advertisement


PFM Special - a simpler and more affordable DIY loudspeaker design

We all know that spikes are de rigeur in loudspeakers, unless you have Shahinians. Somehow, the Ergo E-IIIs and E-IVs aren't particularly fussed about them, and manage to sound coherent and musical without them. However, they do perform much better, without a shadow of a doubt, firmly spiked onto the floor. I thought the PFM-Special would have a similar degree of ambivalence to spikes - that of benefitting from having them fitted, but not critically dependent on them - but how wrong was I, as I found out today.

As I listened more and more to the PFM-Specials with the tweeter pegged back by 4.0 ohms of series of resistance and yielding a near flat treble response, the more I realised something wasn't quite right. The rhythmic and snappy nature of the design with the 1.0 ohm series resistance seemed to have taken a back seat and it was a little bit harder to follow the music. Tonally, it wasn't a problem, but the fun factor seemed diluted significantly. I had put it down to the ups and downs of running in a design, and was patient enough to wait for it to come right again.

It had always been my intention to fit spikes to the lower bass cabinets, and since Steve was planning on coming around with a pair of paving slabs, I thought it would be a good idea to install the spikes before tomorrow. It didn't take much effort to install, and I wasn't expecting much but how wrong I was.

With nothing else changed (identical positioning, toe, XO configuration etc) other than being firmly spiked onto the floor boards now, the difference in intelligibility was staggering. Previously, with the PFM-Specials sitting only on their ample 330 x 400mm footprints on carpet, the top of the upper baffles could move up to +/- 5mm with a gentle prod, though in actual service even with loud bass transients, no movement was perceptible to my eyes. Now firmly spiked, there is no movement when prodded, except for the natural flexure of the timber board, and only with significantly greater force. Now, when music is played, the following attributes of their character come shining through.

Dynamic Headroom - the ability to startle is most obvious in these loudspeakers. More so that the E-IIIs. They go from ppp to fff instanteously and without strain or effort. I honestly thought I'd clip the NAP135s when playing Pink Floyd's live version of 'The Wall' with the 52 volume dial pegged at 11:30 and the room registering 108dB peaks, but strangely it didn't sound uncomfortably loud or strained. On our moderately warm day today, I had expected the 135 fans to be whirring frantically, but they were merely idling. These babies go seriously loud and clean.

Timing - which had gone south with the change in tweeter levels, is now even better than ever. Top and bottom transients are seamless and snapping together as one, and the temporal shifts for musical interplay and tension are remarkably obvious. As a consequence, the rhythmic intent of any well recorded piece is easy to hear and thus understand.

Ambience - though not critically important to flat earthers, is still pretty much intact in this design. The upper most treble regions are still a bit muted and therefore they are not as airy as the roundest earth designs, but there is enough energy - in concert with the dipole mid - to create a very solid sense of space. The PFM-Special was never designed to be an imaging freak, but I cannot help but feel it portrays voices and instruments in the right sizes and places them palpably in the space a metre beyond the side of each loudspeaker and several metres back. They don't project much in front of themselves, though ...

Gravitas - these do bass and how! Tight, deep and fast. It does sound a bit plummier than the E-IIIs, but I still have scope to stuff the cabinet a bit more. As they stand, they are only half stuffed, so to speak.

Limitations - they have a couple. The treble quality is a little bit grainier than I would have liked, but that is a character of the Vifa tweet I'm absolutely certain. It is not offensive, but those that demand CDS-3 level smoothness might not be so chuffed. But at this price point, I'd rather they played music with the vigour and vitality of a good CDX than smooze me with sweet and airy nothings. I'll let Steve describe what he finds tomorrow.

Unlike a closed box loudspeaker, they do sound a bit weird listening from an oblique angle, like side on to them. This is due entirely to the cancellation effects of a dipole and is unavoidable. To this end, I'd hazard that they might be happier to fire down the length of a room than across as mine are at the moment. This will allow a comparatively bigger listening 'sweet spot'. The will still need to be positioned at least 1m clear of the backing wall (as measured from the front of the baffles) to provide adequate breathing room for the dipole to do its job.

Other than that, I couldn't be happier. My next dipole design - tentatively the E-VIs - will feature the same woofer, a Seas Excel M15CH001 mid (now that I know the Vifa woofer is comfortable up to 500Hz), and a yet-to-be determined tweeter, possibly ribbon.

Best I start working on the E-V XO before I get too much ahead of myself.

James
 
Originally posted by James
The treble quality is a little bit grainier than I would have liked, but that is a character of the Vifa tweet I'm absolutely certain. It is not offensive, but those that demand CDS-3 level smoothness might not be so chuffed.
Have you put any thought into using a different tweeter? Since the tweeter is only one small piece, it shouldn't change the overall cost (much).

-=> Mike Hanson <=-
 
Mike,

I've thought long and hard about that one, so the simple answer is yes. But retrofitting a different tweeter will mean a change in the XO design. It won't be a simple plug-n-play affair. It will also mean I have to take a new set of measurements, not to mention route a new upper baffle. In short, a lot of effort will be required for potentially little gain. Also, a significantly better tweeter would mean something like a Scan-speak 2905/9500, which retails for around USD90 each and adding a bit more to the cost of the design.

Don't get me wrong, Mike. The PFM-Special is not at all rough sounding in my opinion. More importantly, it gets all (most of?) the musically important bits right, and sounds engaging to boot. It's just that the E-IIIs have a slightly lighter touch up top, more ethereal if you like, but that is an entirely hi-fi artefact that is not necessary (IMO) for musical enjoyment. Steve is due here any moment, and given that he's become quite accustomed to the sonic signature of my loudspeaker designs of late, let's wait and see what he has to say about their overall performance.

James
 
James,

Looks like you've worked wonders with the little 'un. I take solace in the fact they won't work up against the wall, where my E-III's thrive ;)

Mr Tibbs
 
Mr Tibbs,

You needn't worry at all. There is order in the line-up. Let me explain.

When Steve was here, we did some experiments with concrete paving slabs. These were 450mm x 450mm in size and each weighed exactly 18.4 kg. We tried both the Specials and the E-IIIs with and without them. Here is my take on the proceedings.

PFM-Specials with paving slabs

The biggest change was in the reproduction of voices, where they had more body and substance. Articulation was significantly improved, whereupon music became more intelligible and easier to follow. There were no obvious deletrious effects. The fast and dynamic nature of the loudspeakers was maintained, maybe even improved upon. Clearly an improvement on my suspended wooden floor.

E-IIIs without paving slabs

What was immediately obvious when bringing the IIIs into the room was how the tweeter suddenly disappeared, yet there was plenty of detail and clarity. The OW-1s in the IIIs just get on with doing their job without drawing attention to themselves, and in the process manage to sound exceptionally organic and entirely natural. Perhaps that's what I meant when I suggested that the PFM-Special tweeter was grainier in comparison. It's just a bit more obvious in character, but not disagreeably so I must emphasise.

On the whole, the E-IIIs showed that it had a better grip on musical proceedings, with poise, authority and finesse. It simply sounded better. Then again, they do cost considerably more than the PFM-Specials and in relativity, I guess you can understand why something like the Naim SL2 should sound better than the Naim Intro. But in absolute terms, there is not a lot found wanting for the money. Same goes for the PFM-Special. Only for significantly less money again.

E-IIIs with paving slabs

Given the tall and heavy nature of the E-IIIs, the paving slabs actually made its installation on slabs less stable. It swayed quite easily, but we were curious about the effects of the paving slabs. Cue music. If there was a (musical) difference, it was not significant. But I did note that a frequency sweep in the bass showed that the characteristic loss in my room (via the floor I suspect) between 80 - 100Hz was lessened. Due to the mass and generally inert nature of the E-III cabinets, both Steve and I agreed that the E-IIIs did not gain anything significant from using the slabs, quite unlike the PFMs.

Summary

The PFM-Specials work very well for the money. As you suggest, they are a good little 'un, even if I admit so myself. They make music, they are fun to listen to, and they are easy and cheap to build. But there is also no denying that the E-IIIs are absolutely fabulous. In all honesty, I have some trepidation about hooking up the E-Vs when all their XO bits arrive simply because there is an expectation that they should better the E-IIIs, and both Steve and I feel that will be a tall order. Wish me luck.

James
 
James,

Interesting wee experiment with the slabs. I too have a suspended wooden floor and figured it might be worthwhile beefing up the floor where the speakers sit. I even toyed with the idea of building in a couple of concrete pillars, flush with finished floor level.

In the end I settled on some extra bracing of the joists where the speakers sit, plus a couple of boxes of 3" screws driven in at 9" centres along all of the joists in the room. The screws have a dramatic effect on floor stiffness, as the firm coupling of the floor sheeting to the joists creates a 'T' beam effect.

The end result was that I noticed a slightly better grip on the bass, but nothing too spectacular. The mass of the E-III's seem to make them unfussy about floor quality.

The 'specials' look to be a great way of getting some high performance speakers a very low price. A nice s/h 72/140 would really get them to sing I'm sure, and together would represent a package that would take some beating, even with serious money.

Mr Tibbs
 
I pretty much agree with James' assessment of yesterdays proceedings so I will try to avoid too much repetition.

The PFM specials are excellent at their price point but while they have proved that you can build a good quality speaker for a very reasonable sum they have also demonstrated that for loudspeakers, as for life in general, you get what you pay for. The PFM specials could undoubtedly be made better, but then they would become something different, and possible 50% more expensive, and the original point of the exercise would be lost. A better bet for a higher performance version would be to wait for the EVI.

As a general comparison, the different drivers are more audible then in my Dynaudios, which I was listening to immediately prior going to ‘up country’ or the EIII’s, in that you get more of an impression of listening to a tweeter or listening to a midrange etc. They are not as seamless or quite as coherent as either of the others but they do tighter base than the Dynaudios (which I still think do the best midrange of the three but not by nearly as much as before James rebuilt the EIII XOs).

In my case, paving slabs are a must, as the room beneath my listening room has exactly the same floor plan and, despite having oversized joists, the floor acts a bit like a giant drum. It only took 5 seconds to confirm this in a quick test after loading the slabs into my car. Despite the fact that James speakers sit directly over a supporting beam, the slabs had exactly the same effect on the PFMs as on my speakers, although possibly not to as pronounced a degree, in that voices started to sound more like they were coming from a biological machine rather than a mechanical one. They were both clearer and smoother.

My 450 x 450 paving slabs were not expected to have much effect on the EIIIs, due to the speakers much higher mass in relation to the slabs, nevertheless I did notice some of the same effect on the midrange as for the other speakers. I think the outcome would have been more marked using the 600 x 600 slabs (next size up) as their extra mass and extra width would have somewhat stabilised the speakers, which I suspect might have been able to rock slightly on/with the smaller slabs, in response to base notes, thus softening their edges slightly.

The slabs were sitting directly on the carpet, which is also as I use them. An experiment is in order to see whether sitting them on the heads of screws into the floor will improve things. It may improve base definition by adding stability however it may also remove some of the other advantages of the slabs by coupling them more to the floor.

Steve
 
"The 'specials' look to be a great way of getting some high performance speakers a very low price. A nice s/h 72/140 would really get them to sing I'm sure, and together would represent a package that would take some beating, even with serious money."

Mr Tibbs, thanks for your vote of confidence. I'll let you know how that scenario pans out when Steve finally relinquishes his 72/140 at mate's rates to me. ;)

Given that I had originally hoped to build a latter day AR, replete with walnut enclosures and vintage cloth grills, I think it is the old and big ARs that the 'Specials come closest to replicating in terms of clout, scale, presence and boogie factor. But in new clothes and with a slightly different driver arrangement.

"The slabs were sitting directly on the carpet, which is also as I use them. An experiment is in order to see whether sitting them on the heads of screws into the floor will improve things."

Steve, thanks for your comments. I'm sure it is appreciated by all. Thinking further about the room coupling issue, what we have here is a classical quandary about isolation. There is no denying that coupling a loudspeaker cabinet firmly to a solid floor is beneficial for clarity and definition. But the downside is that any vibration from the cabinet is ultimately passed on to the floor, which will resonate in sympathy at certain frequencies. So we need both coupling and decoupling, which calls for massive bases that are decoupled from the floor, but yet securely anchored in space to provide a reference surface for the loudspeakers and sufficient rigidity into which spurious vibrations can be sunk. I might need to give Mana another go.

Or I can simply lift my carpet and screw my floor boards at small regular intervals onto their joists, ala Mr Tibbs. I think I'll wait until my carpet needs replacement ... which could be a while.

James
 
........ Thinking further about the room coupling issue, what we have here is a classical quandary about isolation. There is no denying that coupling a loudspeaker cabinet firmly to a solid floor is beneficial for clarity and definition. But the downside is....
James
Gut feeling tells me the larger slabs on just carpet will be a better bet than putting the current ones on screws - but I intend to find that out over the next couple of weeks. I wonder if 'anyone' will notice if my slabs suddenly grow a bit?

Extrapolating from the dimensions of the larger slabs and the weight you measured for the current ones, the larger ones should come in at about 32Kg. Since my speakers are 25Kg this will make a dramatic change in the relative masses of the speakers w.r.t. the slabs (from 1.35:1 down to 0.76:1)

Under the EIIIs, the larger slabs would bring the ratio a lot closer to what I currently have. What do the EIII and EV weigh in at?

Steve
 
The E-IIIs and E-IVs are around 50kg and 60kg each respectively. I'm thinking a similarly sized slab of brass, tastefully finished, might get closer to the mass requirement. :D

James
 
Originally posted by James
The E-IIIs and E-IVs are around 50kg and 60kg each respectively. I'm thinking a similarly sized slab of brass, tastefully finished, might get closer to the mass requirement. :D

James
Presumably you mean the E-V not the E-IV. I'm not sure brass is my colour. Have you done a costing?

I am thinking about trying to cast some slabs that are say 500 wide but go 1000 deep and are 55 thick, giving about 62 Kg but not much more front on visual impact than the current ones. This should not only give the mass but the leverage to resist reacting to the movement of the base drivers. Casting them myself, I could also get them to colour match the carpet - which might prove to ultimately be the telling factor in achieving success in the real world. There is a concrete decorative-casting business not far from me, I think I might pay them a visit.

Steve
 
My bad, it's the E-Vs that weigh around 60kg each.

"There is a concrete decorative-casting business not far from me, I think I might pay them a visit."

Have you considered polished granite? Far denser and classier than concrete, don't you think?

James
 
Granite would look good but where from and how much?.

Once I establish the size and shape I need (for the E-V or my version thereof) using relatively inexpensive coloured concrete, then getting a granite version made might be a good idea. Also, if I end up with the system down stairs on a concrete floor then I would feel pretty silly if I had purchased some, then redundant, granite slabs.

Steve
 
Steve,

"I would feel pretty silly if I had purchased some, then redundant, granite slabs."

No need to feel silly. If you purchased the Blue Pearl variant, I'll buy them off you at a mutually agreeable price. The last time I priced granite for the plinth of the E-Vs, I was pretty disappointed so I ain't holding my breath on this one.

James
 
It's good to know you would be there to help out...

I would immagine I might even be able to persuade you to help me test some DIY concrete bases if I asked nicely enough. :)

Steve
 
Since the E-VI is planned (as I understand it) to be an upmarket version of the PFM special using the same base driver (and presumably therefore cabinet volume) but more elaborate construction techniques, better materials and better treble and midrange drivers, would it not be possible for someone to build a version of the PFM special with an upper baffle to suit the new drivers and using the E-VI XO?

This would seem to offer the possibility of a halfway house, using better drivers but the same basic construction of the PFM special. I suspect that the cost of the better drivers would make the cost of the hybrid nearly the same as that of the E-VI but it would still be easily achievable by the woodworking challenged.

Steve
 
"would it not be possible for someone to build a version of the PFM special with an upper baffle to suit the new drivers and using the E-VI XO?"

Well, that kinda depends on how radically different I make the E-VIs. You are right in that I intend to use the same bass driver, but as it is now effectively a special order item only once distributor stocks are depleted, I might go a different route. Not quite decided yet.

As for replicating on the upper baffle of the E-VI to be bolted to the PFM-Special bass bin, the simple answer is yes. But in reality, the constructional approach will need to be quite close because how the baffle is shaped has a considerable effect on the response.

James
 
Thanks, Steve. I've got them but didn't have time last night to strip them off onto my photo server. Might have to wait until the weekend.

James
 


advertisement


Back
Top