advertisement


Tannoy Chatsworth 12" MG rebuild advice.

Fair enough. But I can't keep the positioning the same, as the baffle will be a completely different size.
BTW, I loaded Edge and can't get it to work, Same old, same old. Me, that is. :(
 
I'd just extend the depth of the cabinet as much as you can get away with, but that's me.
Extending the the height of the baffle by extending the bottom would probably be the least damaging thing you can do.

It's changing the baffle width or driver height relative to the top of the baffle that's the most damaging.
 
Did you know that Tannoy used exactly the same crossover for both the 12" and 15" Monitor Golds? In different cabinets? That's a large part of why I have my doubts about them incorporating baffle width calculations in their designs.
Anyway, I will making mine 50-52 cm wide, which is quite close to their various widths. The distance to the top edge is something I will have to look at though.
Paul (Reffc) has added some helpful comments over on Lenco Heaven, particularly about tuning, so I hope to do some drawings and volume calculations soon. Old school, with pencil and paper, and a calculator. I know how to use them! :D
I do appreciate your help, otherwise I would be struggling even more. I grew up in a world without computers, and resisted them as long as I could, so now I find I can't catch up. My working life has been largely spent doing things like signwriting, glass etching and making leaded lights, or in joinery. Hands and eyes, no brain needed. Now it doesn't seem to work! :(
 
Did you know that Tannoy used exactly the same crossover for both the 12" and 15" Monitor Golds? In different cabinets? That's a large part of why I have my doubts about them incorporating baffle width calculations in their designs.
Anyway, I will making mine 50-52 cm wide, which is quite close to their various widths. The distance to the top edge is something I will have to look at though.
Paul (Reffc) has added some helpful comments over on Lenco Heaven, particularly about tuning, so I hope to do some drawings and volume calculations soon. Old school, with pencil and paper, and a calculator. I know how to use them! :D
I do appreciate your help, otherwise I would be struggling even more. I grew up in a world without computers, and resisted them as long as I could, so now I find I can't catch up. My working life has been largely spent doing things like signwriting, glass etching and making leaded lights, or in joinery. Hands and eyes, no brain needed. Now it doesn't seem to work! :(


The measurements I've seen of the 15" Monitor Gold look really nice, with peaks and dips in all the right places. The measurements look far better than any of the latest Tannoy Legacy series to me (on and off-axis). It's obvious to me that real passion went into those older Tannoys. I don't believe for one minute that no thought went into the baffle size and driver position. If Tannoy managed to get the same crossover to work seamlessly with two different drivers, in different cabinets, then I think it shows how bloody good Tannoy were.

Looking at simulations - with a 4" wider baffle and the driver in the same position, you'll lose a touch of energy around the presence region, and gain a bit in the midrange (This is a fact). It should be audible but how much is hard to say. I know that when I'm adjusting the crossover on my own speakers, a decibel over a certain range makes a big difference to my ears, but I suppose some people may not notice or care. BUT, if you raise the driver a couple of inches towards the top of the baffle, it almost matches the frequency response you have now.

To have the best chance, taking measurements of your Chatsworths before you start the build would be the sensible thing to do. You could even temporarily fix something to the sides, to see what the measured difference looks like. Part of your problem is not knowing what the frequency response is like as standard, let alone with the driver being out of spec (20hz fs). Imagine the measured response shows a nasty peak (or dip) right where you least want it. A quick simulation with The Edge would show you if it's possible to remove that peak with a slight adjustment of the baffle width, or a change of driver position on the baffle.
 
Ex my speakers, watching with interest.

There’s a 12” Gold Lancaster which looks wider?
I found it interesting Chatsworth the only sealed Tannoy floorstander, I’d found pics of corner cabs which looked similar volume, but taller. I felt the triangular shape was more rigid.
Another point that’s been mentioned is the wadding and back panel are up against the back of the driver. Done for a reason?
 
Grey line is the 16" wide x 33" high Chatsworth, Blue line is the 21.5" wide x 33.5" high Lancaster (specs are from a quick Google search so could be wrong). This is pretty much what you'd get before the crossover flattens the frequency response, and is just showing the relative difference. About 1db but over a wide bandwidth.

IpmQC8jl.jpg
 
Ex my speakers, watching with interest.

There’s a 12” Gold Lancaster which looks wider?
I found it interesting Chatsworth the only sealed Tannoy floorstander, I’d found pics of corner cabs which looked similar volume, but taller. I felt the triangular shape was more rigid.
Another point that’s been mentioned is the wadding and back panel are up against the back of the driver. Done for a reason?
Yeah, to get the damn things into a cabinet that small! My opinion, of course.
Last comment (honest!) on baffle diffraction . The Chatsworths, in particular, have so much nonsense going on at the the edges of the cabinets, I can't imagine how you could model that. As far as I can see, the Edge assumes a flat sharp corner, not mouldings all over the place. But I am probably wrong. Anyway, If it's not too time consuming I would like to create a useful round-over edge.
Some minor progress; replacing the horrible 4 pin connectors, necessary as one of them was broken internally, so a pair of spare locking connectors installed. And the crossovers re-capped and re-resistored. It all just fits into the original crossover boxes. I will probably redo them after the cabinets are made. Still working those out, but it's looking like 135 litres actual volume, 30Hz tuning, two ports, giving a gentle shelving down to resonance. No peak. But removable lower fronts for altering it all. I've been getting some advice from Paul on Lenco Heaven, and of course Fatmarley here. So I'll have at least two people to blame! :D
51018829976_f84fb643e6_z.jpg


51019875671_fe269299b9_c.jpg

And I just rang Lockwood Audio, who said that rubber surrounds are not available for these, so I will have to suck it up, and resolve any FR deficiencies through the crossover. More fun!
 
The Chatsworths, in particular, have so much nonsense going on at the the edges of the cabinets, I can't imagine how you could model that. As far as I can see, the Edge assumes a flat sharp corner, not mouldings all over the place.

I wouldn't worry about the mouldings. I was trying to show you that different baffle widths will give you a different frequency response. There's obviously one that going to measure with a more linear frequency response than the other, but you won't know which without taking your own measurements.
 
Last edited:
Well, here's a thing. I replaced one 4 pin socket, and on that one the left wires are red. Which I thought was odd, as the they are cross linked in the plug. I would have expected the black wires to be commoned. Well, whatever.
51018829976_f84fb643e6_c.jpg

But on the second speaker the left wires are black! I haven't looked at the plug yet. I had to stop at that point, but it's bugging me now. I can't really look at it until tomorrow, but IIRC, the bass wires are not crossed over, so they will both be wired in the same phase.
But I suspect one tweeter may be wired wrongly! Mark measured the response of these, and had a dip at 1350Hz on one, so maybe that is what caused it? It means I will have to open up the first one again and compare them. Irritating.
49917548701_b474e6438f_o.jpg
 
I finished the other crossover, and checked the socket wiring again. Never mind the wire colours, but the tweeters are wired oppositely on each speaker! I checked bass cone movement with a 1.5V battery, and they are both the same, so one tweeter must be wrong. According to Paul, this is fairly common after they have been fiddled with. A bit of rewiring needed then.
 
The best way to be sure of the correct polarity would be a frequency response measurement at around 1mtr (assuming an even order crossover?).

If that frequency response above is correct, and it really is down around 2db at 100hz, I'd seriously try a 4" diameter x 7" port in the Chatsworth cabinet before building anything (assuming the cabinets not worth much?). With room gain you may be surprised at how much deep bass you get. With a bigger cabinet, all you're doing is trading midbass for deeper bass, so if the midbass sounds fine with a ported 75ltr cabinet, going bigger will just make the midbass weak and give you a bit more low end extension.
 
Last edited:
I finished the other crossover, and checked the socket wiring again. Never mind the wire colours, but the tweeters are wired oppositely on each speaker! I checked bass cone movement with a 1.5V battery, and they are both the same, so one tweeter must be wrong. According to Paul, this is fairly common after they have been fiddled with. A bit of rewiring needed then.
If one tweeter was wired incorrectly I'd have expected to see a much bigger FR variation between the two drivers around the XO frequency than what's shown
in the above graph. I'm also somewhat concerned by what looks to be as much of a 5dB difference in sensitivity between the two tweeters from 2kHz-10kHz. Have you swept them yet?
 
I Have you swept them yet?
No. Not really possible in the loft. But I did test the relative response (sig gen and Tandy SPL meter) and one had a dip. Rewired and dip gone.
So nothing to do until I make new cabs, just run in the crossover caps. I'm not going to butcher the Chatsworths; they are so flimsy it wouldn't be worth the effort. Made down to a price, IMO.
BTW, they have treble "energy" controls, so I am not too bothered by the level difference.
 
I've just had a quick draw, and for a cabinet with a 500mm wide baffle, 320 back, 430 depth, all external, curved sides (max width 530mm) and height 1200mm, with 24mm panels, 30mm front, I get 160 litres. Add 3 cross braces and allow for the back of the speaker and crossover, I still get over 150 litres. Add felt pads, etc., and I'm still above the target of 135 litres. Generous round-overs for the baffle and corner bracing would leave me room to manoeuvre. So at least I can price up the materials. It will have to be the birch ply from my local timber yard, particularly as I get a trade discount there. Although Baltic birch is preferred, what does that mean these days? In the '70s Russian and Finnish birch were very different, but both could be called Baltic, I guess. So Chinese it is, then. :D
 
An idle thought. Has anyone removed the dust cap? Apparently it can be done carefully using acetone, and is considered worthwhile. By some! Of course, that assumes it is stuck on by Tannoy, and not a rebuild. Some more recent Tannoys don't have them, do they?
 
I have when following dodgy steps to re-foam my first pair of HPDs. It's a pain in the arse, leaves a visible glue ring, and makes no difference to the sound (imo). I put them back on again.

Probably fair enough to optionally leave them off if/when fitting new cones, but personally I wouldn't bother removing existing ones unless I had to.
 
I think you're right. Most of the modern ones seem to still have them, and I probably can't hear high enough anyway. :D
I've been fiddling around in WINIsd again, and found some more vent options. I can get a similar modeled response with two long 4" vents, one shorter 4" vent (at the risk of turbulence and chuffing) or even several slots! But balancing the theoretical effects with the probable actual effects takes more expertise than I have.
So, I plan to make the bottom section of the baffle removable so I can try the different options and listen to/measure their effects in the real world. And in my room. Plus, if I make the top part also removable, I can mount the driver in such a way that turning it around gives me two different positions, for baffle step/diffraction testing. And generous round-overs on the vertical edges to minimise the effects.
Lots to think about, plus the shape to allow for close to wall positioning, as there is a theoretical downward shelving of the bass.
 
Hi Awk, did you consider a corner cab that kinda eliminates most of the issues raised above.
There's a Chatsworth corner cab for inspiration, last of the 60s mono era, though it would benefit from a larger volume, too.
Considering vintage Tannoy, one has to bear in mind that the speakers were designed for mono, hence the corner design. Quoting Peter Thomas, owner of PMC about vintage speakers : "You'd have valve amps which had no damping factor at all. You had to design the drive unit in the cabinet to have its own damping, so its bass performance was determined solely by the speaker, not by any control coming from the amplifier".
 
My valve amp has an output impedance of 0.7 ohms. As I am running from the 4 ohm tap that will now be around 0.4 ohms. Which into 8 ohms will give a damping factor of 20. Not high, but not "no damping factor". And at resonance the impedance is around 40 ohms, so a DF of around 100.
Damping factor is determined by speaker, plus crossover, impedance over amp output impedance, as I understand it. Cabinet damping, particularly aperiodic, is something different.
From my limited reading, (and limited understanding!) the necessity for and desirability of, a high DF depends on the Q of the driver among other things. So a low DF may be less of a problem with my Tannoys than with more "modern" drivers.
 


advertisement


Back
Top