-
Get involved.
We want your input!
Apply for Membership and join the conversations about everything related to broadcasting.
After we receive your registration, a moderator will review it. After your registration is approved, you will be permitted to post.
If you use a disposable or false email address, your registration will be rejected.
After your membership is approved, please take a minute to tell us a little bit about yourself.
https://www.radiodiscussions.com/forums/introduce-yourself.1088/
Thanks in advance and have fun!
RadioDiscussions Administrators
Any product similar to Aphex Compellor?
- Thread starter Fenris2
- Start date
DBX makes a decent stereo compressor/limiter similar to the old Compeller. And it's cheap too:
www.sweetwater.com
266XS Dual Compressor/Gate - Sweetwater
Looking for Compressors / Limiters? Sweetwater has FREE Shipping and FREE Sweetwater Support for Compressors / Limiters!
www.sweetwater.com
Fenris2- Orban 6300 AGC can be set to be similar to single band with its cross-coupling controls, and it is likely to co-exist with Optimod at the transmitter. In fact, you might turn off the AGC in the transmitter Optimod and duplicate its function with the 6300 at the studio. In this 6300 application you would not use the multiband portion of the 6300. I believe by using the analog and digital outputs you could simultaneously use the multiband to feed something else at the studio.
www.orban.com
If you'd like to try something different, purchase an Angry Audio Chameleon C-Level. It is a mulitband, but you may find it can co-exist with an Optimod. If not, you can use it for something else. Ask them about another model of the C-Level that I believe is available.
Angry Audio does not offer demos on this product, but you may find a dealer who will.
angryaudio.com
If you are comfortable running a software processor, several software processors can be set up to approximate single band leveling.
OPTIMOD 6300 Audio Processor - Overview — Orban
www.orban.com
If you'd like to try something different, purchase an Angry Audio Chameleon C-Level. It is a mulitband, but you may find it can co-exist with an Optimod. If not, you can use it for something else. Ask them about another model of the C-Level that I believe is available.
Angry Audio does not offer demos on this product, but you may find a dealer who will.
Audio Chameleon - Angry Audio
angryaudio.com
If you are comfortable running a software processor, several software processors can be set up to approximate single band leveling.
Last edited:
DBX makes a decent stereo compressor/limiter similar to the old Compeller. And it's cheap too:
266XS Dual Compressor/Gate - Sweetwater
Looking for Compressors / Limiters? Sweetwater has FREE Shipping and FREE Sweetwater Support for Compressors / Limiters!
Kelly A
I respectfully say DBX is not similar to a Compeller in this application, because the DBX does not have platforming or gated recovery.
Last edited:
It would be cool if such a processor could be run off non-volatile memory on a Raspberry Pi! The interesting thing is that the hardware is so inexpensive that having separate devices for different purposes is feasible rather than depending on switching within a single device.
Clarifying and walking back my DBX comment a bit- The Compeller probably does not have platforming in the purist sense of the term, however the Process balance control can be adjusted to provide a similar sound.
The Compeller and Dominator were designed by Donn Werrbach who implemented his processing concepts in an original, inventive way. The Compellor prevented hundreds of radio stations from sounding much worse, because the Compellor could be adjusted to get away from a cliche compressor sound.
Given that Fenris2 is using an Optimod at the transmitter, the 6300 is a natural choice for leveling action at the studio, prior to an STL with inadequate dynamic range. Fenris2 could also locate a used Orban/Optimod product on eBay that could be modified or adjusted to use only the two-band leveler. Orban 464A would do this without modification, however any old piece of analog gear may need careful refurbishment to original performance.
Fenris2, consider that you may already be using something close to a wideband leveler if the front end AGC in a multiband Optimod is adjusted near that sound character. Thus, your task is to "insert the STL" into the correct location within the multiband Optimod signal path.
David- Stereo Tool offers a build for ARM / Raspberry Pi Linux.
The Compeller and Dominator were designed by Donn Werrbach who implemented his processing concepts in an original, inventive way. The Compellor prevented hundreds of radio stations from sounding much worse, because the Compellor could be adjusted to get away from a cliche compressor sound.
Given that Fenris2 is using an Optimod at the transmitter, the 6300 is a natural choice for leveling action at the studio, prior to an STL with inadequate dynamic range. Fenris2 could also locate a used Orban/Optimod product on eBay that could be modified or adjusted to use only the two-band leveler. Orban 464A would do this without modification, however any old piece of analog gear may need careful refurbishment to original performance.
Fenris2, consider that you may already be using something close to a wideband leveler if the front end AGC in a multiband Optimod is adjusted near that sound character. Thus, your task is to "insert the STL" into the correct location within the multiband Optimod signal path.
David- Stereo Tool offers a build for ARM / Raspberry Pi Linux.
Last edited:
If you're using it for no more than level control in front of an STL, the DBX will do just fine. Set attack and release medium- slow and have a nice day. For as much as some old school radio engineer's want to believe; none of these old forms of audio processing-thing is exactly rocket science, nor some sort of an art form.
Compeller's were decent boxes in their day, but just like most Aphex/Don Werrbach designs, had noisy VCA's.
Sorry Kelly, but that's nonsense. I've used the Compellor extensively in the broadcast chain and in the recording studio. With proper gain staging, the VCA's are not noisy, and it has some very sophisticated processing that cannot be replicated with a conventional compressor.
Thanks Greg, I forgot that the Optimod could be split into two components, one before the STL and one after. I need to re-read the manual, but IIRC it has a wideband AGC followed by multi-band processing. However, we've gotten the best results by using the Compellor into the full Optimod.
Unfortunately I'm dealing with an outside consultant who insists on modernizing the station. I'll have to let him do what he wants, and fix any problems as they come up.
Thanks Greg, I forgot that the Optimod could be split into two components, one before the STL and one after. I need to re-read the manual, but IIRC it has a wideband AGC followed by multi-band processing. However, we've gotten the best results by using the Compellor into the full Optimod.
Unfortunately I'm dealing with an outside consultant who insists on modernizing the station. I'll have to let him do what he wants, and fix any problems as they come up.
I agree. Back in the 80's I did syndication of formats to Latin America. Due to the cost of tape, shipping and customs, we distributed using BASF CrO2 cassete tape and Tascam 122 decks and the Compellor was needed to stay withing the tape's limited headroom. Only the Aphex unit could give the degree of level control without damaging quality that we needed.
Then here's a crazy idea. Why not check EBay for a Compellor?
aphex compellor for sale | eBay
Get the best deals for aphex compellor at eBay.com. We have a great online selection at the lowest prices with Fast & Free shipping on many items!
www.ebay.com
Modernizing with 90's vintage gear? This station must have been in pretty rough shape.
I recently did a recap on a Compellor 320. After a run through the alignment proceedures it beat all original specs. Orignal noise spec was stated "80dB below ceiling level, 20-20KHz, unweighted". This one did 89dB, and sold on eBay for $2.25/dB. ($200).
If the station is analog, a Compellor won't be the thing that establishes the noise floor. But I'll vote for the C-Level or the 6300 if you have to have something a bit fresher, and hardware.
If the station is analog, a Compellor won't be the thing that establishes the noise floor. But I'll vote for the C-Level or the 6300 if you have to have something a bit fresher, and hardware.
If this is an AM station, you're right that additional program chain noise probably wouldn't be of any consequence. If an FM, I would submit that in cascading older designed/aging VCA/analog opamp boxes should be avoided if possible, simply because signal to noise though the chain is cumulative.
I preferred the CRL SGC-800 over the Compellor. It gives you a choice of dual-band, wide band, or a combination of both, plus Dynafex noise reduction which was useful back in the era of hissy tapes.
Technically correct, practically...nope. There are zero FM receivers that actually get to full quieting of -80dB in stereo, in the field. Those are lab-only numbers. Receiver noise is always way above that, like by 15 or 20dB. That, combined with listener's ambient noise, and the noise in any older VCA box is far, far below being a problem. Add to that the fact that a music format typically doesn't pause audio long enough for the entire processing system to recover, especially if at least some of it has a recovery gate, much less leave the system noise floor standing naked in the open with no other mod. A talk format has more acoustic noise at the microphone than any electronic noise in the chain. The one possible problem format would be classical music, and you likely wouldn't use any pre-processing on that anyway, because it's the one format where listeners actually complain about processing.
A compellor was good enough for FM when it was new, and if it's in spec now, it's still good enough, at least from a noise perspective. FM's actual audio performance at the receiver hasn't gotten better in 30 years. If anything, it's worse now. The only improvements in noise floor in the chain would be digital STL, processing and digital exciters, but none of that solves the reception problem.
That is definitely a problem during EAS tests. Every time the announcement comes on, there is such a huge rush-up of noise that it would be more accurate to describe it as "noise-to-signal ratio".
If that happens, the EAS unit is in the wrong place in the chain, or there's a level set problem. Or you're hearing the forwarded audio, which everyone knows is junk. That doesn't indicate a noisy processor, it's a flaw in the EAS system, hopefully soon corrected.
Several years ago I did an experiment on one of my FM stations. Installed a digital STL with AES in and out, got rid of all compellor's, Orban 424's, and anything analog or VCA in the line. Installed an Orban 8200 AES-in straight out of the audio console, with AES feeding the STL. STL balanced AES out straight into a (new at the time) BE FXI-60 exciter. Tuned the BE FM35B transmitter for the lowest synchronous and asynchronous AM noise via a new Belar FMSA-1 modulation monitor, close to -78db. Off that same modulation monitor, I measured total audio signal to noise OTA out the antenna at -88db with all but the mic fader turned up.
Prior to all that work, the station with a hybrid AES/analog program path was lucky to hit -60db signal to noise.
The end result was nothing short of stunning. The stereo separation was amazingly superior to any other station in the market. Quiet passages, even with prerecorded music were noticeably quieter, and that's even prior to the processing hitting the gate. Dynamic range seemed much more pronounced. Even listeners called and E-mailed unsolicited, asking if the station had done something, because it appeared the reception had improved. Comments like 'much clearer', 'sounds more like the actual recording than other stations' were common.
So in closing. I couldn't disagree more that noise isn't destructive to even modern transmission or reception of FM broadcasts. Lower noise transmitted means superior quality at reception. To your point, does that mean practically the station needs to spend the time and money to install superior low noise gear, and then maintain it? Absolutely! Anything else should be considered just 'good enough', but hardly as good as it could be.
You make my case.
You also make another important point: Throwing more boxes at noise and/or distortion problems makes them worse, not better. The KISS principle applies in spades to processing.
Interesting anecdote and "experiment", though I'm not sure what the experiment was. However, if the old system prior to all that work only barely hit -60dB, something was seriously broken or misaligned. You did a whole lot there, and now point fingers at the old analog stuff as the culpret. I would suggest that a slow, step-by-step upgrade path might have revealed the defective bit. You didn't say the old STL was analog, but if it was, that right there would be the most likely of all to raise noise and, if composite, mess with separation. Wide-banding the FM35B for low sych AM was a good move too, but would affect noise only in some rather specific multipath conditions. I don't recall that it was even possible to narrow-band an FM35B to the point that separation was impacted audibly, but I think not. So wide-banding didn't change much other than to minimize multipath artifacts.
To scale the anecdote just a little, though...stereo separation. Yeah, well, that's actually not something you can hear improving once you're at 35dB or so. First, no program material panned much wider than that, second, psychoacoustics...you just can't pick that up except under very controlled test-only conditions. Second, psychoacoustics come into play. It is in fact possible to reduce measured separation while increasing percieved separation. So who knows. The clue might be:
You did a major processing change, right? All of those new settings most certainly could do that. Wider dynamics will sound better separated too, and lower distortion, and perhaps lower noise. But a lowering of noise and a change in separation wouldn't do any of that directly.
As to "coudn't disagree more"...I never said noise isn't destructive, so you're not disagreeing there, but you are on the edge of misquoting me. And certainly lower noise is always a desirable goal. But "lower transmitted noise means superior quality at reception" is a huge generalization that I can both agree and disagree with once we start putting numbers on it. So let's try that.
First, put your listener in his environment. It's probably fair to assume he's not cranking up the radio to concert levels. The typical listening level for radio has a couple of thresholds. One is, it's desirable to be at the same level as average speech, slightly below or above, circumstances depending. Average speech hits us at 65dB SPL, so lets say 70dB for loud speech. Peak to average ratios after processing are not much more than 8dB. That puts 100 % peak mod at 78dB SPL and 70dB SPL average, and a -60dB (re 100% peak) noise level at 18dB SPL, unweighted, and weigthing will take it down 4-6dB more or so. Ignoring weighting, we're at 18dB SPL with our horrible -60dB noise, so what's ambient noise level like in our listener's room? If he's at home, he might hit a 25dB SPL noise floor with his HVAC not running. If he's in an office, it's going to be 50dB. If he's in a car, it's 70dB. That's his noise floor, not the transmitted -60dB. Now we clamp active noise cancelling headphones on him, take the average SPL up to 85, and block a lot of background noise. Now perhaps he can start to hear our -60dB, but everything has to pretty much stop to do it. Does that represent typical listeners?
These days most listeners are in cars. Cars ambient noise is 70dB on average. If the radio is turned up so it comes in at 85dB (and that's so loud you can't talk over it!), our -60 noise is now 45dB below the listener's ambient noise.
Circling back around, in my world, -60dB for transmitted noise would have been unacceptable for FM at any point for the last 50 years for the signal going to the antenna. That wouldn't even be "good enough". Something was broken, and it wasn't just because it was old analog. You replaced the broken pieces with new stuff all at once. It could have been new analog gear too, and would have worked almost as well. And here's the "agree" part. I'll be the first advocate for lowest possible transmitted noise. I worked very hard to get it to the mid 80dB area 40 years ago when exciters were analog and we used phone lines for STL. And it was possible to do. But never, and I do mean never, could that transmitted noise level be realized in a receiver in the field. In fact, back then, the current mod monitors were noisier. -86dB was a nice feel-good number, but it wasn't audible on anybody's radio. 55dB of separation was also a nice feel-good number, but never receivable either once you've run the RF through an IF strip and an antenna in even a little multipath. You can't hear that separation because it isn't there to beging with, and the music you're listening to doesn't have it baked in either. I'm not saying we shouldn't bother transmitting the best possible signal, not at all. We should all obscess about getting the last dB of S/N. But it is a $$ vs dB issue too, and many times the $$ win. I'm just scaling the expected result to reality.
And in reality, there are plenty of Compellors that won't damage today's audio perceptibly with noise, because they never did. Analog gear isn't just all noisy because it's analog either. And yes I do advocate an all-digital transmission system, but let's be real. Some stations can't afford all of that. That doesn't mean they can't be competitive with an all analog system, it just means it takes more effort and understanding to pull it off.
Are you saying that cascading analog VCA/Opamp devices don't contribute to overall audio path noise?
Then I maintain they shouldn't kid themselves that they're doing everything to improve their broadcasts. But, that's largely part and parcel for radio anyway. It's typically done on the cheap.
Competitive with similar radio stations? Or competitive quality as compared with the new world of competition from streaming?
- Status