Tube DIY Asylum

Do It Yourself (DIY) paradise for tube and SET project builders.

Return to Tube DIY Asylum


Message Sort: Post Order or Asylum Reverse Threaded

Looking for a high pass filter kit

173.75.242.249

Posted on July 4, 2014 at 18:53:01
Cleet Torres
Audiophile

Posts: 188
Joined: July 11, 2013
Preferable tube, being a tube head, but not essential.

I'm planning to bi amp my speakers. I'm leaving the speaker crossovers intact. I want to use a SET amp for my tweeters and only feed around 500Hz and up into the set. If the SET doesn't need to work on the low frequencies I'm sure it's will sound sweeter.

The transistor power amp I'mm be using for the bass I don't really care and I doubt it's will make an audible difference if I do remove the high frequencies from its input.

Thanks
Cleet.

 

Hide full thread outline!
    ...
In the meantime, posted on July 4, 2014 at 20:28:26
1973shovel
Audiophile

Posts: 10117
Location: Greenville SC
Joined: February 25, 2007



You could try the simple DIY High-Pass filter shown in the article linked below. Simple to build, and they'll plug right into the RCA input jacks of your SET.

Footnote #7 gives you the formula to calculate the frequency you're shooting for.

 

RE: Looking for a high pass filter kit, posted on July 4, 2014 at 20:31:47
Tre'
Industry Professional

Posts: 17302
Location: So. Cal.
Joined: February 9, 2002
"I'm leaving the speaker crossovers intact. "

That's actually a big mistake.

One of the big advantages of bi-amping is getting rid of the crossover parts between the amp and the woofer.

If you build a Sallen-Key low pass filter and place it before the SS amp, removing the passive crossover parts for the woofer, it will sound much better than what you have planned.


Tre'


Have Fun and Enjoy the Music
"Still Working the Problem"

 

RE: Looking for a high pass filter kit, posted on July 4, 2014 at 21:31:18
Eli Duttman
Audiophile

Posts: 10455
Location: Monroe Township, NJ
Joined: March 31, 2000
Check out Phil Marchand's offerings. An all tubed electronic crossover does not come cheap. Phil's opamp based stuff is not junk and might be your best bet.


Eli D.

 

RE: Looking for a high pass filter kit, posted on July 5, 2014 at 03:41:18
kugelwellen dude
Audiophile

Posts: 4
Joined: June 27, 2014
How do you place a filter before an amp?

 

And if you don't like foot-notes, posted on July 5, 2014 at 06:51:15
Chip647
Audiophile

Posts: 2651
Location: The South
Joined: December 24, 2012

 

The XM1 was the very first thing I ever soldered!, posted on July 5, 2014 at 07:04:09
Chip647
Audiophile

Posts: 2651
Location: The South
Joined: December 24, 2012
I still have them.

 

Thank you!, posted on July 5, 2014 at 07:05:17
1973shovel
Audiophile

Posts: 10117
Location: Greenville SC
Joined: February 25, 2007
That's WAY easier than me having to relearn how to use my scientific calculator all over again, on those rare occasions I need it!

 

RE: In the meantime, posted on July 5, 2014 at 07:17:55
Cleet Torres
Audiophile

Posts: 188
Joined: July 11, 2013
Thanks shovel, that is a nice easy and simple way of achieving what I want.

 

RE: Looking for a high pass filter kit, posted on July 5, 2014 at 07:25:45
Cleet Torres
Audiophile

Posts: 188
Joined: July 11, 2013
Thanks Eli. I had seen them before years ago and forgotten about them.

Regards
Cleet

 

RE: Looking for a high pass filter kit, posted on July 5, 2014 at 08:05:44
Tre'
Industry Professional

Posts: 17302
Location: So. Cal.
Joined: February 9, 2002
Preamp out to filter in, filter out to power amp in, power amp out connected directly to speaker.

Tre'
Have Fun and Enjoy the Music
"Still Working the Problem"

 

Completely agree with Tre'..., posted on July 5, 2014 at 08:44:33
Lew
Audiophile

Posts: 10911
Location: Bethesda, Maryland
Joined: December 11, 2000
One of the major benefits of bi-amplifying is getting rid of reactive components between amp and speaker. I strongly suggest you do that. But also, you mention choosing 500Hz as a crossover point. In my direct experience, the ear is very sensitive to qualitative differences in that region of the audio spectrum, and I would recommend you cross over at no higher than 200 Hz, for maximum invisibility of the transition between the two amplifiers. Lower is even better. Any decent SET should be able to power your high efficiency speakers down to 100 to 200Hz, I would think.

 

RE: Completely agree with Tre'..., posted on July 5, 2014 at 09:16:09
Tre'
Industry Professional

Posts: 17302
Location: So. Cal.
Joined: February 9, 2002
Lew, I don't think Cleet is crossing at 500Hz. He just wants 500Hz and below out of the amp driving the tweeter and it's passive crossover parts.

"I want to use a SET amp for my tweeters and only feed around 500Hz and up into the set. If the SET doesn't need to work on the low frequencies I'm sure it's will sound sweeter."

Of course, as you know, with the SET amp connected to a tweeter through a passive crossover the impedance of that load at the lower frequencies will be very high and the SET will not be doing any "work" at the lower frequencies to start with.


Tre'


Have Fun and Enjoy the Music
"Still Working the Problem"

 

RE: Looking for a high pass filter kit, posted on July 5, 2014 at 09:54:03
drlowmu
Manufacturer

Posts: 9730
Location: East of Kansas City
Joined: January 10, 2005
That is the wrong way to biamp !!

TWO things are negatives to consider carefully :

(1) You will have to go through an active crossover, which involves one extra active electronic stage and parts for the signal to transverse, extra cabling,........... ALL with their attendant losses.

(2) On any good system, you will EASILY hear the different timing and signatures of differing amps, differing topologies, and it will be far less than optimal. The ear - brain can easily detect that.

Optimal to biamp is to have SAME AMPS EXACTLY on top and bottom, and same wire and wire lengths also. But then, you STILL have Problem (1) as above, a whole other crossover with its attendant losses to go through before you even get the signal to the amps.

Ideal is often the SIMPLEST : a single, great, truly great amp, a full-range amp, and a passive speaker-level crossover,........ if into tubes : a high efficiency speaker, well wired.


Have fun.

Jeff Medwin


 

RE: Looking for a high pass filter kit, posted on July 5, 2014 at 12:37:31
Skip Pack
Audiophile

Posts: 144
Location: Hollister, CA
Joined: November 20, 2005
This is one of the few commercial examples of a passive filter that serves could be used for a simple crossover (though this one is not).

http://bottlehead.com/smf/index.php/topic,5145.0.html?PHPSESSID=29ssr4qa9tq1gacs57fu1thgb6

I use a fixed equivalent of this in my system to compensate for open baffle low frequency cancellation. The key to predictable performance is knowing and setting up the filter to deal with your preamp/source output impedance and power amp inmput impedance correctly. Done right it works very well.

Skip

 

Isn't the net result the same?, posted on July 5, 2014 at 18:29:21
Lew
Audiophile

Posts: 10911
Location: Bethesda, Maryland
Joined: December 11, 2000
Some other amplifier will have to provide the drive for frequencies below 500Hz. Doesn't that amount to crossing over at 500Hz? And part of my thinking also was to eliminate that passive crossover between the amp output and the tweeter, if possible in favor of using a simple capacitor on the input side of the tweeter amplifier, which could equally well provide the necessary high-pass filter. Anyway, I don't want to be re-inventing his wheel.

 

RE: Isn't the net result the same?, posted on July 5, 2014 at 18:46:21
Tre'
Industry Professional

Posts: 17302
Location: So. Cal.
Joined: February 9, 2002
"Some other amplifier will have to provide the drive for frequencies below 500Hz."

Yes but that amplifier and speaker could be playing well above 500Hz.

Again I think he just wanted to limit the frequencies going to the amp that's using the passive crossover to cross the tweeter at the original crossover frequency (one that I'm assuming is a good amount higher).

"Doesn't that amount to crossing over at 500Hz?"

Not really. Again I think he just wanted to limit the frequencies that the tube amp has to deal with, not cross the tweeter at 500Hz.

I could be all wrong about what he trying to do. He wasn't all the clear.

I do hope he will try pass filters (used as the crossovers) in front of the amps and eliminate the passive filters between the amps and speakers.

I like active low pass on the woofer amp and just a cap for the highs but that will depend on the frequency response curve of his tweeters and how that all lines up with his desired crossover point.

One caveat, to protect his tweeter from "bumps in the night" he might want to add a cap between the amp and tweeter with a -3db point well below his crossover point. No sense in losing a tweeter from a turn-on surge.

Tre'
Have Fun and Enjoy the Music
"Still Working the Problem"

 

Oh dear ... let me rephrase, posted on July 5, 2014 at 19:26:04
Cleet Torres
Audiophile

Posts: 188
Joined: July 11, 2013
I don't want to remove my speaker crossovers, I have no intention of removing my speaker crossover and I am not going to remover my speakers' crossovers.

My need is very simple. Use a SET for the tops but spare it from having to consume what little energy it has to amplify frequencies the speaker crossover will block, but leave enough so that the speaker crossover will work as designed.


 

RE: Looking for a high pass filter kit, posted on July 5, 2014 at 19:27:03
Cleet Torres
Audiophile

Posts: 188
Joined: July 11, 2013
'Optimal to biamp is to have SAME AMPS EXACTLY on top and bottom'
I cannot figure out what the "optimal" part is about listening to high frequencies using amplification that I find unapealing at high frequencies.


'you will EASILY hear the different timing and signatures of differing amps".
Exactly, that is what I am trying to achieve.

"and same wire and wire lengths also"
Poppycock old fellow. YOU will NOT be able to tell the difference between lets say 5 vs. 6 foot of the same cable ever. Not you, not me, not any single living human being.

 

RE: Completely agree with Tre'..., posted on July 5, 2014 at 19:45:22
Cleet Torres
Audiophile

Posts: 188
Joined: July 11, 2013
'Of course, as you know, with the SET amp connected to a tweeter through a passive crossover the impedance of that load at the lower frequencies will be very high and the SET will not be doing any "work" at the lower frequencies to start with.'

No, I don't agree with that one.

For starters, the intermediate stage(s) will not see the low frequency impedance of the speakers and be amplifying the complete signal w/o any change.

Secondly, I have a hard time thinking the output tube would not be amplifying everything it sees on the plate. Especially in configurations like parafeed. Best case lower frequencies will be amplified at a lower gain.

The amp will most certainly benefit greatly from low frequencies deleted from the input.

Look, I'm not trying to build darth vader spaceship here. I'm looking for a simple solution to get a bit of a bump in sound. That's all.

 

RE: Oh dear ... let me rephrase, posted on July 5, 2014 at 20:16:59
Tre'
Industry Professional

Posts: 17302
Location: So. Cal.
Joined: February 9, 2002
That's very simple to do.

Just put a cap in series with the input of the tube amp.

A cap chosen against the input impedance of the tube amp (the grid resistor value of the first stage of the SET amp).

Here's the math. The cap value in uf = 159000 / (the frequency times the resistance)

If the input impedance of the amp is 100k ohms and you want to limit the amp's low end frequency response to 500Hz then a .0038uf cap placed between the input jack and the top of the grid resistor of the first stage will give you a 6db per octave high pass filter with a -3db point of 500Hz.


Tre'


Have Fun and Enjoy the Music
"Still Working the Problem"

 

Having been down this path, posted on July 5, 2014 at 21:27:51
Russ57
Audiophile

Posts: 3754
Location: South Florida
Joined: November 16, 2001
The problem isn't the signal your amp has to handle.....it is the signal your speaker (individual drivers) have to handle. Let that sink in real good and hard.

The simple solution you are after leaves you with a two outcomes. One is you have such a shallow slope that your amp and tweeter sees a lot of energy below this 500 hertz spot you pick (and that is a rather bad spot BTW). The other is, if your speaker crossover is such, that you leave your SET driving a complex impedance that changes with hertz (because they don't make a tweeter that can run 500-20,000 hertz without serious problems/crossover help). Neither is what you want.

Bottom line, you will walk away thinking bi-amping is flawed. That would be a shame. It is hard to get right but the rewards are there. You just have to think it through a lot more and get the right parts.

I'll suggest you look at what speakers (drivers) can do on the flat and level. We are talking about what range of hertz they can handle while keeping a ruler flat impedance. I doubt you will find much (in a single driver) that can handle more than about three octaves.

What you should be looking for, unless you want to tri-amp, is a mid-range driver that can run down to around 125-250 and up to around 1250-2000 hertz at 100db/watt (minimum) and mate that to a compression driver/horn that can run from there to 20,000 hertz (more or less). You would then have a passive crossover between mid-range and horn. Your SET would drive those. You'd be surprised what a couple/few class A watts can do.

You would use an electronic crossover after your preamp to send everything below 125-250 hertz to a solid state amp (say a 15" speaker on it) and everything above to the SET. If your musical tastes dictate it, a dedicated subwoofer will be needed.

You are better off using pro-audio drivers. Pick a crossover that allows you to adjust slope and cutover point. There are many things to take into account from room conditions to mechanical slopes of the drivers so you need adjustablity to begin with.

Don't think this will be easy. Chances are high you will decide amps were never your problem and that speakers were. There are some 12" and 10" vintage (alnico) speakers that will do and a few newer 8". DSS and 18sound made some good (cheap) horn lens.

 

RE: Oh dear ... let me rephrase, posted on July 5, 2014 at 21:54:08
Paul Joppa
Industry Professional

Posts: 7296
Location: Seattle, WA
Joined: April 23, 2001
You have not mentioned the speakers' passive crossover frequency or slope. That's important, your high-pass filter muse not disturb the phase of the tweeter signal in the range where it will affect the acoustic crossover performance. It is not difficult to imagine a situation where a simple first-order highpass two octaves below the speakers' crossover will still have an audible effect on the tonality. If it's a commercial speaker, you might have to get the attention of the actual designer to get a useful answer to these questions.

 

RE: Having been down this path, posted on July 6, 2014 at 06:00:30
Cleet Torres
Audiophile

Posts: 188
Joined: July 11, 2013
Did you actually read my post at all?





 

RE: Oh dear ... let me rephrase, posted on July 6, 2014 at 06:03:53
Cleet Torres
Audiophile

Posts: 188
Joined: July 11, 2013
Hi Tre

Yep, I was thinking of making a little switch-able setup with a couple of cap values allowing me to adjust the cutoff frequency.

I'm trying to see if there is an easy way to get a steeper slope. I am not worried about phase issues as I intend to keep the cutoff frequency way below the speaker crossover point which I believe to be 2500Hz. So if I have some phase shift anomalies at 250Hz - 300Hz I'm not particularly worried.

Thanks
Cleet.

 

RE: Oh dear ... let me rephrase, posted on July 6, 2014 at 06:17:11
Cleet Torres
Audiophile

Posts: 188
Joined: July 11, 2013
Hi Paul

I believe the crossover frequencies is around 2,500Hz and first order slope. I am thinking that for the signal I feed my SET if I keep the cutoff at below 400Hz I should keep all the phase issues I will introduce well away from being audible or bothering the tweeter.

The bass driver and signal I am not touching.

thanks
Cleet

 

RE: Looking for a high pass filter kit, posted on July 6, 2014 at 09:40:08
hifipaul
Audiophile

Posts: 735
Location: NY
Joined: December 22, 2008

You have a good idea, it's worth pursuing. One of my systems is set up in a similar fashion (Altec VOT bass-300hz xover-300hz horn) using passive and active xovers. Some suggestions: use a passive filter at the input of the tube amp, as mentioned above, or, better yet, rig it between amp stages. 1-2 octaves out of the pass band will be fine; skip the active xover, or use it only for the low pass section. The SS amp you use for bass will make a huge difference. My favorite amp for bass is a rotel, it sounds terrible on top but great down low.

All of this effort is great for LOUD music, but not great overall. I use a SET amp-full range/single driver for critical listening. When I use speakers with good, invisible, xovers like Dynaudios, I use a single, higher powered, tube amp because multiamping creates as many problems as it solves.

Should you decide to go this route, be prepared to do a lot experimenting and make a lot of changes.

 

RE: Oh dear ... let me rephrase, posted on July 6, 2014 at 15:11:40
Paul Joppa
Industry Professional

Posts: 7296
Location: Seattle, WA
Joined: April 23, 2001
That helps a bit. A first order crossover has only 90 degrees of phase difference between the highpass and lowpass signals, and (ignoring temporarily the response of the drive units) the tweeter signal is still substantial two octaves below the crossover (625Hz in this case).

Without writing a book on crossover design, I think you'll get the fastest and most reliable answers by experimenting. Put in the capacitor that Tre' described, but lay in a supply of several values. Use clipleads or a switch so you can compare the sound with the cap in and out, or between caps. Don't listen for sonic quality, just the frequency response/tonality.

It is not uncommon to reverse the tweeter phase, especially in first and third order crossovers, looking for the best sound. Add that to the experiment if you have thee time and patience.

You may well end up with something different from the designer's intent, but that works better in your room with your music etc... But if you do the experiment, you will know what compromises and choices you have made, and will have no need to worry whether it might have been done better.

 

LOL! [nt], posted on July 6, 2014 at 17:42:03
Ray Moth
Audiophile

Posts: 2784
Location: Jakarta, Indonesia
Joined: November 10, 2003

 

RE: Oh dear ... let me rephrase, posted on July 6, 2014 at 22:24:20
Paul Joppa
Industry Professional

Posts: 7296
Location: Seattle, WA
Joined: April 23, 2001
The post from Tre' to which I referred seems to have changed or disappeared, so my reference has no meaning anymore. I was talking of an RC filter and so was the post that has gone away. Sorry for the confusion.

For what it's worth, higher order filters have more phase shift.

I took a class in filter design in grad school; it barely scratched the surface of what matters in speaker crossovers. That's why I suggest experimenting, and I suggest keeping the experiments simple with few variables at first. Or, of course, asking the designer of the speakers you are considering.

 

RE: not so sure ..., posted on July 7, 2014 at 03:53:31
"and same wire and wire lengths also"
Poppycock old fellow. YOU will NOT be able to tell the difference between lets say 5 vs. 6 foot of the same cable ever. Not you, not me, not any single living human being.

In ARTA LIMP, there is a section for cable compensation that can be used when taking impedance vs. Hz plots.

The cable inductance in nH and resistance is subtracted from the measurements.

The length of wires does apparently make some difference here.

For determining Zobel networks, I would think it could be important.

As for Drlowmu hearing the difference in wires - he is using parallel runs of wires for everything in his amps and maybe in his speaker wires.

All that extra wire might be having some audible effect.

 

Harrison labs at parts express, posted on July 7, 2014 at 06:07:56
hawkmoon
Audiophile

Posts: 903
Location: cleveland
Joined: July 11, 2003
They plug directly into your amp to modify the output. You may want to try these before doing more work. I use them on my subwoofers. Good luck.

 

RE: not so sure ..., posted on July 7, 2014 at 06:47:22
Cleet Torres
Audiophile

Posts: 188
Joined: July 11, 2013
"All that extra wire might be having some audible effect."

Methinks an approach I should avoid then. ;-)

 

RE: Harrison labs at parts express, posted on July 7, 2014 at 06:52:39
Cleet Torres
Audiophile

Posts: 188
Joined: July 11, 2013
That's about as cheap as one can get and save the cost of another pair of IC's.

thanks
Cleet

 

This one? (see link), posted on July 7, 2014 at 07:51:50
Tre'
Industry Professional

Posts: 17302
Location: So. Cal.
Joined: February 9, 2002
It didn't go anywhere and I didn't edit it.

Tre'
Have Fun and Enjoy the Music
"Still Working the Problem"

 

RE: Harrison labs at parts express, posted on July 7, 2014 at 08:05:22
Tre'
Industry Professional

Posts: 17302
Location: So. Cal.
Joined: February 9, 2002
I see a problem with those.

Let's look at a high pass plug-in filter using the math from my post, http://www.audioasylum.com/forums/tubediy/messages/22/229184.html

I believe the plug-in is just a cap in series followed by a resistor in shunt.

If the plug-in doesn't take into account the input impedance of the amp then the crossover frequency will be wrong because the shunt resistor is in parallel with the input impedance of the amp and Harrison has no way of knowing what the input impedance of your amp is.

I'm not saying they won't work, they just won't work at the stated frequency.

Tre'
Have Fun and Enjoy the Music
"Still Working the Problem"

 

RE: Harrison labs at parts express, posted on July 7, 2014 at 08:09:12
Tre'
Industry Professional

Posts: 17302
Location: So. Cal.
Joined: February 9, 2002
You don't need another pair if IC's.

Just put the cap inside the amplifier.

Replace the piece of wire between the RCA jack and the grid of the input tube with the proper value cap.

That cap and the grid resistor create the high pass filter.

Tre'
Have Fun and Enjoy the Music
"Still Working the Problem"

 

More information, posted on July 7, 2014 at 08:31:06
Tre'
Industry Professional

Posts: 17302
Location: So. Cal.
Joined: February 9, 2002



If this is indeed a schematic for your amp the input impedance is 470k ohms.

A .00068uf cap, placed where shown, would cause a 6db per octave high pass filter with a -3db point of 500Hz.

Tre'
Have Fun and Enjoy the Music
"Still Working the Problem"

 

Pure speculation on my part, posted on July 7, 2014 at 09:00:24
1973shovel
Audiophile

Posts: 10117
Location: Greenville SC
Joined: February 25, 2007
But given their size (and perhaps cost) I'd bet they use a ceramic cap in series.

I'd much rather make my own. That would give me the ability to experiment, and to use a better cap than what's likely in there.

Thanks for your input on this topic, Tre'. I'm learning something.

 

You're welcome. (nt), posted on July 7, 2014 at 09:00:59
1973shovel
Audiophile

Posts: 10117
Location: Greenville SC
Joined: February 25, 2007


 

RE: More information, posted on July 7, 2014 at 09:36:06
Cleet Torres
Audiophile

Posts: 188
Joined: July 11, 2013
I do follow what you are saying. If I do go that route, I'll put it on a switch.

I prefer to have something obvious and in the eye so that I'm aware of the change whenever it's used. Years pass and you forget you modified a component and then it becomes a real head scratcher.

Regards
Cleet

 

A lot of Hogwash!, posted on July 7, 2014 at 10:54:24
gusser
Audiophile

Posts: 3649
Location: So. California
Joined: September 6, 2006
Yes, you will be adding another active stage comprised of one or more tubes. But consider why you are doing this in the first place. The benefits of sparing the SET from amplifying bass frequencies may more than outweigh any small losses from the active crossover.

As Paul said you do need to be careful with phase shifts from the combines filters.

Look for the Steve Bench crossover circuit. This is a simple one tube design. Also look at Marchand Electronics. They even publish manuals with schematics if you want to customize their designs. The Marchand crossovers are well regarded.

As for the matched amps, matched wire lengths, timing (other than phase shift), this is all a bunch of hogwash at audio frequencies.

 

Consistent and predictable , posted on July 7, 2014 at 14:18:25
drlowmu
Manufacturer

Posts: 9730
Location: East of Kansas City
Joined: January 10, 2005
You are. To get it wrong, totally.

Jeff Medwin

 

RE: This one? (see link), posted on July 7, 2014 at 14:27:28
Paul Joppa
Industry Professional

Posts: 7296
Location: Seattle, WA
Joined: April 23, 2001
That's teh one. It didn't show up yesterday for some reason. Thanks.

 

RE: Isn't the net result the same?, posted on July 7, 2014 at 14:37:57
Lew
Audiophile

Posts: 10911
Location: Bethesda, Maryland
Joined: December 11, 2000
"I like active low pass on the woofer amp and just a cap for the highs but that will depend on the frequency response curve of his tweeters and how that all lines up with his desired crossover point."

That's what I am doing with my Beveridge 2SW speakers. I took a cue from you and used a Russian silver mica on the Beveridge input. The woofers receive signal from the low pass side of a restored Dahlquist crossover via a vintage Threshold amplifier. At 100Hz, the crossover is seamless to my ears, or at least agreeable. I like that most of this system is at least 25 years old.

 

Very cool. (nt), posted on July 7, 2014 at 15:59:20
Tre'
Industry Professional

Posts: 17302
Location: So. Cal.
Joined: February 9, 2002
.
Have Fun and Enjoy the Music
"Still Working the Problem"

 

RE: Consistent and predictable , posted on July 13, 2014 at 06:30:36
Ray Moth
Audiophile

Posts: 2784
Location: Jakarta, Indonesia
Joined: November 10, 2003
No, Jeff, what Gusser says is plain common sense. (Remember, this is AUDIO frequency, not RF.)

 

Page processed in 0.039 seconds.