New Posts New Posts RSS Feed - Adventures In Transmission Line Loudspeakers
  FAQ FAQ  Forum Search   Events   Register Register  Login Login

Welcome to the Graham Slee Audio Products Owners Forum

 

Open to all owners plus those contemplating the purchase of a Graham Slee HiFi System Components audio product and wishing to use this forum's loaner program: join here (Rules on posting can be found here)

This website along with trade marks Graham Slee and HiFi System Components are owned by Cadman Enterprises Ltd


Adventures In Transmission Line Loudspeakers

 Post Reply Post Reply Page  <1 19202122>
Author
Graham Slee View Drop Down
Admin Group
Admin Group
Avatar
Retired

Joined: 11 Jan 2008
Location: South Yorkshire
Status: Offline
Points: 16298
Post Options Post Options   Thanks (1) Thanks(1)   Quote Graham Slee Quote  Post ReplyReply Direct Link To This Post Posted: 06 Feb 2017 at 9:19am
Now using SPICE modeling to be able to see the filter response and it showed me a lot more than the online caculators.

All wound components comprise a resistance and inductance in series, and so, the driver is resistive and inductive. The specs say so. And therefore the load comprises 5.4 Ohms and 21.3uH.

The filter feeding this is an inductor: which cannot escape having resistance, in this case 0.5 Ohms; in parallel with a resistor. The values decided upon are 80uH and 8.2 Ohms.

The product is a notch filter! The theory holds that it will not be a notch filter, but will simply shelve down. The practice is different. The driver inductance causes the output to rise further up in frequency.

To prevent this happening, the online calculators say you need a zobel network. The simulation model shows it to be ineffective - very little changes. However, there is a bigger phase difference using it.

As I've commented many times to many people, phase differences are to be avoided as they confuse the brain. But phase differences outside our hearing range don't.

The situation where you have -45 degrees and -3dB level is OK, but where the level does not change (i.e. it is still flat), the brain has trouble in dealing with it as it is abnormal. A confused brain is an agitated brain, and a music system causing that is also very fatiguing because the brain is working hard to correct something that can't be corrected. It will either get your back up, or if it's less accute, will make you zed off (fall to sleep).

Why so many hi-fi manufacturers don't accept or even know this I will never understand. Possibly it's because it isn't in their text books.

The phase response given by the above component values gave the biggest phase difference: 18 degrees, at 7.5kHz, 3dB down, which should result in confusion, but as the driver response is like a jagged rock, it will already be quite confused. Looking at my measured driver response there is a peak roughly at this frequency, which goes the opposite direction to the 3dB cut, and may be levelling things out.

The filter's aim is to bring down the quite pronounced peaks in the 10-12kHz region, which are 4-5 dB up, and the filter opposes that level by roughly the same amount. Here the phase is about 16 degrees out, and at least it's not 45 degrees which would lead to a confused state.

I tried a few running modifications, such as adding a parallel capacitor to the filter (0.1uF), which altered it from 1st to 2nd order - not that you could see much difference in the amplitude plot over such a narrow range - but the filter slopes became slightly steeper, and the listening got a whole lot harder. Likewise, the zobel network when tried, even though the simulation said was not going to be as effective as the simple maths would suggest, made the going a lot harder.

So, now I end up with a simple shelving filter which by accident is really a notch filter, and I'm just about happy with the HF response. Speech sounds just about right, and most of the veiling caused by having too much harmonic content, is gone.

The inductor in the filter brings 0.5 Ohm series resistance to the prototype, which given a zero impedance drive, results in a damping factor of just 10. It will be more like 5 - 8 when amplifier output impedance is factored in. There will not be much cone control, but this is common to most loudspeakers and makes a mockery of those who champion amplifier damping factor.

So, taking stock:

1. Baffle step has been taken care of by making the cabinet 12" wide, so sound level blends nicely into room reinforcement frequencies.

2. Internal reflections have been reduced by interior padding which removes a lot of coloration.

3. The above mentioned filter (resistor and inductor) remove the top end sharpness.

The imaging is still reasonably good with nothing seeming to come directly from the drivers.

Now to hopefully finally tackle the bass. The final box is going to be just about 6" deep to keep the same volume, seeing as the cabinet will be considerably wider. A front mounted port will have to be short to allow enough air space between its back end and the back of the cabinet. This suggests a small diameter, but that can result in "chuffing".

Because the driver, box, room combination results in 60-63Hz boost, my thought is to use the port to boost at 0.7 those frequencies, which should steer away the attention from 60-63Hz.
That none should be able to buy or sell without a smartphone and the knowledge in how to use apps
Back to Top
JamesD View Drop Down
Senior Member
Senior Member
Avatar

Joined: 05 Nov 2012
Location: Bolsover
Status: Offline
Points: 246
Post Options Post Options   Thanks (0) Thanks(0)   Quote JamesD Quote  Post ReplyReply Direct Link To This Post Posted: 11 Feb 2017 at 7:08am
Does it make things easier and better to do all this correction using passive components or would there be any benefit in doing it actively with a unit that goes in series with the input to your power amps? Just curious :-)

Edited by JamesD - 11 Feb 2017 at 7:09am
Aren't ears brilliant
Back to Top
Drewan77 View Drop Down
Senior Member
Senior Member
Avatar

Joined: 25 May 2013
Location: Chester,England
Status: Offline
Points: 1544
Post Options Post Options   Thanks (0) Thanks(0)   Quote Drewan77 Quote  Post ReplyReply Direct Link To This Post Posted: 11 Feb 2017 at 7:50am
James, it's certainly easier & more convenient to make changes, measure, listen, adapt etc by going active but I'm not sure a 'purist' like Graham would like that, especially as he knows exactly what he is doing  & has components already to hand (also the cost).
Older than I once was, younger than I'll be
.............................
Andrew
Back to Top
Graham Slee View Drop Down
Admin Group
Admin Group
Avatar
Retired

Joined: 11 Jan 2008
Location: South Yorkshire
Status: Offline
Points: 16298
Post Options Post Options   Thanks (0) Thanks(0)   Quote Graham Slee Quote  Post ReplyReply Direct Link To This Post Posted: 11 Feb 2017 at 7:40pm
Because of the backlog of work piling up after spending a considerable period of time experimenting with the full-range drivers, hoping eventually to achieve some sort of transmission line result, I seriously need to catch up, and so must put my contributions to this topic I started on hold.

Just like all my designs, they don't happen easy, and they don't just happen for the sake of wanting to show I can make particular product. There's a lot more to it than that. If they don't do it for me then what chance have they in doing it for you?

But I still had to overcome the room/speaker imbalance to reinstate my reference Stirling LS3/5A speakers, and went searching for help, but there's not much about, and what there is isn't very helpful - mainly advice from wannabes suitable for wannabes.

I found a paper written by the BBC which helped a little, and found they too have room/speaker problems to address (or did...).

Part II of this paper was of real help: http://downloads.bbc.co.uk/rd/pubs/archive/pdffiles/monographs/bbc_monograph_78.pdf

After running tones at the frequencies the room/speaker imbalance happens, I was able to position the LS3/5A's on their stands in their best positions, which will never be ideal, but close enough for me to monitor my work.

Accuracy of position was unbelievable - an inch one way or another made a profound difference - but eventually I was able to get sufficient adjustment to minimise what the sound block plasterboard was causing (that stuff ought to be blacklisted!).

Listening position was also troublesome - a few inches nearer than the optimum and the bass boom was OTT.

Imaging still wasn't perfect, but torquing-up the bolts on the left hand stand brought it back into focus.

And finally, dealing with reflections. There was a strong one to my left at the position of the "false window" curtains put there just for the purpose. Obviously not absorbent enough, and so a large piece of polyurethane foam was attached to the wall behind them.

I then found a less obvious reflection from the patio door on my right - you could "see" the "extra speaker" and right in the middle of the midrange - around 500Hz. Again, the curtains were too thin to absorb this, but luckily I had an identical spare curtain which I was able to double with the existing one.

This set the sound stage symmetrical, and actually better than I'd had it in the other room.

So, does it work? Well, I can definitely hear the difference between two versions of the Accession MC. One has a standard fixed integrated voltage regulator, the other an experimental discrete voltage regulator. The difference is now immediately noticeable where it wasn't before. I'll not go into which regulator sounds best to me, but at least I can get on with my job.

That none should be able to buy or sell without a smartphone and the knowledge in how to use apps
Back to Top
ICL1P View Drop Down
Senior Member
Senior Member
Avatar

Joined: 04 Aug 2012
Location: Faringdon, Oxon
Status: Offline
Points: 2341
Post Options Post Options   Thanks (0) Thanks(0)   Quote ICL1P Quote  Post ReplyReply Direct Link To This Post Posted: 11 Feb 2017 at 7:54pm
It's been interesting and I look forward to hearing more if/when you return to it, but it's good to hear you'll be back to development work. I take it Accession MC is the priority? Would also like to hear more on the rim drive TT.
Ifor
=====
Reflex M & ACCESSION M, CuSat50, Majestic DAC, a Proprius pair.
Back to Top
Richardl60 View Drop Down
Senior Member
Senior Member


Joined: 04 Nov 2014
Location: Yorkshire
Status: Offline
Points: 1468
Post Options Post Options   Thanks (0) Thanks(0)   Quote Richardl60 Quote  Post ReplyReply Direct Link To This Post Posted: 11 Feb 2017 at 9:39pm
Here here, great to read this thread but the show must go on and look forward to Accession MC developments.
Back to Top
Fatmangolf View Drop Down
Moderator Group
Moderator Group
Avatar

Joined: 23 Dec 2009
Location: Middlesbrough
Status: Offline
Points: 8958
Post Options Post Options   Thanks (0) Thanks(0)   Quote Fatmangolf Quote  Post ReplyReply Direct Link To This Post Posted: 11 Feb 2017 at 9:42pm
At the risk of being selfish about my passion for Graham's work on Hifi System Components, I'm glad we are back we are back on the development and production line. If the treatment of new room makes the Stirling LS3/5A speakers as good as they were in the workshop I think we have some more treats ahead of us. Thumbs Up
Jon

Open mind and ears whilst owning GSP Genera, Accession M, Accession MC, Elevator EXP, Solo ULDE, Proprius amps, Cusat50 cables, Lautus digital cable, Spatia cables and links, and a Majestic DAC.
Back to Top
 Post Reply Post Reply Page  <1 19202122>
  Share Topic   

Forum Jump Forum Permissions View Drop Down

Forum Software by Web Wiz Forums® version 12.01
Copyright ©2001-2018 Web Wiz Ltd.

This page was generated in 0.094 seconds.