Doc modding Marantz imperial 7

[karatestu]

From the picture in the last post you will see I am faffing about with the positioning of the sphere above the up firing mid bass. I had opted for right above the mid bass previously when I had three or four tweeters in a sphere because that is what sounded best with that configuration.

Since the tweeters are history (good riddance to tweeters and 1st order filters) then it is time to revisit this part of the design. I liked it directly above the mid bass because my tiny brain told me that it was helping to disperse the directional frequencies from a beaming driver (it runs out of steam at 7kHz approx) around the room and for the reason I could turn the tweeter sphere any way and adjust the tilt towards the ceiling.

I have gone back to having the tweeter roughly in line with the surround at the back of the mid bass' cone. I have had a single tweeter here previously, before I progressed through to four of the buggers. I now feel that the sound is better by not having the sphere above the mid bass. the bass is not diminished and the sound is more open. I will experiment with having the 2" cone at different points across the mid bass cone up to directly above the dust cap. There is going to be some reflections going on but in my quest to time align the two drivers as best I can then that is a price I will have to pay with this set up.

So, working out which position is best will have to be done by ear possibly with the help of a bit of science if I can't detect any difference, in which case I want to try and work out what is likely to be the best measured solution without measuring Maybe Rex can help here if he is feeling generous

I had thought that lining up the approximate position of the two voice coils would be good enough but I don't think it is. Most stuff written about time aligning two drivers is concerned with point and squirt not when you have the two drivers at 90 degrees to each other. When they talk about time aligning two drivers with a first order then there is a 45 deg phase shift with both low and high pass. It is my understanding that the low pass delays the sound whilst the high pass works a miracle and actually brings the sound forward in time before it has even happened. If that is the case then I require my 2" tweeter to be behind the centre of the voice coil of the mid bass because the mid bass has no low pass filter - it is running fully open. My high pass is a series line level polystyrene capacitor at the input of the power amp board.

Anyone make sense of that

[karatestu]

Believe it or not it has been listening to my JBL Charge 5 bluetooth speaker so much that brought about the dissatisfaction with the doped driver bipole with multiple tweeters. It does so many right and stops you thinking about the gear and just enjoy the music. That brought about the arrival of the 2" cone full range used as a single tweeter. What had I been doing messing about with tweeters pointing all around - I must be totally fick for not realising a tiny full range would solve my dissatisfaction with tweeters and their crappy distortion with 1st order filters.

So that led to trying my old B&W P4's to remember what an undoped driver sounded like. The little 2" injected so much get up and go into proceedings that I started to become suspicious of the doped cheapies. Very glad I did and tried a worm of blutack which has not only increased qts for better sealed box use but lowered Fs(lower bass) and dropped the sensitivity a bit to enable a perfect marriage with the 84dB 2".

The good bye product of listening to transmission line bass again has been I have been able to compare it to sealed box and my crap passive radiator experiment. The TL does some good things like increase the lowest bass and stops the driver having to move so much but those advantages just do not trump the tightness of closed box. End of

So I just have to sort out this time alignment business. I am not sure about trying a 3 way anymore with down firing bass and up firing mid even though it has some clear advantages. What it will have going against it is distance of bass and mid from each other plus the fact it needs more electrical filters. This simple 2 way only needs a high pass cap on the tweeter.

Going filterless where possible just sounds so right to me. The music just flows and sounds natural. This current version I have got to now does not really impress me in any way. It just sounds right, natural, energetic, vibrant but soothing and relaxing at the same time. I have listened to quite a few albums this morning and it has been one sonic wet dream after another. Currently playing The White Stripes Elephant and I am one happy bunny.

[antonio66]

One last experiment before the day is out. No farming to do Dave

I took the legs off which provided 15cm distance from floor to bottom of the cube. Mid bass was 45cm from the floor and the top of the tweeter sphere was 60cm from the floor.

Now they are resting on the floor so 30cm from floor to mid bass and 45cm to the top of the tweeter sphere. Excuse the usual mess of what is my work room / office.



I am currently listening to it. The bass has improved because of more boundary enforcement. Cancellations have probably improved, not sure. The reflection points in the room will have changed and both drivers are further from the ceiling which the up firing driver in particular relies upon. I can see and hear more of the up firing mid bass. I have obviously had to tilt the tweeter up more which is a luxury i have as I can more or less put it where ever I want. The music still doesn't sound like it is coming from so low down.

I must listen to this more. From a visual point of view then maybe some short 15cm legs would look better for what is going to be a BB8 look alike with two spheres one on top of the other. Bit of a trip hazard as it is now. Sounds ace though.

[karatestu]

It didn't take long to realise that sitting them on the floor works for the bass but not higher up. You can hear the forward firing driver's output in the cone of the up firing mid bass and that can't be heard when the enclosure is raised up. So legs are required after all.

There is a trade off going on here. The tweeter behind or above the mid bass definitely sounds much better. But there is some reflections between the two drivers that wouldn't happen if the tweeter was on the front like nva cubes. Imo the result is much better with it behind or above the mid bass even with the reflections.

Siegfried Linkwitz (RIP) solved it in the Pluto (below) by placing the tweeter at the front of the up firing mid bass and then applying delay to the tweeter. Also he crossed the two over much much lower than I do. I am not doing delay or crossing low so I have to live with it. There is some clever closed end transmission line backwave soaking up going on.



Next experiment is having the 2" cone above the dust cap of the mid bass. That may prove to be the best compromise especially when the tweeter is tilted up a bit. Tried putting some stuffing material over the mid bass but I covered the whole thing and it killed the sound a bit. Maybe a small amount under the tweeter will work. Will try that.

[karatestu]

Bit of a visual preview of what the basic shapes will look like. It will be finished properly of course and some short legs between 15 and 30cm long. The two steel hemispheres will be separated by a circle of baltic birch ply with holes in it. I will also make a mounting ring out of birch ply for the mid bass. It will enable me to rebate the driver and a roundover with a router will blend it into the curve of the sphere. I think the legs will be wood also. Don't really want to make it all steel - mix it up a bit.



The enclosure will be too big so enough spare volume for bracing the back of the mid bass to the lower hemisphere and some holy panels to try help break up the internal dimension. There will still be room for the bottom of the sphere to have kiln dried sand poured in. The sphere won't really be a sphere inside. Some fibreglass will go in too.

I plan to use damped threaded rods and nuts to secure the driver and to clamp the top hemisphere to the birch ply dividier and the ply driver mounting ring to the steel hemisphere. IThat stops it ringing. The lower hemisphere will be clamped to the ply divider by a single bolt to stop it ringing but the sand in the bottom will also help with that. The hemispheres already have an 8mm hole exactly at the pole of the hemisphere. The whole thing should be able to be dismantled if needed.

Not decided on finish yet but the steel will be painted. I plan to leave the ply parts and wooden legs uncovered. The tweeter sphere will probably be plastic and painted the same colour as the steel. I will have to flame treat it first to get good adhesion.

It's finally going ahead but first I need to slice the top off two of the hemispheres with my 300mm chop saw which I am going to modify for this job. It will go back to normal afterwards.

[r3xj0hn570n]

It's finally going ahead but first I need to slice the top off two of the hemispheres with my 300mm chop saw which I am going to modify for this job. It will go back to normal afterwards.

If was asked to cut a precise hole in a steel sphere i'd use a steel sheet nibbler attached to a circle jig.

[r3xj0hn570n]

So, working out which position is best will have to be done by ear possibly with the help of a bit of science if I can't detect any difference, in which case I want to try and work out what is likely to be the best measured solution without measuring

You are allowed to measure, mics are cheap and REW is free. That said, if you can't hear the difference (and it won't blow up), it probably doesn't matter that much.

I had thought that lining up the approximate position of the two voice coils would be good enough but I don't think it is. Most stuff written about time aligning two drivers is concerned with point and squirt not when you have the two drivers at 90 degrees to each other. When they talk about time aligning two drivers with a first order then there is a 45 deg phase shift with both low and high pass. It is my understanding that the low pass delays the sound whilst the high pass works a miracle and actually brings the sound forward in time before it has even happened. If that is the case then I require my 2" tweeter to be behind the centre of the voice coil of the mid bass because the mid bass has no low pass filter - it is running fully open. My high pass is a series line level polystyrene capacitor at the input of the power amp board.

Anyone make sense of that :-?

Don't sweat trying to phase align the 2 drivers too much. Capacitance, by definition, is a derivative measure, meaning phase lag (current, which drives the coil lags voltage applied by the amp), so having the tweeter forward more is a perfectly valid thing to do. To measure it, you'd need a square wave source, quite a bit of patience and imagination and as soon as you changed measurement position, it'd change completely, especially with an omni as it's interacting with the room, a lot. There was a great article written by (i think), the founder of Quad, in HFNRR about room acoustics. I was unable to find it online. Pity.

[karatestu]

So, working out which position is best will have to be done by ear possibly with the help of a bit of science if I can't detect any difference, in which case I want to try and work out what is likely to be the best measured solution without measuring

You are allowed to measure, mics are cheap and REW is free. That said, if you can't hear the difference (and it won't blow up), it probably doesn't matter that much.

I had thought that lining up the approximate position of the two voice coils would be good enough but I don't think it is. Most stuff written about time aligning two drivers is concerned with point and squirt not when you have the two drivers at 90 degrees to each other. When they talk about time aligning two drivers with a first order then there is a 45 deg phase shift with both low and high pass. It is my understanding that the low pass delays the sound whilst the high pass works a miracle and actually brings the sound forward in time before it has even happened. If that is the case then I require my 2" tweeter to be behind the centre of the voice coil of the mid bass because the mid bass has no low pass filter - it is running fully open. My high pass is a series line level polystyrene capacitor at the input of the power amp board.

Anyone make sense of that

Don't sweat trying to phase align the 2 drivers too much. Capacitance, by definition, is a derivative measure, meaning phase lag (current, which drives the coil lags voltage applied by the amp), so having the tweeter forward more is a perfectly valid thing to do. To measure it, you'd need a square wave source, quite a bit of patience and imagination and as soon as you changed measurement position, it'd change completely, especially with an omni as it's interacting with the room, a lot. There was a great article written by (i think), the founder of Quad, in HFNRR about room acoustics. I was unable to find it online. Pity.

Right you are Rex. Good point about the omni.

So I will just choose the best position judged by my ears. If I get into measurements now then this speaker will never get built. In fact no speakers will ever get built I will finish this one then consider learning how to measure cos there is no point in doing it badly.

This semi omni thing complicates the application of theory. I try to get the two drivers within a quarter wavelength at xover but it is impossible. At 20 deg C 7kHz has wavelength of 5cm. Quarter of that is 1.25cm. Nobody crosses over this high for a good reason - you can't get the centre of the two drivers close enough.

[karatestu]

If was asked to cut a precise hole in a steel sphere i'd use a steel sheet nibbler attached to a circle jig.

Nibblers won't cut through 3mm steel unfortunately. Plus I want the cut edge to be a flange for the rounded over ring of plywood that the driver will be rebated into , Hope that makes sense. This flange needs to be exactly parallel with the bottom flat edge of the hemisphere (the equator line of the sphere when the two hemispheres are put together).

That is why i was going to modify my 300mm chopsaw. I need to construct a perfect right angle with the saw mounted on the vertical surface and the flat of the hemisphere will sit on the horizontal flat surface. I can then spin the hemisphere and cut all the way around and have the flat flange as I would like.

[karatestu]

Playing about with the look of spheres on top of each other.

How about this one



If two spheres were joined up with15cm diameter connecting hole that would be about 26 Litres. That would be enough for some more bass. Getting into 8" driver enclosure size with that.

You may notice I have tidied up my wiring Now using some 1 metre lengths for the left speaker and 1.5 metre for the right. Made out of 2mm solid core mains wire and LS2. Does the job. Trying to evaluate the difference from the sprawling mess of vastly over length LS6 I had on before.

Also gone back to the sphere directly over the 6.5" mid bass. I am all about symmetry (OCD). I did comment that this way chokes the mid bass a bit but I think I had it too close before. The pole of the sphere was lower than the surround before but I have jacked them both up a bit and that seems better.

With the 2" cone behind the mid bass I couldn't live with the thought of it's output getting reflected out of the deep mid bass cone. Although I couldn't hear the effect once the speaker was raised up again on legs, I couldn't live with the thought of those reflections messing up the off axis response.

Another reason for it being centrally over the top of the mid bass is I think that will redirect the waves equally all around which in my mind should be better.

Right that's that decided then.