Hi Everyone,
I was perusing the internet, and after finding B&W 801 clones on Ali Express I came across a request for help regarding a Snell type D. A 3 way floor stander with a rear tweeter. Never mind why they wanted help, what it reminded me of is just how BIG and MESSY snell crossovers all seemed to be for the era. Below is an example.
I'm wondering if anyone else has noticed this and has any comments. Of course, I'm a fan of those Snells but looking at these now I kind of shiver. What are your thoughts on the driving principles that got the late, great Peter Snell to these designs?
The only other crossovers I can think of remotely like this might be the M&K S-1B crossovers I got to look at long ago.
I was perusing the internet, and after finding B&W 801 clones on Ali Express I came across a request for help regarding a Snell type D. A 3 way floor stander with a rear tweeter. Never mind why they wanted help, what it reminded me of is just how BIG and MESSY snell crossovers all seemed to be for the era. Below is an example.
I'm wondering if anyone else has noticed this and has any comments. Of course, I'm a fan of those Snells but looking at these now I kind of shiver. What are your thoughts on the driving principles that got the late, great Peter Snell to these designs?
The only other crossovers I can think of remotely like this might be the M&K S-1B crossovers I got to look at long ago.
It is my understanding that Snell used a reference speaker for all of their builds. Every speaker that left the factory was compared to the reference. So, with variations in drivers and crossover components, there was a lot of hand trimming of crossover values. Additional caps and hand winding/un-winding of handmade inductors. Snell speakers traditionally used higher order filters, thus more components. It does look kinda messy and the inductor arrangement could be better, but it's hidden inside the cabinet and got the job done!
Dan
Dan
There were a few years in the late 80s and early 90s where loudspeaker designers had discovered computer aided loudspeaker design.
This is speculation on my part, but there were a few years there when designers were kinda "fetishizing" achieving textbook response curves for each band in the loudspeaker. I remember a review of a Thiel speaker in particular, where they were bragging about the size and the complexity of the crossover, despite the fact it uses first order slopes.
In other words:
As the debate raged for a few years there, I think manufacturers mostly stopped doing it, especially because it was pretty easy to end up with a xover that cost more than the loudspeaker drivers. I'd argue that the embrace of ULTRA simple designs, even crossoverless designs, was a bit of a 'pushback' against that trend of huge complex crossovers. It was around 1998-ish that it seemed like people "re-discovered" Lowther and Fostex and various other type of full range or nearly full range speakers.
This is speculation on my part, but there were a few years there when designers were kinda "fetishizing" achieving textbook response curves for each band in the loudspeaker. I remember a review of a Thiel speaker in particular, where they were bragging about the size and the complexity of the crossover, despite the fact it uses first order slopes.
In other words:
- It's true that a carefully designed speaker with first order xover slopes can reproduce a really nice step response and potentially even play back square waves
- But due to the introduction of computer aided loudspeaker design (IMHO), designers began obsessing about getting the individual response curves to be nearly perfect
As the debate raged for a few years there, I think manufacturers mostly stopped doing it, especially because it was pretty easy to end up with a xover that cost more than the loudspeaker drivers. I'd argue that the embrace of ULTRA simple designs, even crossoverless designs, was a bit of a 'pushback' against that trend of huge complex crossovers. It was around 1998-ish that it seemed like people "re-discovered" Lowther and Fostex and various other type of full range or nearly full range speakers.
More fun when you add charge coupling. (that was done under the cap and resistor board).
This is my experience with the Snells I've owned, every crossover is a little different and quite obviously tuned to the drivers.
The rule of thumb is measure what you have in the capacitors and replace with like, and don't mess with the inductors.
Serious QC at the production line incorporates adapting crossovers (mainly part values) to individual driver responses (and combined response). Any self-respecting company would do so, KEF did that for ages for their reference models.
The coupling between those adjacent air-cored inductors will be horrendous. Obviously, nobody told them about right angles.
Those Snell speakers were known to be very neutral for their time. Now we would do that with dsp, but in that time it had to be done analog, and without computer measeurements or simulations. So it's trial and error untill it was right, and crossovers got complex because they tried to eq the speaker to a preference curve with the crossover.
Simulations were available with a calculator and pencil, using measured responses and impedance. A bit slow by today's standards but quite doable.
Just got done skimming the article that John Broskie wrote on his cross over improvements. Seems that a Zobel network has some advantages. Then splitting the inductors in half, using them in a series configuration of sorts loses me. This stuff reminds me of a mobile over my crib.
Well, I think Excel was a thing already by then, so hopefully not literally using a pencil, but definitely the early days of computer aided crossover design.
I think this has fallen out of favor thanks to more consistent driver manufacturing, but in the 1980's this was certainly more important.
Also, given just how really terrible some "high end" speakers measure I don't know if anyone even cares anymore.
What -in commercial hi-fi? Probably not all that many -for all the claims about demanding 'accuracy', to my eyes most tend to gravitate toward systems with some kind of 'euphonic distortion' (using the generous, and sometimes generously-meant term 😉 ).
And I suppose -why not? It's just that people 'need' to understand the different priorities rather than assume it's a one-size-fits-all. Hi-fi as originally envisioned was technically 'the most accurate possible reproduction of what's been recorded'. It can't be 'more' -it can certainly be less. But that's not always everybody's goal anyway -after all, this is just a hobby, not a dictatorship. In fact, I suspect it's not the goal of 95% of the people who are into what they call 'hi-fi', and probably an even higher percentage of 'audiophiles', whether they realise (or acknowledge) it or not. What they're really doing is optimising systems to produce a sound they like on a given day or period in history. Their money, their ears, their eyes, their choice. It's because it all gets lumped under that catch-all term, and many honestly believe they're after 'faithful reproducton' [matron] that's a major cause of confusion. That and a shed-load of marketing twaddle which hasn't helped matters much.
And I suppose -why not? It's just that people 'need' to understand the different priorities rather than assume it's a one-size-fits-all. Hi-fi as originally envisioned was technically 'the most accurate possible reproduction of what's been recorded'. It can't be 'more' -it can certainly be less. But that's not always everybody's goal anyway -after all, this is just a hobby, not a dictatorship. In fact, I suspect it's not the goal of 95% of the people who are into what they call 'hi-fi', and probably an even higher percentage of 'audiophiles', whether they realise (or acknowledge) it or not. What they're really doing is optimising systems to produce a sound they like on a given day or period in history. Their money, their ears, their eyes, their choice. It's because it all gets lumped under that catch-all term, and many honestly believe they're after 'faithful reproducton' [matron] that's a major cause of confusion. That and a shed-load of marketing twaddle which hasn't helped matters much.
In fact, that is exactly what I am in the process of doing right now. I know what I like to hear, and as Walter Cronkite says, "That's the way it is."