AH, gotcha.
The passive line level transformer matrix doesn't have any adjustments in itself. It's just purely converting L/T to Sum/Difference.
So that means that the side speakers in your scenario would be fully R-L, and (same signal inverted) L-R, probably not what you want. Presumably you could use a suitable resistor to bleed some signal from the Sum signal to the Difference signal, but then you'd also end up bleeding the same amount of signal the other way.
You DO get to control the front vs side balance (from mono to crazy stereo) however, just using your amp's balance control.
I don't want to make it sound like I built some amazing matrix encoder box. I just did a quick proof-of-concept using four 1:1 transformers and a bird's nest of alligator clips. I was amazed that it worked!
The passive line level transformer matrix doesn't have any adjustments in itself. It's just purely converting L/T to Sum/Difference.
So that means that the side speakers in your scenario would be fully R-L, and (same signal inverted) L-R, probably not what you want. Presumably you could use a suitable resistor to bleed some signal from the Sum signal to the Difference signal, but then you'd also end up bleeding the same amount of signal the other way.
You DO get to control the front vs side balance (from mono to crazy stereo) however, just using your amp's balance control.
I don't want to make it sound like I built some amazing matrix encoder box. I just did a quick proof-of-concept using four 1:1 transformers and a bird's nest of alligator clips. I was amazed that it worked!
I should actually try the SSS with the full M/S, there should not be that much difference - and as you say, changing the M/S ratio will give me some chance for tuning if necessary. Do you have any recommendation for the transformers? I was looking at the Neutrik 10k:10k line transformers some time ago, it seems they could work fine.
With a Sum/Difference matrix, you don't technically need two side speakers.
You can also run a single speaker sideways in an open baffle, with pretty much the same result.
I have a Fender Acoustasonic SFX acoustic guitar amp that does this, and it's quite effective. Here it is with the speaker cloth off:
Around the same time, Groove Tubes made a smaller amp for keyboards using the same technology.
You can also run a single speaker sideways in an open baffle, with pretty much the same result.
I have a Fender Acoustasonic SFX acoustic guitar amp that does this, and it's quite effective. Here it is with the speaker cloth off:
Around the same time, Groove Tubes made a smaller amp for keyboards using the same technology.
I don't know very much about transformers, unfortunately. I bought the cheapest ones on Amazon, Bojack El-14, 600 ohm, 10-pack for $11.
Again I was kind of surprised it worked! I haven't done any critical testing to see how good they sound, but at $1 each I don't expect perfection.
Ok. Yes, I do not need two side speakers, but the box already has them anyway. But unfortunately, you have just inspired me to build a new one with just two speakers🙂 The prototype will be 3D printed and in two parts to find the best position for the open baffle driver - top, bottom or behind the mid one.
Awesome!
Keep in mind that if you’re using the same type of driver for both positions they’ll respond quite differently, since one of them is open baffle. The open baffle one won’t get all that much bass, but you wouldn’t get that on the side (Difference) channel anyway, since bass is rarely panned to one side, or out of phase left-to-right.
I suppose one COULD make both the sum AND difference channels open baffle. Hmmm…
I am more afraid of the front to rear difference in FR, the LF rolloff should not matter that much - it is not so much different from the very small closed box. That is why the 3 speakers are a good solution.
This is the circuit that I've been working with. Varying the shunt cap and resistor to attenuate the balance of FR (slight LP to the front, slight HP to the sides).
This was my prototype with a bass unit al-in-one. It worked really well dependant on the room placement and fiddly wiring. I found it impossible to create a passive crossover to the bass unit without smashing the impedance and creating a really complex (for me) crossover. This unit used a crappy 2.1 amp. I'm not very electronically literate. However, I may revisit if I see other people having a go. Doesn't solve the awkward room issue. Just posting for info.
This was my prototype with a bass unit al-in-one. It worked really well dependant on the room placement and fiddly wiring. I found it impossible to create a passive crossover to the bass unit without smashing the impedance and creating a really complex (for me) crossover. This unit used a crappy 2.1 amp. I'm not very electronically literate. However, I may revisit if I see other people having a go. Doesn't solve the awkward room issue. Just posting for info.
Attachments
Last edited by a moderator:
Yeah, that's the circuit I've experimented the most with as well.
But I never got committed enough to doing it properly to find an amp that could do that bizarre configuration without complaint.
I think the biggest downside to this method is that all three drivers are too interdependent. You can't to 2 or 3-ways, you can't EQ front and sides separately, and you can't really mix and match components, without it becoming a big balancing act.
But I never got committed enough to doing it properly to find an amp that could do that bizarre configuration without complaint.
I think the biggest downside to this method is that all three drivers are too interdependent. You can't to 2 or 3-ways, you can't EQ front and sides separately, and you can't really mix and match components, without it becoming a big balancing act.
My experience too.
RE the amp. If you haven't already, might be worth trying a cheap class a/b chip amp from ebay as a test to see if that improves things. I has some limited success with this, but couldn't find an inexpensive one that was also 2.1. Same amp problem with full M/S S/D I guess?
I'm imagining the open baffle "side" driver mounted on a circular baffle within a 180degree sphere (Ikea bowl). If the flat baffle were fully circular and mounted at 90degrees inside the bowl so there was half the baffle outside the bowl, it could help the separation as well as directing the side reflections away from walls. Not sure where the "mid" speaker would go however.
The only other method that I can imagine would involve some esoteric time-delay or phase DSP, which is way out of my understanding. Although I would be very up for copying!
RE the amp. If you haven't already, might be worth trying a cheap class a/b chip amp from ebay as a test to see if that improves things. I has some limited success with this, but couldn't find an inexpensive one that was also 2.1. Same amp problem with full M/S S/D I guess?
I'm imagining the open baffle "side" driver mounted on a circular baffle within a 180degree sphere (Ikea bowl). If the flat baffle were fully circular and mounted at 90degrees inside the bowl so there was half the baffle outside the bowl, it could help the separation as well as directing the side reflections away from walls. Not sure where the "mid" speaker would go however.
The only other method that I can imagine would involve some esoteric time-delay or phase DSP, which is way out of my understanding. Although I would be very up for copying!
Wired up 3 identical fullrange cabinets in the 0.5 matrix under the tv. It's interesting, made the mixed down 2 channel going into it bigger, as if it were coming from the screen, but I lost a decent chunk of vocal intelligibility. All drivers were just facing forward, I wanted to see what it did and if it worked with my amp correctly. What is the impedance of three 4 ohm drivers, it looks like 4 + (4/2) = 6 ohms?
Going to do some more experimenting, this is fun.
Going to do some more experimenting, this is fun.
If you use a line level matrix solution (passive transformers, maybe some active option, or even digital - though Mini DSP doesn't seem to be able to pull it off), then you can use any amp you want, no issues there. And you can eq and crossover the line level signal, either pre matrix (processing normal L/R signal) or post matrix (processing the Sum/Difference signals separately).
I tried my best to visualize the IKEA bowl open baffle, but I have to admit I'm a bit puzzled about how it would look (or sound).
Interesting! I can't wait until I have a little time to mock this up in foam core.
Reduced intelligibility might be a real problem - blending a signal with it's inverse is, quite literally, causing all kinds of phase issues.
Yeah, I'm going to point the side ones off to the sides when I get back to it. I don't really have any side walls so I guess I'll see what happens. The four speaker matrix might be better as the center is two drivers rather than summed into one - I think.. I do like the way it sounds more involving with movies but I really don't want to lose the intelligibility. 
Strangers in the night
Drinking martinis
Having sex at night
Sinatra playing on the TV
Tell me you're not playing that LP of Sinatra on that record destroying rotating device.. .aagh...
Of course I am!
It's not a record destroying device, please keep your thoughts before typing "assumptions".
With a highly tweaked/modified/fully serviced Garrard model 3000 - Synchonous motor -3.5 gram Pickering elliptical magnetic cartridge - adjustable antiskating as a custom add-on to the changer - and a heavy cast aluminum platter from a Garrard SL-65.
Dead-on speed accuracy, inaudible rumble/noise - the changer approaches fine Dual turntable performance.
Any more questions?
Nice, clean deck! 😎
I read that thousands of Garrard 3000s were sold with Pickering P-15 cartridges, which performed very well in the low mass arm.
I read also that Garrard would custom build the motors for companies - 2 pole, 4 pole and the Deluxe 3000 could have a Synchro-Lab motor.
Indeed, they can be capable of very nice performance, particularly the "deluxe" versions like mine is.
It was specifically made for the KLH music systems of the mid-1960's, and yes, with the synchonous motor.
The Pickering cartridge is a special version similar to the X15/PATE but with a modified mounting and fitted for a "plug in" receptacle, resulting the very slim headshell.
The Sylvania Exponent 440 "upscale" portable stereo phonos also used a similar version of the 3000 but with a 4 pole non-syncho motor.
I serviced one in the shop for a customer, and for a portable with swing out wing speakers, it rocked!
Whenever I get snarky comments related to vintage record changers, I immediately attest them to people who have their head in the audiophile "clouds" and just shake my head.
Yes, perhaps, a "high end expensive single play" machine would give satisfactory performance to that crowd, but without the actual listening test, to put down a specific changer model is just babble.