the 1954 Limiter!!

GroupDIY Audio Forum

Help Support GroupDIY Audio Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.
The input section looks very similar to BA6A the RC network look the same .0006 sec attack 0.33 discharge release for RCA.. could the 00006 extra 0 be a typeo... theres also a audio engineering 1950 artilcle here somewhere about the new BA6a with an explanation of the single and daul attack & release of the Ba6 it sounded very much like the Radio craft explanation...could they have been reworking the BA6A circuit for 1954 & as the RC network was similar thought attack release times would be the same so re wrte the audio eng piece... Im shootin in the dark a bit..if I understand correct the 1954 would be slower attack due to less power to drive rectifier ??....
id be very interested to play with this circuit

bottom line is tho 1954 sounds good as well as looks great.
 
[quote author="PRR"]This dual-stage plan was floating around, but I suspect someone saw a Fairchild. Which does have 50uS attack, but from a much beefier sidechain than one 12AT7. [/quote]

This got me thinking; Since the 12AT7 doesn't have the poop to drive the 6AL5 as we'd like, and we might not want to add a beefy sidechain or output amp, like 6V6 pair in Collins, et al., why not use a 6SQ7 in the sidechain as in Federal Limiter? It looks like the 6SK7 plate drives the 6SQ7 grid just fine, or is that only possible with the 12k plate resistors, as opposed to 47k. Would we load the 6SK7 plate too much?

If that would work you'd shed 2 tubes and have a 3 tube limiter. Then it could fit in less than 7RU box, too. :green: Of course, I think cannikin's look positively smashing and magisterial in their big boxes. :thumb:

My other immediate thought was that a lesser stepdown in the output transformer could yield higher levels into common higher Z inputs. Even though we like our tube gear to drive lowZ input passive EQs, a healthy level into modern higher Z inputs would be plenty useful. Maybe an interstage transformer for plate to P.P. grids used backwards as ~100k:10k?

Just sittin' here thinkin'.
 
> why not use a 6SQ7 in the sidechain

Makes things worse.

6SQ7 (half a 12AX7 only lamer) has higher output impedance than 12AT7. Actually 5 times higher.

That means C1 will charge slower.

> as in Federal Limiter?

The Federal is also slow. Even slower than the 1954. It was never made to be great audio.

That's not bad. I was taught to avoid clipping more than a few mS. Some systems are hyper-sensitive and need 50uS. But clipping is widely used for "sound". You can't put drumset in a mix without violent peak reduction, which pretty much means clipping, and often quite a lot of it. Fortunately a ADC accepts overdrive without telling you about the lost bits, no rail-bounce/stick such as some analog channels do.

If you want "FAST" audio limiting, with the usual hold cap values, you need under 1K source impedance. And since GR voltage needs to be larger than peak signal voltage on the vari-tube grid, you want, 5, 10, 20, even 40 Volts behind that 1K. You want a 0.1 Watt to 1 Watt Power Amplifier.

Several of the Classics used the existing 0.1 Watt line output stage to also drive the rectifier. This makes some control difficult; also tends to load-down the peaks. Both flaws are very acceptable, either in safety limiting or limiting for effect.

Sand-state devices dump current tons better than tubes: faster attack. Gain is cheap too: flatter limiting. If you wanna "make it better", easy: go sand. Heck, put a DSP in the sidechain and compute any dynamic response you like. (Also program it to self-test and compensate tube unbalance.)

lesser stepdown in the output transformer

To a point. 10K instead of 600 would give 4 times the level, except many of our inputs really are 10K so we could lose half of that, get only double the level.

The real question is: what is the output power? To get 20dB compression (surely more than TV commercials merit) from the operating point they seem to have selected, 6SK7 current has to fall to about 0.2mA. Depending on the load, that's not much over a milliWatt, well short of traditional Line Levels which reference 1mW and go up from there. No doubt this beast can jam a ADC to max, but a lot of engineers are not happy unless levels are HOT, even though they end up padding-down.

Getting a limiter to just "work good", control audio with minimal damage, is harder than about any other "basic" audio design. Doing it well in tubes is just awful hard. Fortunately most modern music needs "color" more than accuracy.
 
Looks amazing - great!

But the schemo has gone from page one. Any chance you could post it back up, please, or email me a copy?


EDIT - Just found it here
http://www.groupdiy.com/index.php?topic=27372

:thumb:
 
Hi all! Recently, I've started breadboarding the 1954 Limiter with a 6SN7 in push-pull as make-up amp. So far, the audio amp is completed and it passed audio without much problem. I did some quick voltage measurements on the unit and was quite surprised that the DC voltage on the 6SK7 cathode is 83V!! ??? Could this high voltage be due to the use of a high value resistor in the cathode circuit (it looks like 24K in the schematic)? Am I way off from the original value?

Many thanks to cannikin for posting the schematic. This is a great forum and I have learnt a lot of things here. Thank you guys!!!  :)

Cheers!
Gavin
 
Looking back at this, it's pretty much an update of an AGC that NBC used around 1940, and the adjustable bits looking like BA-6 or 86 have hallmarks of RCA. 

index.php



 
Feel bad resurrecting this but doing it anyway cause I'm genuinely interested and I've been told multiple times to ignore that red notice up there...

So what exactly was the problem with this thing again? It doesn't put out enough juice while significantly reducing gain?

I was thinking about building this.... mostly because it seems extraordinarily easy compared to a lot of the tube limiter circuits and I have all the parts except the voltage regulator tube and the transformers. But if it can really manage only half a volt while reducing gain by 20 db... maybe it isn't worth it??

Anyone have any ideas on modifying it to account for that in a relatively simple/elegant way?

Is it also really necessary to build in the 6.3 VAC part that allows you to bias the screen grid via phase cancellation?? Just seems a bit weird to build a biasing tool into an interface....

Also, anyone have any idea to what degree the pot near the voltage regulator tube will actually allow you to control the limit threshold?
 
davemascera said:
Feel bad resurrecting this but doing it anyway cause I'm genuinely interested and I've been told multiple times to ignore that red notice up there...

GOOD ON YOU!  No reason for a new thread. 

davemascera said:
So what exactly was the problem with this thing again? It doesn't put out enough juice while significantly reducing gain?

PRR covered it on page one.  It's basically the first stage of a limiter, so it needs additional make-up gain to be useful at line levels. 

davemascera said:
Anyone have any ideas on modifying it to account for that in a relatively simple/elegant way?

put a 990 SS amp after it.

davemascera said:
Is it also really necessary to build in the 6.3 VAC part that allows you to bias the screen grid via phase cancellation?? Just seems a bit weird to build a biasing tool into an interface....

Tubes change parameters over time and many of these controls on vintage limiters get tweaked frequently. 

davemascera said:
Also, anyone have any idea to what degree the pot near the voltage regulator tube will actually allow you to control the limit threshold?

That's what it does, it's limited to a low compression range.  Make that R string a 25K for higher ratio/threshold. 
 
EmRR said:
GOOD ON YOU!  No reason for a new thread. 

PRR covered it on page one.  It's basically the first stage of a limiter, so it needs additional make-up gain to be useful at line levels. 

put a 990 SS amp after it.

Tubes change parameters over time and many of these controls on vintage limiters get tweaked frequently. 

That's what it does, it's limited to a low compression range.  Make that R string a 25K for higher ratio/threshold.

Hmm I guess it makes sense if they're to be set that often....

The 990 is not a bad idea at all, but since I do have a 6SN7 right in front of me, is there any reason I can't just plop the output stage of the federal right in there?
 
davemascera said:
Hmm I guess it makes sense if they're to be set that often....

The 990 is not a bad idea at all, but since I do have a 6SN7 right in front of me, is there any reason I can't just plop the output stage of the federal right in there?

lassoharp did a similar build......

https://groupdiy.com/index.php?topic=44011.msg549861#msg549861


BTW, if you don't mind me asking. what substitute transformers did you use for your BA-2 build?

 
scott2000 said:
lassoharp did a similar build......

https://groupdiy.com/index.php?topic=44011.msg549861#msg549861


BTW, if you don't mind me asking. what substitute transformers did you use for your BA-2 build?

CMQEE-3440A and CM-27101

I was having problems on the low end (-10 at 20 Hz)... talked with Dave at Cinemag and it was that the CM-27101 wasn't beefy enough.... I loaded it with a 600 ohm resistor on the secondary and then I was only about -3 at 20 Hz, nearly flat at 30. Left it that way and use it almost every day. Bought the originals from Doug to add in a second preamp and ended up reselling them.

Hmm interesting. So he's basically just using the output stage from the federal combined with a lot of elements from the BA6A...
 
Hmm so I'm doing this build now... gonna try to put a 6SN7 in and attempt to implement the cathode return meter... might try to use an interstage transformer like an A-18 for sh*ts instead of some coupling caps (course I know the 12at7 will still need-em).. not sure..

Does anyone know if there is any downside at all really to increasing the value of C1 if one wants a slower attack?? What about making R1 a pot to ground in case someone wants to make the release faster? So far as I can tell, I don't really see the harm in it... seems like

Was thinking of making a selector switch with a bunch of increasing cap values on it, maybe maxing out at ~30ms....
 
Does anyone know if there is any downside at all really to increasing the value of C1 if one wants a slower attack?? What about making R1 a pot to ground in case someone wants to make the release faster? So far as I can tell, I don't really see the harm in it...
Most of the old compressors were designed for radio transmission, so the timings were fixed  (they were playing mastered recordings after all).  For recording purposes, you need a selection of attack and release times.  The normal way to do this is to keep the cap value constant but switch in various series resistors to feed it for attack, then a similar switch in parallel for release.  The advantage of switches is that you can make a record of the settings for the track, not really possibly with just a pot.

0.1uF can allow a slight ripple which on some feedback compressors can turn into motorboating,  1uF is used on the Fairchild and others, but many other comps get by with 0.25 or 0.5uF.  Resistors that add with each position are mainly used to avoid switching noises, changing caps will probably cause some  sudden charging noise.

DaveP
 

Latest posts

Back
Top