DBX 263A De-esser - Any opinions or tips on performance?

[K1JJ]

Hola,

I just bought a NOS  DBX 263A de-esser. A friend recommended it.  Will have it tmw.  
The purpose is to attenuate some of the sharp SSS's I have in my voice / teeth and also get rid of highs above 6 Khz. I'm hoping to get rid of my digital audio filter as a result.

It has line in and out on the back.  Where in the chain would you put it?    Mic > 528E preamp > EQ > 6 band processor...  My 528E has a de-esser, but hoping for a better box in the dedicated DBX 263A unit.

Has anyone used one and any tips or opinions on its use?   Being designed back in 1987 era, this is an analog box as far as I can tell by the schematic, right?

Thanks.

T

[N2DTS]

The VX2000 has a de-esser with an adjustable frequency and threshold but it does not seem to be very effective,stuff still gets through.
It seems to be made to reduce it, not block it.
Since I can not hear highs, i can not tell what it sounds like, but I see stuff getting through on the spectrum display.

[Steve - K4HX]

Just eat marshmallows when you are on the air. No dbx thingie needed.

[W2VW]

Get some tips from the ssb guys on 3872 every Friday afternoon. It's the bad false teeth net.

None of them can make s or t sounds.

[KL7OF]

Tom....I would try the de esser between the EQ and the processor....Good Luck...Steve

[K1JJ]

Tom....I would try the de esser between the EQ and the processor....Good Luck...Steve

Tough crowd, indeed...  

I could just talk with my arm in front of my mouth.


Steve, OK on between the EQ and processor.   I'll try that and as the last box in the chain, just to see which is best.


Or, I could just get rid of it all and plug in a Golden Eagle D-104, caw mawn.

Hoping to hook it up tonight.

T

[K1JJ]

* Report Update:

I gots the DBX-263 De-esser tonight and hooked it up.  

It's just what I had hoped.  I was able to get rid of my digital filter and let the analog De-esser do the job of sloping off the extreme highs.

With the digital filter, the slope was so sharp I think I was missing subtle voice harmonics  - anything above 5 Khz was gone.  But the de-esser has a softer roll off and lets the 6 to 7 Khz come through, though down 10 - 30 Db.   If the rig is clean, I don't think these high frequency wisps will be a problem.  It sounds better to my ear.

The de-esser comes in aggressively on the highs above 5-6 KHz and uses a mild compresser. I see no attenuation or alteration whatsoever of the frequencies below 5 Khz.   I now hear a smooth "ss" sound rather than a harsh tearing.  I could not get the same effect using my 6-band 9024 processor.

I think it's a good addition and allows me to run the important frequencies below 5-6 Khz more aggressively.  The audio spectrum now looks busier.

T

[flintstone mop]

YO Tom
The Behringer 9020 will lop the hell out of the high end very nicely.
I have mine set for -70dB compressor threshold and the level set for minimum. You can Un-link the bands and it's like 6 separate compressors. From 5kc up 20kc the level is greatly reduced.
I kept my DBX, 32-band EQ in-line to cut the highs a little bit more. Everything is flat up to 6.3kc and I attenuate rapidly to 20kc.
This has been wonderful for the bright over-processed audio at WBCQ. If they change something, I can do the opposite to some extent.
If the battle persists, I quit!!! I'm getting tired of fighting technology and non-communicating people.
Sorry for ranting
Fred

[K5UJ]

I think it's a good addition and allows me to run the important frequencies below 5-6 Khz more aggressively.  The audio spectrum now looks busier.

what are you running for a modulation peak limiter?

[flintstone mop]

* Report Update:

I gots the DBX-263 De-esser tonight and hooked it up.  

It's just what I had hoped.  I was able to get rid of my digital filter and let the analog De-esser do the job of sloping off the extreme highs.

With the digital filter, the slope was so sharp I think I was missing subtle voice harmonics  - anything above 5 Khz was gone.  But the de-esser has a softer roll off and lets the 6 to 7 Khz come through, though down 10 - 30 Db.   If the rig is clean, I don't think these high frequency wisps will be a problem.  It sounds better to my ear.

The de-esser comes in aggressively on the highs above 5-6 KHz and uses a mild compresser. I see no attenuation or alteration whatsoever of the frequencies below 5 Khz.   I now hear a smooth "ss" sound rather than a harsh tearing.  I could not get the same effect using my 6-band 9024 processor.

I think it's a good addition and allows me to run the important frequencies below 5-6 Khz more aggressively.  The audio spectrum now looks busier.

T

harsh tearing.......
Are you monitoring the audio transmitted from your antenna or dummy load?
The tearing I hear from my Flex comes from the R390A in the 8kc BW. When I flip it to 16kc BW.,sounds like FM radio.
Recording the Flex TX audio from the software is always clean ss's and nice audio.
Surprised you can't get same effect from the 902........hmmm

Fred

[K1JJ]

Without extreme highs limiting I'm getting full modulation out to 7 Khz, with still some at 10 Khz. Maybe it's not as bad as tearing, but certainly is excessive. I'm listening in an AM diode detector and watching the SDR and scope. I'm generally keeping mod % below 100% positive.

The 9024 does have a peak limiter in it which I need to play around with more.  

I do have the 9024  5Khz to 10 Khz band set at -70 for full compression.   I'm finding a touch of the 9024, a little high cut from the EQ, and a hard limiting from the de-esser is all I need to get the desired slope.  Must investigate unlinking, assuming it's not already unlinked.

I use a dummy load 99% of the time, but notice the audio monitor is a little hotter into the antenna.

I'll try a recording as you suggested, Fred.

T

[K5UJ]

<<The 9024 does have a peak limiter in it which I need to play around with more.  >>

Okay, the Beheringer compressor is not an AM broadcast peak limiter.  It may have some kind of limiter function in it but it is a studio piece of equipment.  If you are using it to try to limit mod peaks then that's part of the problem.  Only a broadcast limiter has the instantaneous attack that eliminates virtually all overshoot. 

<<
I think it's a good addition and allows me to run the important frequencies below 5-6 Khz more aggressively.  The audio spectrum now looks busier.>>

To achieve the dense sound you want in the mid-ranges you need to use the 5 band compressor (adjusted accordingly) to drive a broadcast mod. limiter into 6 to 10 dB gain reduction and roll the highs off with an equalizer or some sort of 5 kc LPF instead of a de-esser except when the band isn't crowded--then the de-es can be used.  They usually only act on sibilance highs above around 6 kc.  This is what will allow you to achieve a high average peak power level in your sidebands.  The compressor levels but it doesn't prevent carrier clipping so you can't safely get the high averages you are probably looking for.  You need a transmitter with mod.iron and audio power reserve to handle it.   Aren't you running something like a 4x1 modulated by a pair?  that should do it     A mistake some ops make is getting a LPF that sharply cuts off at some frequency.    a gentle attenuation is needed--I have a fairly linear drop off from 3.5 kc to 5 kc of about 30 or 40 dB using a graphic eq. for this.

[W2NBC]

The DBX-263 and a few other stand alone de-essers have variable crossover points with aggressive attack times and variable release .. The crossover for Tom's unit is anywhere from 800Hz to 8kHz .. This particular DBX model is still used at CBS Television, News Radio 880 NYC, all for Broadcast and highly effective ON THE AIR.

From what I understand, the transmitter Tom is using with this setup is his Class E, with a tremendous amount of "headroom" (170% peak positive capable, with built in negative peak limiting). He is running now at only 100% positive peaks with about 80% neg. Any "broadcast limiting (clippers)" would be useless within those operating conditions.. BTW, "studio" positive peak limiters are exactly that.. Infinite compression or 'limiting".. slower than clipping, and hopefully add minimal distortion.. With Tom's set up "overshoot" could only be defined as his preference for high frequency content control as it pertains to bandwidth, and the DBX is certainly up to the job.

[K1JJ]

The DBX-263 and a few other stand alone de-essers have variable crossover points with aggressive attack times and variable release .. The crossover for Tom's unit is anywhere from 800Hz to 8kHz .. This particular DBX model is still used at CBS Television, News Radio 880 NYC, all for Broadcast and highly effective ON THE AIR.

From what I understand, the transmitter Tom is using with this setup is his Class E, with a tremendous amount of "headroom" (170% peak positive capable, with built in negative peak limiting). He is running now at only 100% positive peaks with about 80% neg. Any "broadcast limiting (clippers)" would be useless within those operating conditions.. BTW, "studio" positive peak limiters are exactly that.. Infinite compression or 'limiting".. slower than clipping, and hopefully add minimal distortion.. With Tom's set up "overshoot" could only be defined as his preference for high frequency content control as it pertains to bandwidth, and the DBX is certainly up to the job.

Thanks for the explanation, Jeff.

Yep, this is the class E 24 pill PDM rig.  I've found the THD sweet spot is around 80% negative, so I run it around there most of the time. I can always turn the audio up if need be.

You mentioned the release is variable? This is an internal function, yes? With the DBX-263 I notice the release time is maybe 100 mS or so. The point is when the sibilance hits the DBX-263's compressor hard, the rest of the lower frequencies are pulled back too by the same dB compression, being a single band compressor. Is there a way to deal with this?  I find the DBX works the best, at least for my use, at 15 Db of gain reduction, which is 3/4 of the slider range.

Still, I am very pleased with it.

After testing it, the 9024's limiter does appear to be an average based circuit rather than peak.


BTW, I'm a little cornfused on whether to run rthe DBX in broadband or high frequency mode.  The broadband works best with the frequency knob at about 4 Khz. But in Hi-Freq mode it does similar performance when the freq knob is at 2.5 Khz.  Any ideas here?   The manual seems to say voice is best with Hi-Freq mode and broadband for mixing, but try them both.


T

[W2NBC]

Tom ,
Some of the units at CBS have been modified for the variable release, with an access hole (like the one I saw here).. Just spoke with the audio maintenance supervisor who said that is not necessary for nominal use.. For the price, they are still pretty cool and without the digital artifacts!

As Rob mentioned, gradual EQ curves help retain the "sparkle" and harmonic content without sharp cutoff artifacts with LPF's (especially digital ones)..

The unit should be set to "high frequency mode".. to compress only the stuff above the crossover.. I wouldn't use the nomenclature on the knob as accurate.. I don't own one, but I will bring one home on loan and dig into it OM!

that'sssssssssss it for now.. oops.. forgot my de-esser..   "that's it for now".
Jeff

[K1JJ]

Tom ,
The unit should be set to "high frequency mode".. to compress only the stuff above the crossover.. I wouldn't use the nomenclature on the knob as accurate.. I don't own one, but I will bring one home on loan and dig into it OM!

Jeff

Glad I axed.  In the Hi-Freq mode I don't hear that compressor action effecting the lower freqs.  It's working only on the higher stuff, as it should.  That's a good thang. 

Tnx, OM.

T

[flintstone mop]

Hey Tom
The 9020 un-linking is in the setup screen. Just tap on setup and upper right of the display is band-link...on off.
When un-linked each band has it's own separate setting. When they are linked, whatever you adjust for band # 1 affects, bands 2, 3, 4, 5, & 6.

It would be nice to hear this monster TX and your audio.
give us some air dates and time and freq.
Fred

[K5UJ]

This isn't worth an argument so I'll make a couple of points and leave it at that:

<<Any "broadcast limiting (clippers)" would be useless within those operating conditions.. >>

I disagree.  Firstly, there are clippers and there are limiters.  It may be a matter of terminology, but to me, a clipper, is an older technology device (or a homebrew diode clipper) that has an extremely sharp knee and does just what the name implies.   When I speak of limiters, I'm thinking of more modern and sophisticated tools that employ look ahead or feed forward techniques to start acting on a fast rising level and start to insert a curbing attenuation so even a fast rising pulse is flattened out ahead of the rig.   Once set up, this among other things allows the op to relax and not have to keep an eye on the scope knowing no matter what sounds hit the mic, the carrier is not going to clip.   Anything else not designed for that particular goal is simply not going to do that.  No matter what up or down mod. percentages you operate at, you are not going to be able to drive anything going into the rig at a consistently high level with a reliably solid ceiling without something made for that purpose. 

Secondly, if you have never operated with an AM broadcast peak limiter, why not keep a look out for one and pick up an old CRL PMC400, or 450, or inovonics 222 and experiment with it.   If you wind up deciding you don't need it you'll always be able to sell it.

[W2NBC]

 " tools that employ look ahead or feed forward techniques".. many $$$$$ and with latency.. How do I know? I have an Orban 9200..

This topic was about a de-esser  and its application..

Ps. none of the your second example devices  CRL PMC400, 450, inovonics 222 are "look ahead limiters, and what makes them "Broadcast" limiters are adherence to NRSC pre-emphasis/de-emphasis curves (response) and asymmetrical limiting. If you play them hard you can increase "density", but at increased distortion PERIOD..

  I have heard Tom's audio and it is LOUD and distortion free..
  We will just leave it at that..and is DEFINITELY not worth an argument

[Ralph W3GL]

Jeff,

What  essentially is the difference between the DBX 263A and the
DBX 163X

I have  an 163X that I've never tried to put on line along with a DSP
9024, also not on line.   

What I don't have is a booming "radio voice" like some folks on here
but with my American DR332 mike driving my VX-2000, my reports
are "sounds like you" from people who know me in person...

Now the big question, will my DBX or DSP increase the density of
my transmitted audio?   

[W2NBC]

Good Luck !! Hey Ralph,

The 263A that Tom has is a "de-esser" which is basically a fast acting compressor that works to reduce sibilance or higher frequencies associated with the "S" sounds.. It is mainly used as a recording tool for vocals with bad fitting dentures.. (just kidding).. It is effective at what it does but doesn't add squat to 'density" !

The 163X that you have is a single band (wideband) compressor that works across the entire audio spectrum and is useful in keeping "average" levels higher which (can) increase perceived "density or loudness"..
The 9024 that you have is a multiband compressor limiter that has more control over individual crossover points and can give you even more "density" (loudness) by compressing each range of frequencies at thresholds you set tailored to your voice.. BUT.. with even all this junk to add density , I have heard your signal and it too is loud and clear and in my opinion, may not be worth throwing any of it inline!!

PS. your VX2000 has compression and de-essing built in! you're all set..

You sound great sir and nice to hear from you!!

Jeff

[W2VW]

I really do not like the way the VX-2000 compressor works. It can't keep up.

Ralph sounds almost exactly the way he does in person except for an occasional kazoo peak sound. That makes me sad.

[N2DTS]

Not sure I like it so much, but its small, cheap, and it does a lot.
Mic preamp with phantom power, phaze switch, line input, expander (that part works ok), poor compressor, basic EQ, de esser.

I think you can not push it much because its not fast acting enough.

I really do not like the way the VX-2000 compressor works. It can't keep up.

Ralph sounds almost exactly the way he does in person except for an occasional kazoo peak sound. That makes me sad.

[W2VW]

Not sure I like it so much, but its small, cheap, and it does a lot.
Mic preamp with phantom power, phaze switch, line input, expander (that part works ok), poor compressor, basic EQ, de esser.

I think you can not push it much because its not fast acting enough.

I really do not like the way the VX-2000 compressor works. It can't keep up.

Ralph sounds almost exactly the way he does in person except for an occasional kazoo peak sound. That makes me sad.

Yes Brett the features price and size are all nice. Having a mic preamp is nice too. I just wish the compressor didn't have that transient response to fast rising sound.

I've heard you on the air recently and do not think this is an issue on your audio.

[N2DTS]

Well, I should try and actually set the thing up right.
I have not really played with all the levels and settings, I could try fast and hard limiting then run just below the threshold, as it is, I run soft compression with the fast button out just to keep the audio level up.

For EQ, I get reports I have no highs, other say I have no lows, so I guess its about right...