Doc modding Marantz imperial 7

A place for DIY project discussions.
User avatar
karatestu
Posts: 5945
Joined: Sun Jan 08, 2017 4:40 pm
Location: North Yorkshire
Has thanked: 1865 times
Been thanked: 1378 times
Great Britain

Re: Doc modding Marantz imperial 7

Unread post by karatestu »

Put the tweeter pod back to normal for now. Will investigate the Shahinian style tweeter firing at a later date.

I forgot I had the tweeter capacitor to play with as well. I had settled on 1uf and 4.7uf in series to make 0.82uf. That was ok with the 5" bipolar. I tried two 1uf in series to make 0.5uf. That was obviously too low (too high corner frequency) as it sounded like a hole in the frequency response.

I found the 3.3uf Russian paper in oils and put them in series with the 1uf PIO. That gives 0.76uf. That is much better than either of the others. It has calmed down cymbals and brass instruments quite a bit. I wish I had a 2.2uf to try in series with the 1uf as that would give 0.68uf which I would like to try before settling on anything.

Sounds crazy but my crossover (if we take the natural roll off of the mid bass to be the low pass) with 22 ohms and 0.76uf is 9.5 Khz. With 0.6b uf it will be 10.6 Khz :shock: But then a tweeter rolled off at 6dB per octave is going to be producing significant output a long way below 9.5 kHz. An octave is a broad frequency range up at this frequency level. Not sure what rate the mid bass rolls off as there are no graphs I can find for it and it is doped now so wouldn't be correct anyway. Can only do it be ear really.
DIY FREE ZONE

antonio66
Posts: 1006
Joined: Mon Jul 29, 2019 8:16 am
Location: Grimsby/Phnom Penh
Has thanked: 1449 times
Been thanked: 408 times
Great Britain

Re: Doc modding Marantz imperial 7

Unread post by antonio66 »

Love that little tweeter pod you've got there Stu, and I like the idea of being able to adjust, so none of them are firing at the listener directly. Unsure about tilting though, but trust 'your' ears, and don't forget Shahinians sit low/lower to the ground than most speakers.

User avatar
karatestu
Posts: 5945
Joined: Sun Jan 08, 2017 4:40 pm
Location: North Yorkshire
Has thanked: 1865 times
Been thanked: 1378 times
Great Britain

Re: Doc modding Marantz imperial 7

Unread post by karatestu »

Good point about the height of Shahinians Dave, thanks. You should know having owned some :dance: Was it Cubes that replaced them ?

I keep bangin' on about possibly having a hearing problem (sensitivity to certain frequencies) so I suppose I ought to get a test once covid has calmed down a bit. I have mild tinnitus probably from years of noisy drumming and machinery at work as well as wax bunged up ears. I remember hammering some metal without ear defenders years ago and my ears were whistling for a hour, very unpleasant. Maybe I should get them looked at before finally signing these speakers off.

Since playing with the tweeter capacitor I don't feel the need to go to Shahinian style tweeter arrangement although I will at some point (maybe not with these speakers).

Something I have seen in my own diy speaker behaviour- try something new, get very excited, listen to it for a while, start to dislike certain things about it, dismiss it as flawed, fall back to the known good. It would seem the known good for me is a bipole with up & down firing mid bass, tweeter(s) behind the up firing mid bass.

I need to try one forward firing tweeter again but behind the mid bass, just to check that the four tweeter arrangement is really better. I have my original visaton g25ffl sitting idle that I can quickly swap in. Really hope I haven't been kidding myself all this time :pray: - it's good to have a reality check now and again.
DIY FREE ZONE

User avatar
karatestu
Posts: 5945
Joined: Sun Jan 08, 2017 4:40 pm
Location: North Yorkshire
Has thanked: 1865 times
Been thanked: 1378 times
Great Britain

Re: Doc modding Marantz imperial 7

Unread post by karatestu »

Which is best ? A wide or narrow baffle ? This graph seems to advocate for minimum baffle

Image

But it is only true on axis, which makes me think that the worst baffle is a baffle oriented towards the listening position. The more off axis the best, width and driver positioning, edges rounding, etc... being secondary.

Image

So Cube users (and me :dance: ) who are always completely off axis (90 deg when sat down) to the up firing (and down firing) mid bass enjoy very low diffraction (approx zero) from said driver. Better than a sphere on axis in fact.

But diffraction mainly affects tweeters in conventional speakers. But Cubes are not conventional speakers in that the mid bass is not rolling off until much higher frequencies than conventional speakers. This may make diffraction a more important consideration but the large mid bass drivers (especially the 8") are starting to beam the higher frequencies long before we get to 6,7 or 8 Khz. So who knows.

With the tweeter being front firing and the listener usually on axis then Cubes are completely at the mercy of diffraction from the tweeter baffle. Round overs would probably help. It would make sense to put the tweeter in a small sphere or as close as you can get. For me that is a rounded over dice shape as I have four of the blighters. I don't need the flat faces on the back or the bottom of the dice though so maybe I could improve it by sticking on dome shapes that will complete the circle shape on those two faces.
DIY FREE ZONE

User avatar
karatestu
Posts: 5945
Joined: Sun Jan 08, 2017 4:40 pm
Location: North Yorkshire
Has thanked: 1865 times
Been thanked: 1378 times
Great Britain

Re: Doc modding Marantz imperial 7

Unread post by karatestu »

There will be diffraction ripples in every tweeter's response. But the (in)sensitivity of our hearing masks them quite easily above 6kHz apparently. Humans are most sensitive to sound deviations in the range between 2-5kHz, which is exactly where most two way speakers crossover are and also where the worst ( first order) diffractions of conventional hifi-loudspeakers are.

Can diffraction be measured by the human ear :think: :think: :think: From the graphs above you can see the magnitude of the ripple starting at low frequency and continuing up as a function of baffle dimension and shape. In some cases it's more than +/-3db and you can hear that apparently.
DIY FREE ZONE

Ordinaryman
Posts: 475
Joined: Tue Apr 16, 2019 4:50 pm
Has thanked: 100 times
Been thanked: 77 times
Great Britain

Re: Doc modding Marantz imperial 7

Unread post by Ordinaryman »

Wow graphs and measurements. In the end I'm sure you will let your ears decide, Don't know for sure but I think its what "The Doc" did. You will know so much better than I. Fascinating.
These users thanked the author Ordinaryman for the post:
karatestu (Thu Feb 11, 2021 3:16 pm)

User avatar
karatestu
Posts: 5945
Joined: Sun Jan 08, 2017 4:40 pm
Location: North Yorkshire
Has thanked: 1865 times
Been thanked: 1378 times
Great Britain

Re: Doc modding Marantz imperial 7

Unread post by karatestu »

No measurements here (those graphs aren't mine) but certain things can be used as a guide and hopefully bring better music without selling out the core principles. If Doc was here he probably would have deleted the post or banned me :lol: More likely being I wouldn't have had the guts to post it in the first place. :grin: :whistle:

Doc always said just use your ears which is what I have done so far and will continue to do so. Doc said no stuffing. I prefer just a little bit with my bipole. Doc said sealed box was best. I think I prefer aperiodic to air tight in my situation but that might change next week. :roll: Doc didn't give any thought to diffraction (I don't think). The Cubes had no rebates for drivers, no round overs and even raised sides at one point. Difficulty / expense of machining acrylic was probably the reason for that.

When I first moved my four tweeters out of the 12cm square wooden cube into the foam dice with lovely rounded edges the sound was transformed. Apart from a bit lower vibration from the mid bass (foam) I can't think of any reason for the dramatic improvement other than the rounded edges helped with diffraction (which apparently spoils FR and imaging).
DIY FREE ZONE

Ordinaryman
Posts: 475
Joined: Tue Apr 16, 2019 4:50 pm
Has thanked: 100 times
Been thanked: 77 times
Great Britain

Re: Doc modding Marantz imperial 7

Unread post by Ordinaryman »

I'm sure you would never sell out the core principles. Whilst I understand little of the technical terms, I cant wait till you come to a conclusion, build them and DOMINATE the speaker world :epopc: :clap:
These users thanked the author Ordinaryman for the post (total 2):
karatestu (Thu Feb 11, 2021 4:37 pm) • antonio66 (Thu Feb 11, 2021 4:54 pm)

User avatar
karatestu
Posts: 5945
Joined: Sun Jan 08, 2017 4:40 pm
Location: North Yorkshire
Has thanked: 1865 times
Been thanked: 1378 times
Great Britain

Re: Doc modding Marantz imperial 7

Unread post by karatestu »

"DOMINATE the speaker world" :lol:

I doubt it. Too many people turn their noses up at speakers that aren't point & squirt rectangular boxes with table top frequency response graph, Linkwitz Riley 4th order hifiness, point source, imaging to die for and lovely shiny badges that align with their fanboyism (Tannoy, Yamaha, Harbeth, Spendor ). :grin: Oh and only the most expensive drivers get you taken seriously.

They will do for me and that's about it I should think. My speaker design / building abilities are very much amateur and probably seen in the wider audio community as such along with hilarious, Heath Robinson and generally a magnet for ridicule. That's why I only post here, nobody understands what I am on about or doesn't give a feck.

These will hopefully be my forever speakers - that is the aim. Can't be doing with box swapping. Although I have done loads of experiments which might be seen by some as even worse than box swapping :lol: Saying that I will need some speakers at work which might end up being open baffle jobs on 8ft x 4ft sheets of plywood :lol: :lol: :lol: :lol:

Speakers that play music in a dynamic and coherent way (sounds like real musicians) has always been the goal. That's why everything clicked here when I first joined - The Doc and most other members have the same objectives. I don't want something that measures perfectly and throws the baby out with the bath water.

The main core principals being
1) Semi omnidirectional - non of that point and squirt nonsense :naughty:
2) No filters or as few as possible. We need one on the tweeters but I might have a cunning way of avoiding that and doing it at line level by removing the ones on the output of the source components and putting one of them in the input of the power amplifiers. The resistance to ground is so high there that I can use some small value polystyrene caps (yum) to get the correct high pass filter. Usually the caps on the output of source components are quite large in order to not roll off the bass but still block DC. Same amount of filters at line level as before but one less between the power amps and the speakers. There is no switch on thump with NVA amp boards that I can hear so should be safe in that respect. But if the amp goes DC for any reason then it's bye bye tweeters.
3)Bass energy in phase with the ground (up firing mid bass). Speaker doesn't then rock about as it would with forward firing. I have gone one further and added an identical down firing mid bass which not only reduces cabinet vibration by opposing force but compensates for baffle step loss for free :dance:
4)As few components in the signal path as possible. Yes that old chestnut. I managed to do away with the padding resistor before the single tweeter by adding three more tweeters and wiring them in series. Some could say I have actually added more components rather than take away (one resistor removed but three coils added :grin: ) There is some gain to be had from these added tweeters though rather than just drop voltage and produce heat.
5)No damping or stuffing. It was going great until recently when I added an internal box divider made of 1" open cell foam and liked it :whistle: Lost browny points on that one :shhh:
6)No evil electrical BSC circuitry, notch filters, shelf filters or impedance flattening bollox. Tick.
7)Stiff, good quality cabinet materials. So far I am a big fail on this apart from the steel lining I have done. Stiff and heavy might not be the best way to go. Stiff for bass yes but heavy usually raises the frequency of resonance and stores energy. I am starting to think that building out of something light and non resonant may be the way forward. Something like EVA foam with some 3 mm steel to make it rigid although 20 - 30 mm EVA foam is said to be quite rigid already. Also it can be sanded and bent into nice shapes.
8)Bored yet ?
9)As low distortion as possible. Well, distortion can take many forms. The main one for me has been adding more drivers and sharing the work out. It was quite remarkable the difference that made with both tweeters and mid bass. With the tweeters it was important because they only have a 1st order filter and so only roll off at 6 dB per octave. The mid bass are cheap and probably anything which lowers excursion for them will be rewarded with lower inter modulation distortion.
10)Doping - I am pretty sure that this improves these mid bass no end. Even with a first order low pass on a mid bass you are going to get some of the cone break up coming through unless you crossover low. Good luck finding a tweeter to match up - hello three way. Yes sensitivity is reduced a bit but if you start off with a relatively efficient driver and add more of them (in parallel) you can easily increase spl to that of the tweeter. Happy days.
11)The little housekeeping things like rounding over edges, driver rebates generally reducing diffraction where possible without compromising any of the core principles
12)Closed box for tight bass - I may have strayed from the path a little but at no point have I considered bass reflex, transmission line or other group delay, fart inducing stunt bass. Timing is everything - the Doc's principles all point to that.
13)Something nice to look at. Totally failed there so far :hand: Although the final build in those oak tables will put that right :guiness;
14)Have fun :handgestures-thumbupright: :handgestures-thumbupright: :handgestures-thumbupright: :handgestures-thumbupright: :handgestures-thumbupright: :handgestures-thumbupright: :handgestures-thumbupright: :handgestures-thumbupright:
15)A good musical experience wherever you are in the room. I don't want to have to sit in one place with my head in a vice whilst the rest of my family get an inferior experience. Fighting over the sweet spot :lol: The only other way to do it would be to do my listening on my own in a dedicated music room. I don't want to be on my own, music appreciation is a social thing for me. I don't have a room spare anyway. This has been the source of much head scratching. Semi omni with one tweeter does this better than most. Fully omni takes it further but is hard to do (diffusers etc). The best I could do was to have several tweeters pointing in different directions but firing no less than 90 degrees in regards to another. Seems to work and no hideous comb filtering that I can detect. Pointing tweeters 45 degrees at the ceiling like Shahinian does interest me but may be a step too far.
16)Anything else I have forgotten. Oh, dynamic and musical at low volumes which is all I ever listen at. It seems I can defeat the Fletcher Munson curve by making my speakers a bit on the bassy side of neutral. Adjustable aperiodic vent can help with that :grin: The other thing that occurred to me the other day was that with the oak table legs splaying out by 5 degrees and if I tilt the bottom baffle lower towards the back of the speaker, I will have no internal enclosure walls parallel. Whether that will make any difference I don't know.
DIY FREE ZONE

User avatar
karatestu
Posts: 5945
Joined: Sun Jan 08, 2017 4:40 pm
Location: North Yorkshire
Has thanked: 1865 times
Been thanked: 1378 times
Great Britain

Re: Doc modding Marantz imperial 7

Unread post by karatestu »

Didn't know whether to put this here or in the amplifier thread. I have talked about removing the capacitors in the speaker level high pass filters for a while now. Well now I am back to biamping I am able to do it. And I have :dance:

What this requires is the addition of a capacitor in series with the signal at the input of the power amp board. Combined with the resistor to ground at the input of the amp board it creates a high pass filter. At speaker level the capacitor combined with the drive units impedance does the same thing.

Image

So now I have no components between any of the four amps and the drivers they are connected to :guiness; The capacitors I have used at the input of the amps are polystyrene which are arguably / probably the best caps you can use.

I tested this using some old tweeters that I don't care about. Stuck them on top of the mid bass cabs. Played just the tweeters at first to check everything was ok. It sounded good even through the crap metal domes.

Image

Turned everything on and off several times to check no surges of DC was going to blow anything and then swapped my regular four visatons back in. There was a moment when I thought I had fecked it up as some crackling was coming from the tweeters on the left hand side only. Turned out to be a speaker wire I had not clamped up tightly enough in the binding post :roll:

I am listening to it now. I don't have the cap value quite right as there is too much of a dip in the FR. A larger polystyrene cap should sort that out. There is an increased clarity to the music but the cap being too small might have something to do with that. Very very promising though :epopc:
DIY FREE ZONE

Post Reply