LOFT 410 expander compressor limiter

I thought I would start a general thread for informational purposes.

here is an interesting piece. LOFT 410

It appears to be from mid 80's based on some '1985' I.C. date codes.

It is a pleasure to use,  it is accurate enough,  and it sounds nice to my ears.  (It will be even better after servicing)

The  IC's are socketed for easy removal/replacement, and it contains a pair of Valley People TA-101 transistor arrays

EDIT:  it has a LOFT serial number 72 and PHOENIX serial number 04832

EDIT sept.1 :  after recapping,  installing new regulators, and cleaning pots and contacts, it is now in my recording chain and sounding great, much detail has been restored..

pic 2

component side of PCB

JR might have designed this.  I'm sure he'll chime in. 

envelope said:

I thought I would start a general thread for informational purposes.

here is an interesting piece. LOFT 410

It appears to be from mid 80's based on some '1985' I.C. date codes.

Click to expand...

Yup, early-mid '80s.

It is a pleasure to use,  it is accurate enough,  and it sounds nice to my ears.  (It will be even better after a full servicing regimen)

The  IC's are socketed for easy removal/replacement, and it contains a pair of Valley People TA-101 transistor arrays

Click to expand...

I suspect the TA-101 are the only really obscure parts in there, a NE772*** that may be getting old, probably an obsolete transistor array for the log conversion, but there is a lot stuff going on inside there.

Maybe later I'll post more about the design philosophy but in short strokes, it is not a conventional above/below threshold compressor  for the central dynamics section, but a full range compressor. So twiddle the compression ratio and the entire dynamic range gets squeezed.  Since even a little compression applied full range will raise the noise floor up uncomfortably high, I overlay the compressor with a downward expander.  This downward expander threshold basically sets the unity gain inflection point for the noise floor before the downward expander starts reducing gain again.  Calculations are made in the side chain to start reducing compression as needed to smoothly pass through this unity gain inflection point.

So this is unlike other compressors that have thresholds and knees, or whatever... You set  the peak limiter for where you want the peaks stopped, and set the expander for where you want the noise floot to end up, and than squeeze between with compression ratio.

I was very excited about this compressor as it was a unique approach to studio compression. Unfortunately I had a falling out with my business partners and left around when this was released.  I don't think it was ever advertised so never really got a fair shot in the marketplace.

Glad to see one alive in the wild.  8)

JR

PS IIRC there were a bunch more small details but i am going from memory.

*** 572 were cheap compander ICs, but I didn't even use their internal gain elements, I just used them as decent quality compact precision rectifiers for use in my side chain.  That circuit board would be even more crowded if I had to flesh out a few precision rectifiers too.

pic 3 attached

The solder side of the PCB.  The PCB is labelled Rev. 5 File. A

envelope said:

I thought I would start a general thread for informational purposes.

here is an interesting piece. LOFT 410

It appears to be from mid 80's based on some '1985' I.C. date codes.

It is a pleasure to use,  it is accurate enough,  and it sounds nice to my ears.  (It will be even better after a full servicing regimen)

Click to expand...

I wouldn't expect much improvement from newer op amps, The T07x were adequately fast.

The Valley People (Paul Buff) VCA was IMO SOTA for the time, the most recent THAT corp VCA is better on paper. For dynamics processing side-chain manipulations are more audible than VCA path performance, just saying.

The  IC's are socketed for easy removal/replacement, and it contains a pair of Valley People TA-101 transistor arrays

Click to expand...

After three decades at least re-seat the socketed parts, and replace any obviously faulty ICs.

=======
I will attempt to explain (remember?) what the controls do.  ;D

From Left to right..

meter-  IIRC I might have done something slick like indicating + dB from the bottom up, and - dB from the top down to make a more intuitive to read display.

--(downward) expander---
in/out  -  looks like defeat for the expander section

release- release time for just the expander function

atten switch- limits the amount of downward expansion,  3dB is more natural sounding where the noise floor does not completely go away. -15 and -40 dB allow full gating of noise floor.

threshold- this sets the unity gain inflection point where gain shifts from boost to cut at the noise floor

Slope- the gentle 1:1.3x slope is most natural sounding. It doesn't take much expander slope to sound too sudden.  1:2 is like a listening to a companding NR playback that wasn't compressed. 1:20 is full on gating

---(full range) compressor----

attack time

out/key  - a little fuzzy but I think this selects a rear panel jack to feed the side chain instead of normal audio input.

release -

comp sw- FM  IIRC there was a broadcast application there the side chain was shaped by FM pre/de emphasis curve.  out- is probably bypass, and duck, is for use as a signal ducker. The duck function might grab the input from the other channel, but I am not real confident about that,
 
Ratio- varies from 1:1 (no compression) to infinity, but it really can't do infinity so some fixed high ratio (8x?).  Remember this is a full range compressor so a little goes a long way.  2:1 cuts the dynamic range in half.

Threshold- since this is a full range compressor this sets where the 0dB unity gain transition occurs during the full range compression, provides enough range to grab a -10dB(V) nominal 0VU or +4 dB(u).  There is no need to drop the threshold down lower to get more compression, just turn up the ratio knob,

------Limiter-------
This is a fast attack/fast release limiter that operates on top of the compression.

The red LED indicates when signal is above/below limiting threshold to help when setting this threshold.

Threshold sets limiter threshold.

out/in defeats just the limiter.

==== S-D switch=====

Stereo link switch for stereo operation, or Dual mono...


Caveat... This is from 30 years ago so my recollections are not very sharp.  After three decades of hindsight, this fancy compressor can deliver similar transfer function to a premium conventional compressor-limiter-expander, hopefully just better and a little easier to dial in for good results.

As you can see I could have added even more controls (pots in place of the 3 pos switches), but I worried it was already too complex for many users. Recall that dbx had great success with their one knob squeezer back then.

JR

PS: No I don't have a schematic.  8)

Valley is back in the market and you can still get the ta-101

I have one of these too, Serial #56. I believe 1 channel is dead. I got this throw in a box of stuff for free over 4 years ago, and I think its about time to get her up and running again. Thanks for the theory of operation John, I wasnt initially sure if it was worth fixing, but it sounds like a unique and powerful piece. Cant wait to get it up and running and smash some drums.  ;D

So glad to find this info.  As well as this message board.
I just acquired a Loft 410 at an auction here in the Greater Cleveland area. Looks beat up but I'm about to plug it in and see if it works. I'll keep you posted.

I am really glad to hear that a few of these are still around. I thought it was a far more logical approach to dynamics processing but regrettably there was never any advertising invested to sell this "new improved" way, so it didn't gain much traction in the marketplace.

The design concept was instead of the conventional tweak pre-gain, compression thresholds, and compression ratio, to get more compression, just have one knob vary how much compression is applied over the entire dynamic range.  The unintended consequence from this is that the noise floor will get boosted by the compression gain build up.

To manage this noise floor a downward expander is integrated into the side chain compression management to gracefully pull back gain following a preselected slope to pass through a noise floor unity gain intercept and continue expanding downward below that.

Overlaid on top at high level was also a fast limiter.

After this roundabout journey we end up with a transfer function not unlike (some) conventional compressors but with more precise control of various aspects.

If i was designing this today (using analog ) I would use the latest THAT corp VCA, but the convoluted side chain processing might be easier (certainly cheaper ) using a microprocessor. 

In hindsight trying to change how the industry applied compression without advertising or a detailed education process was more than a little optimistic.  : (good training for later working at Peavey, they didn't advertise much either).

JR

Sound like a cool piece! Lots of control.......

There's a couple 400 gate/limiters on musicgoround.......... none of the 410s around from what I can tell.....

scott2000 said:

Sound like a cool piece! Lots of control.......

There's a couple 400 gate/limiters on musicgoround.......... none of the 410s around from what I can tell.....

Click to expand...

The gate/limiter was a completely different animal.... It was designed for live SR to squelch low level PA noise between sets (you know when the music stops and hear hum and buzz, and whatever).

To keep it cheap and clean, instead of a VCA audio path active all the time (and VCAs were not that good back then especially used for the full dynamic range), I used an OTA shunt in parallel with the feedback loop of an op amp (TL07x). So when not gating or limiting you had the path fidelity of a unity gain op amp (i.e. very good).  8)

When noise gating or limiting, the distortion from the OTA was inconsequential compared to the gain modulations being performed on the audio.  So cheap and dirty when gating/limiting, but clean as a whistle when not active. 

JR

So here we are a whopping 2 years and 1 day (oh...so close) since the last post on the subject and I just picked up my second Loft 410. Got if for a bank-busting $11.50 because it’s as is, and had 3 broken pots. I have to admit I like these things. Definitely has a sound, unique to it. As a previous member said, the side chain detection circuitry has more of a profound effect on the “sound” of a Compressor then the I/O circuitry or VCA used.
Interesting coincidence... My first unit is serial number 71. One prior to the original poster’s. Not surprising however is that we’re both in Canada.
Anyway... I plan on picking it up from the seller (also here in Toronto... Loft seemed to sell well here), in a few days and then rebuilding the thing. Any interest in Schematics? If so I may start to reverse-engineer it while I’ve got it on the bench. That is of course if nobody has any objections. Otherwise I’ll keep the schematics private.
Cheers, Dean.

I have one of these, and, until now, I've found zero references to it on the net. I'm very glad to see informed comments on the unit here. I'm inclined to do the recapping and other maintenance on it, now.

yamobedeh said:

I have one of these, and, until now, I've found zero references to it on the net. I'm very glad to see informed comments on the unit here. I'm inclined to do the recapping and other maintenance on it, now.

Click to expand...

Good to hear that another one of my old children are out in the world and kicking around.  8)

The 410 was my blank sheet of paper approach to dynamics processing. Instead of juggling thresholds and compression ratios to get more/less compressions one knob increased or decreased it for the entire range. The noise floor with downward expansion was set by its own control and the high level limiter likewise... This in principal eliminated all the interactions between controls that required constant re-tweaking with typical technology (still).

I learned from 15 years at Peavey that customers do not appreciate having to learn new controls to perform existing functions, even if better.

JR