I have a cunning plan. HP Filter.

[Kipper4]

I've seen them there King Tubby HP Filters and I wanna try that for myself.
We're talking Dub of course.
There's some altec 9069 HPF selling for money I don't have and I'm wondering what's inside that mahoosive box since it's advertised as a passive HPF.

My intent is to use it in conjunction with my mixer.
So to me line level seems like it might work best.
I Like the big knob with set points from 70hz to 7khz.
Anyone got any ideas what's in the box since I can't find and dub filter schematics.

If not what can I use preferably to fit in a stomp box enclosure.
Thanks for reading and any advice is great fully received .
Thanks
Rich

Dub skiddley dub, skiddley dub dub doooo.
SIREN
ECHO N VERB WASHED.

Confused yet? Go here for more info....

https://audio-merge.com/

[Mark Hammer]

Nice to see you back here, Rich.  Been a while.  Hope all health issues are well behind you.

Cheers from the land of snow.  I was hoping to finish a few things this afternoon, but I suspect the driveway is going to occupy my afternoon.  This is the view out of my workshop window at the moment.

[Rob Strand]

My understanding is the circuit is like this,



You could easily translate that to an opamp circuit without inductors however emulating a pot for RL will take a some experimenting to come-up with a circuit.

[Kipper4]

Hey Mark. Thanks for the welcome back.

I don't envy you with the snow and driveway.
A bit of good music should help pass the time though and a stout shovel.
We're on a similar latitude to you and the snow is due here anytime, it's the same most years.

Hope you and yours are keeping well?


Stephen will be jealous in the heat in Aus........

[Rob Strand]

Here's some official documentation:

https://www.vintageradio.com/history/Altec1966.pdf

The Rs and RL are present by virtue of the source impedance and load impedance.

[Kipper4]

Thanks Rob. Saved to ebooks. Smashing.

[ElectricDruid]

Here's some official documentation:

https://www.vintageradio.com/history/Altec1966.pdf

The Rs and RL are present by virtue of the source impedance and load impedance.

"weighs 15 pounds"!! Ok, so there's some big iron in there! That's for the dual LP/HP unit, but still...that's several kilos per filter!!

18dB/Oct isn't a slope you see that commonly these days, but there's no reason it can't be done. I can think of several ways, *far* from the original passive design.

[Kipper4]

Hey up man. Hope you're ok?

"Far" from is fine as long as it's not above my pay grade I'd like to make something to try.
I did wonder about an Svf too. At least I've made them before.
Take care Tom.

[PRR]

They do not do anything you can't do a LOT cheaper with a dual op-amp and some R and C.

Get Lancaster's Active Filter Cookbook. https://www.tinaja.com/ebooks/afcb.pdf

The Altec is a 3rd-order high-pass. Switched frequency. Oh heck, Don's image is so good you may not need any other info. Altec's alignment may have been near "Flattest" and that also leads to round R values. (RD2 only has to equal Rd1 but can be 3k to 100k.) He suggests varying C to shift frequency but for a few octaves varying R works fine and you have much more picks of R values in the catalog.

[Rob Strand]

Here's some ball-park estimates for the part values.

Read the comments on the spreadsheet.   The Altec response plots and the textbook circuit values don't quite match up.  Normally a textbook constant-k filter design ends up with a Butterworth response.   

I added a tweak parameter 'x' which you can see bends the circuit towards the plots.   The way the bending is done is kind of in the spirit of a constant-k design.   You can see the uncertainty in the part values is then 20%.

Some uncertainty from back-engineering the response can come from discrete cap and inductor values in the real unit.   No doubt the inductor is a custom inductor, probably multi-tapped, so there's no reason the inductances have to stick to standard values.   The tweaked cap values are perhaps closer to standard values.





It occurred to me later the reduced attenuation at freq/5 could be due to the DC resistance of the inductor.

The size of the inductor determines the resistance and the distortion.   We have no distortion spec on the unit although we do have a maximum voltage.   The attenuation at freq/10 could be used to work out a DC resistance for the inductor but each frequency setting will a different DC resistance because the inductor is different.   Another issue is at the lower frequencies, where the size of the inductor is more demanding, the plots don't go low enough to see the attenuation at freq/10.

All in all there's not enough info in the catalog to split hairs.   

All I can say is my tweak shouldn't be treated as the answer.  The inductor DC resistances are an equally valid explanation, perhaps even more likely.

To get the right attenuation at freq/5 for the 1kHz range a DC resistance of around 15 to 20 ohms would be a good ball-park.   However that could mean the DC resistance of the 70Hz range is somewhat higher - it can't be too high as it would change the roll-off too much.

[duck_arse]

nothing but pouring rain and flash flooding here. glub glub glub mate.

so will this be the U-Rich, I-Rich or King Kipper soundsystem? Struth at the Controls. will the knob be big enough.

I looked at some gut-shots of one of those altec boxes, dual filters - man oh boy, there is some mechanicals going on in there. and thousands of those black capacitors.

[ElectricDruid]

Hey up man. Hope you're ok?

Oh, y'know, surviving.

"Far" from is fine as long as it's not above my pay grade I'd like to make something to try.
I did wonder about an Svf too. At least I've made them before.

SVF would be possible, but how about just cascading three voltage controlled single-pole HPF stages?

There's a nice example in the SSI2164 datasheet, Figure 13, page 12:

https://www.soundsemiconductor.com/downloads/ssi2164datasheet.pdf

This is shown with the SSI2164 (obviously) but it can be done with the V2164 or AS2164 instead, although they use 30K resistors in place of the SSI's 15Ks. You'd get one VCA left over, so you could add level control too.
Lowpass would be basically the same thing, except with the very-slightly-simpler lowpass stage.
Since it's a voltage-controlled filter, the control is dead easy. Also the 2164 has an exponential control response, so it follows octaves nicely. Very handy. You'll get a wide range (many octaves) for a CV between 0V and -3.3V (relative to whatever the chip thinks "ground" is) so you can set up particular frequencies with a chain of resistors and a rotary switch like the original, or use a pot with resistors top and bottom to trim the range a bit. And a ma-hoooooss-ive knob, obviously

The 2164 will run down to +/-4V, so it's ok on a 9V supply if you want to make it into a pedal, but for a desktop dub box, I'd probably make a +/-12V or 15V supply.

For the name, I think the King Kipper Soundsystem sounds brilliant. I can smell the ganja smoke from here.

[PRR]

> cascading three voltage controlled single-pole HPF stages?   SVF...

I thought we wanted "King Tubby HP Filters", 3-pole passive switched?

There's an infinity of filter types. But a specific type was specified. Do we need to tromp all over the field first?

[ElectricDruid]

I thought we wanted "King Tubby HP Filters", 3-pole passive switched?

Well, we *could* that, but that's what Altec already did, so it's not so much fun as designing something else to do the same job!

In some ways, we can swap mechanical complexity and big chunks of iron for electronic complexity. I was wondering how you'd avoid needing triple-gang pots or multi-pole switches to control an 18dB filter, which is why I thought of a voltage-controlled design. Makes the control side much simpler, at the cost of a more complicated filter circuit.

[Rob Strand]

Another alternative is to switch components using analog switches, and perhaps two push buttons (up/down).

[PRR]

> how you'd avoid needing triple-gang pots or multi-pole switches

Build 12 filters. The opamps are 19 cents a pair. The C will be much more but you have to buy those in any case. A 12-throw switch is a buck or so. Copy-paste-paste-paste.. your layout. You could even opt to make a few other-type filters in the assortment: 7-pole low-pass, bypass, whatever rings your chimes.

Yes, Altec done it, sold a few, most of those got scrapped in studio renovations. They are apparently valuable now. There are clones of various types (heavy iron box to software emulation).

[ElectricDruid]

> how you'd avoid needing triple-gang pots or multi-pole switches

Build 12 filters.

Yeah, that works too!

[Kipper4]

Thanks for the great input fellas. I have yet to decide what to make.

[Shoeman]

My reply has nothing to do with the technical aspects of your question, it's towards the title of the thread.  I hope you were able to make Baldric and Lord Percy understand your cunning plan.   
  I also got to thinking that I really don't know what any dub stuff sounds like. Then a light bulb went off from my younger days....
  "my daddy was a bankrobber, he never hurt nobody"   Does that count?

[nocentelli]

I also got to thinking that I really don't know what any dub stuff sounds like.