Tannoy had a commercial model using the principles we are discussing here, the ALF625. It was while reading a review of this in Hi - Fi World sometime in 1994 that I was inspired to try doing something similar. Later that year, the ALFie 625 was reviewed in What Hi - Fi? who described it as "slow"
They also ridiculed the acronym, "Advanced Low Frequency improved electronics", saying; "More Arnie and less Alfie would be better". Maybe such a review in the popular hi - fi press can stifle a product in the marketplace, I think it ceased production in 1996.
[ ... ]
I have here, opened front & back in front of me, a Tannoy 625ALFie (as it's labelled, and how I found this old thread on a search for it - little/no information other than your mention - I have yet to search out the Hi-Fi World review) - and I can state that it uses no MFB - there is no connection to the woofer other than a rather thin pair of wires to a standard voice coil in a rather cheap looking stamped metal basket woofer (although it does have a small vent in the pole piece), and the "improved electronics" uses a TDA7294 monolithic chip mounted to the back panel plate which acts as heat-sink.
TDA7294 Audio Amplifier Circuits
The amplifier plate is recessed onto a second 18mm layer behind the main back, and under that is the actual back of the woofer cabinet, unbraced, and very ringy... I was impelled to take it apart, having recently acquired it from a friend whom I'd been searching for a sub for, for his large caravan, found at a bargain price, and advised him on the basis of the original 625ALF brochure which does mention MFB and a second sensor coil. but he decided it was too large, untried, so passed it on to me... and driving it alone, & with various set-ups over two amplifiers and large and miniature speakers, I found a rattling, worrying enough in itself, and a sort of subdued thumping low bass...
The rattling may have been the wires in the front part which are loose and long (there's a 3-way twisted pair to the R-G LED indicator as well) but also perhaps from the undamped unbraced inner back panel, even wires on the other side rattling against it, so I'm going to tape the wires, have just cut an oak brace for inside the amplifier compartment inner actual sub back, and found a couple of stainless steel plates 275x100x1mm that I will glue with contact cement, hopefully making a CLD constrained layer damping effect for the other side, from the front, for the two halves between the oak brace at back... which I will have to position carefully, to fit just between the PCB and torroidal transformer, and allow wires to pass over.
As for the damping in the front part of the cabinet, which actually is only 19cm deep behind the woofer, (32.5 overall deep external!), it's just 3 identical portions of approx 40mm white wadding, I don't know what you call it, one on the back, two cover the sides all around, very efficient production.
I have some green, loose fiber "Sound-Stop" I think it was, fine fiberglass I think acoustic wadding designed for interior walls, leftovers saved from a professional installation, that I have thought I might try instead, big chunks.
I remember other speakers that were almost completely filled with foam 'rubber', whatever you call that... even as passive radiators.
At least this a sealed cabinet. So perhaps better stuffing will lower F3, etc.
But in summary: the 625ALFie does not have MFB.
Maybe the 'improved electronics' from marketing was from 'make it cheaper'.
I'm disappointed, naturally;- I had hoped to find a MFB system within.
But no. Although it does say "CURRENT DRIVEN ACTIVE SUBWOOFER" on the back panel, and I don't know how they've configured the amplifier chip, except to note that there are a couple of low ohm parallel high watt resistors in series with the output, but I think the least I can do is improve the cabinet damping.
Photos later, I just wanted to take notes while I have it apart in the lounge!
Back from the dead, both this thread and that Tannoy speaker.
Odd the cabinet is so junky but then manufacturers have always been applying MF to make poor speakers sound like great speakers. (Around here, we want MF to make SOTA speakers greater.) I'd first listen to see if the electronics have mechanical hum before going too far remediating the cabinet.
From your description, I am thinking the speaker has a series (current) MF circuit. All it takes is to have a low resistance resistor in the driver path and that may not be discernible to the naked eye. A bridge (which is not much different electrically) could also be as easily hidden. You don't need an external sensor to have wonderful MF.
To test, you simply see if the driver is getting a constant voltage across the freq compass or if the voltage changes in expected MF ways. Or simply put in one note (say 80 Hz or anywhere low) and see if the voltage to the driver changes when you put a finger on the cone. Or push on the cone and see if the amp talks back.
And then you might want to think about a better enclosure instead of remediating the old one.
Ben
Last edited:
RdM; The woofer in the original Tannoy design used two coils. The main one for driving, and a longer thinner one for the feedback. I don't know how it was executed, but that was the idea. The designers (Mills and Hawksworth?) used the feedback to aim for a specified output impedance that would give the desired response, and that is the reason for their "current driven" claim on the box. Has your sample been modified previously? It is over 20 years ago now.
Reading the posted wisdom of bolserst, I think the second-custom-coil might be a very smart approach (although not feasible for DIYers who can't add a coil to the VC). Sony used it too. With a custom coil, you can avoid the mutual inductance problem (although not a big problem for subs), loss of power drive by grabbing the second winding, and get a suitably big signal to work with.
I think the endless squabbling over positive, negative, current, velocity, and other MF dogmas is largely irrelevant. I suspect all MF systems operate as negative impedance amps whatever the nature of the error signal.
Moreover, it is silly today for members of this forum to fuss over hitting the precise low freq speaker freq response by fussy modelling or worry about velocity versus acceleration correction if they want to explore MF. The FR gets settled only after the speaker is in place in your room and you run REW.
Some of the challenges of implementing MF arise from trying to make it work above, say, 200 Hz. Not easy to work with phase and not easy to make the whole cone (and surround) move as one piston. Now-a-days, clear you need to break-off the low frequencies no higher than maybe 150 Hz and so these MF challenges aren't relevant.
I read through this ancient thread and want to add that flat FR is not an important goal (as noted above) nor is distortion reduction beyond the horrible mess at lowest frequencies. It is the gross cone mis-motion that MF addresses well because it is so gross... the woofer cone is slow to get going and slow to stop and the speaker rumbles on.
Ben
Last edited:
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.