonance around male voices had gone. Replaced by an even less frequent (but still dissatisfying) resonance from the female actors' voices. I put up with this for a while.
But last week decided to buy a third pack of the above to complete the job.
Pleasingly this seems to have done the trick.
At the same time I also properly glued the speaker gasket that I had previously just rested in place. (The Goodmans Audiom 18P were intended to mounted behind the fascia and so have a built-in foam gasket on their fronts. But mounting behind looks a bit shit, so I had to buy new one to go behind the drivers, front mounted on the fascias.)
The whole sound is (even) clearer. Imaging is better and bass is (even) more defined. As I did both jobs at the same time, I'm not sure whether the clearer sound is down to the soundproofing or the better sealed bass driver. Or both.
Best thing is that this is the first time I've tweaked the speakers and I haven't needed to adjust the L-Pads on the mid and tweeter to compensate for some subjective sonic change to frequency levels. They just sounds better. So that's a bonus! :-)
When I first built the speakers I gave little thought to the cabinet. "It's just a box!"
How naive. Probably the steepest learning curve for me has been how influential on sound even a simple cabinet is. Retrospectively bracing and dampening the cabinet has been a bit of a pain (fortunately the hole for the 18 inch driver makes for a decent access point). If I do a build again I will certainly invest more time and energy on making the cabinet as rigid and inert from design-day 1.
On a more self-congratulatory note, when I first conceived of these I did not imagine getting such pleasing results. The build was just a bit of hobbyist fun. Now I absolutely love them and the revamped Goodmans Goodwoods and Magisters don't get a look-in in the main system.
Think I've spent about £800-900 (and a bloody lot of hours) on them altogether.
#chuffed
Loudspeaker cabinet construction, and vibration control in general is very interesting. I've been reading some thoughts on the subject recently by a very experienced engineer, and it's way beyond my grasp!
I have tried to further educate myself about the technical details, but yeah, beyond me too.
I understand the principle of reducing cabinet vibration as far as possible, but that’s about it.
(So not merely wool, but sheep flesh too...?)
Then fleece or foam inside.
I bought fleece on Colin's advice when I fettled the Triangles, but misunderstood. I bought a bag of fleece rather than a hide, which is what he meant. I had to stitch bags together out of mesh to hold it.
I'll go off and look up what I was reading, and link to it.
https://www.audiosciencereview.com/forum/index.php?threads/speaker-cabinet-design-considerations.11263/
Here is there other. Actually, this is best to read first.
https://www.audiosciencereview.com/forum/index.php?threads/harbeth-monitor-30-speaker-review.11108/
You want to pay attention to an experienced engineer named Frank Dernie. He posts in both threads. Bonus points if you work out who he is and what his engineering pedigree is. 😉
Baaah
For a while I’ve been thinking about possibilities for a build in the summer when I have more time.
Take a look, especially at the first four or five pages, of the linked thread relating to Harbeth. I think that's where the best info was.
I certainly learned a lot about cabinet behaviour I wasn't aware of.
So far Alan Shaw has been quoted as saying thin (9mm) walls are likely to create less resonance issues than 18mm ones. And further to this, that lining 9mm cabinet walls with bitumen (or similar) will also help.
interesting.
Also a link to Toole’s thoughts, that I will try to make time to pursue later.
I haven’t read anything about sandwiching yet.
https://www.pinkfishmedia.net/forum/threads/building-120l-cabs-for-12-tannoy-monitor-gold-rs.237172/
Here is a very excellent Tannoy build thread on pfm. I think the builder has likely read many of the same resources I referred to, and the too-and-fro about fixing tightness is rather interesting as well. Just four pages with pics.
Excellent tip to use PU18. I don't know why I didn't think of this, I've used it for other things in the past. All in all, a very fine DIY implementation of constrained damping and minimal bracing in a loudspeaker cabinet that should be easy to copy.
Will do a proper read over the next few days.
Thanks! :-)
That was useful. And not overwhelming for me.