Visaton B200 - Taming the shout

[planet10]

Originally posted by johninCR
BTW, Dave I picked up a used pair of 167's that have your phase plugs. Will those fit a B200, since it has the same VC diameter?

Not really. A 1" VC with whizzer requires a smaller diameter plug than if no whizzer. The B200 plugs are also longer.

dave

[Ricky]

I have used the B200 myself in a commercial design. I cured the top end by putting a 0.3mH coil in series with the driver and an R-C network in parallel before the coil to bring the impedance down again.

Ricky.

[scottw]

Geoff H wrote

The circuit is designed to run into a 100K load

Hey,

Doing the correction at the line level is an interesting solution. Is there a way to make the filter suitable to a 50k (or 20k) input? (There is too much I still don't understand about LL circuits.)

Thanks,

Scott

[johninCR]

This is not a substitute for phase plugs or any other worthy mod to the driver. It is an alternative to cutting. We should always endeavour to solve these sorts of issues where the problem starts.

That's why I don't try delving into the electronics side. I leave that to you experts, and stick to exploring the mechanical side. With OB's there's plenty of room to explore, which is quite rewarding.

"I got picked on for no graphs by telling too much of what my ears tell me"

That's a shame. It's the ears that count the most. Why do many not trust them? I look forward to seeing Johns double sided horns, with or without graphs.

Listening to music, moving around in the room, and sweeping through tones with 1hz increments around problem areas may not tell the whole story, but I think it paints a better picture than any other method. Sure our ears/brain can be fooled, but they are incredibly sensitive with highly evolved and flexible data processing. I'm anxious to measure only as a test of my tuning by ear, and to generate FRD's to import into my active EQ/XO software to eliminate the guesswork. Plus there are some instances where measurements will prove helpful.

[ultrakaz]

Hi JohninCR,

I've mounted my B200 in a front horn (200Hz cutoff or so). The throat was around 4.5". I got a boost (2-6db) between 200 to 900Hz and attenuation (1-4db) from 1000-6000Hz and flat after that. Spacing between driver and mouth changes the balance. This is not an exact science. If I knew how to balance the boost/attenuation I'd really be happy.

I'm not surprised you are getting great results with your waveguide. It almost looks like the front loading section of the Tannoy Autograph.

I'd like to construct those waveguides. If you don't mind, I'd appreciate it if you could post details on the construction of the waveguides (dims, process, and caveats). Thanks.

[johninCR]

Ultrakaz,

I'll be sure to share everything, probably more than some want, because I tend to fully explain things. I need to get them to finished form, and once I get something I can listen to my work pace typically slows to a crawl.

These are different since I need to get to finished form before filling voids with sand, which I believe will have a marked improvement. Despite using doubled up real 2"x4" solid laurel to construct the terminus roundovers, the roundovers vibrate a lot, but the relatively thin plywood panels forming the waveguide are quite dead. I may not fully understand the cause of vibration, but I do know 40-50kg of sand per speaker will cure it. I've always wanted some really massive OB's anyway, so this is my chance.

Also, I want to include enough of the science, so others understand it in hopes of furthering the design through technical or aesthetic (esp size) improvements.

[Geoff H]

Impedance scaling. For a 50 K ohm input amp, halve the Rs, and double the Cs. 20K, 1/5 the Rs, 5x Cs.

There are several advantages in LL eq. The amp, be it tube or SS is in direct connection with the VC. There is less chance of ringing and other issues. Reactive components reflect back through transformers, placing strange load conditions on the plates.

Another advantage, the slew rate requirements of the amp are reduced if we are shelving the top end. That may not be a problem with a good amp, but why amplify something and then attenuate it? For too many years, I ran my Wharfedales in a 100w pc system, with 10db l-pads to pull them back to 93dB/W. EL84s would have been a better solution.

Slew rate is a measure of how fast an amp can respond to a signals rise time, usually in Volts per Millisecond. Exceed that, then the feedback circuit can't keep up, resulting in transient intermod distortion.

John wrote "With OB's there's plenty of room to explore, which is quite rewarding"

You bet. I trust you are watching Beyond the Ariel I am currently extracting as much as possible from a pair of 6" F/Rs on OBs only 460mm high. Chasing down panel vibes.

Geoff

[scottw]

Impedance scaling.

Thanks Geoff,

I thought it was that simple, the main requirement being a correspondingly low source impedance.

If I were to use this ciruit, housed with the amp, and also have a CR to roll off the low frequency (around 100 Hz or so) it would go something like: 1M to terminate the interconnect, your LL eq. filter, then the appropriate CR to roll off the lows(this R is also setting the input impedance for the amp)? Just wondering if the CR would be interacting with the eq. filter.


Scott

[Geoff H]

Hi Scott. I would have the bass roll off first. Use lower values, so that the following network doesn't load it too much. ie the shunt R = 10K, Xc=10K, and I would try to stick with 6dB/oct.

I had a look at some of the other links provided. Most seem to be similar with that on the Visaton site. Ricky's situ is a bit different, as it was a commercial design. (Interesting reading about R-A, will the real Richard Allen / Allan stand up?) Trying to ship a LL Filter with a speaker system could give the impression of an afterthought. Magazines would have a field day with it.

However, in the DIY world we can be more flexible, design for performance, not just convenience, and benefit from the effort.

Just a word on capacitors. I think good quality polystyrene are as good as any. Silver Mica were normally used at RF. I don't know of any advantage at AF. S/Ms offer greater temperature stability, not one of our requirements.

Geoff.

[Geoff H]

Speaking of afterthoughts: My network has an input Z of 84K at low freqs. A series C with XL of 84K at the desired freq and your done at 6dB/oct. ie 0.018uf for 100 Hz.

Now for a bonus. An additional pot with a series R shunting after the input C, will give you an adjustable bass roll off freq. 80 to 240 Hz could be very useful in setting it up.

Cheers,
Geoff.