500 series Level-Loc project
« on: January 20, 2014, 06:23:11 AM »
Hello all,

I’ve been working on putting the Level-loc circuit onto a 500-series card with a few added extras.

My principle want for the circuit is for it to be used in mixing or after a mic pre so I’ve simply omitted the I/O transformers, used INA134 and DRV134 for I/O and padded down the input (initially with a 40dB pad) later a 30dB pad (still undecided).

Quick overview of what I have (probably incorrectly) done:

Primarily I have assumed that this would never be used as a HIFI solution, therefore the bypass is not ‘true-bypass’ as in – it passes through the INA/DRV chips whatever happens.  The bypassing is done with the unbalanced signal and controlled by relay.  The circuit initially ran off a 9v supply so I have replicated this here using a 7809 off the +16v supply.  I also only have the basic version of Eagle so I've put it on to a 160mmx100mm board.

So after some laying-out and some double checking I made up the proto using the toner-transfer method and some FCl:



But I also want a mix/blend control – I have recently made a splitter-blender for bass guitar so I whacked that on the input/output to then realise that the output is phase inverted so I made a quick change to invert the ‘send’ element and breadboarded the revised blender:






In faffing with the circuit I have also tried a pot in place of the ‘distance’ switch (which I guess works just like a threshold) but the switching as it was worked really nicely – so why fix what ain’t broke?

Anyway I figure an HPF on the input (pre-compressor) may be useful so I’ve tacked on a simple RC filter (10k/100nF -160hz?).  I also thought an LPF would also be useful so, similarly I tacked on another arbitrary RC filter (this example was 10k/6n8) to the output (post compressor) to tame any top end so it can be blended back ‘underneath’ the dry.

I also read about changing the release time – ‘when in Rome’ right? – so I used the faster value of 1uF and added a 2uF in parallel to slow it down a bit.

There are some sound clips here:

https://soundcloud.com/minoian/sets/lvl_lok-500

So far I'm happy with the progress and once I've updated the basic schematic to include the blend and other circuit mods I'll upload it here.  I'm clueless when it comes to circuit element interaction so these separate elements have literally just been put together without any real consideration.  Once I've uploaded the schematic I would be delighted if others would provide some constructive criticism...  :)

On a massively more important note   ::) - I've not decided on a name yet...  ;)  'LVL_LOK-500' is my current reference...

Ian
 8)
« Last Edit: January 20, 2014, 06:42:05 AM by irfrench »


mylesgm

Re: 500 series Level-Loc project
« Reply #1 on: January 20, 2014, 07:07:24 AM »
sounds great… will you put a self etch file or similar?  I'd be interested in trying...
Check my work at www.mylesmumford.com

Re: 500 series Level-Loc project
« Reply #2 on: January 20, 2014, 07:21:53 AM »
Thanks.  I've been looking around for a level loc project and found the Sherlock (now discontinued) but decided I'd prefer it in a 500-format.  So I thought I'd give it a go.

sounds great… will you put a self etch file or similar?  I'd be interested in trying...

Yup - the only awkward bit is cutting the board down to fit in the 500-slot.  Not difficult though.

I'm still uhhming and ahhing about input levels and I have a cheap-o 1:10 transformer here that I may try.  Once it is a bit more finalised I'll be a bit happier for people to try it (don't want anyone wasting time/money) but I can't see why it won't end up being a nice little project.

I'm hoping that some more clever-sorts will be able to help out and make sure things are functioning properly.  And more to the point that my hamfisted 'techniques' aren't ruining the party.  :)

Ian.
 8)

Re: 500 series Level-Loc project
« Reply #3 on: January 21, 2014, 06:09:15 AM »
Hello all,

I have attached the schematic that now includes blender and filter sections.  As mentioned before - these were just sections of circuits that I have tacked together (the blender is a variation on a bass guitar pedal schematic)

If anyone can provide guidance on sections to remove/change/improve/poke-fun-at please feel free.  8)

Ian.
 :)

Re: 500 series Level-Loc project
« Reply #4 on: January 29, 2014, 12:40:41 PM »
Evening all.

Just a quick update.  I have added the mix circuit to the layout and included switchable input pad, hpf, lpf and release controls.  Hopefully all useful additions.  With the exception of the pad control simply omitting the modifier switches would result in a 'normal' circuit, so the build can be tailored to individual wants/needs.



Eventually the front panel could look something like (there are some changes I can already see are needed):



I'm really busy at home/work ATM but will try to get a test build done ASAP.

 :)

okgb

Re: 500 series Level-Loc project
« Reply #5 on: January 29, 2014, 03:39:45 PM »
Nice work , if there's a mini kit, I'll buy one to support your effort!
GKB Audio / Greg Boboski

Niketouille

Re: 500 series Level-Loc project
« Reply #6 on: February 04, 2014, 05:58:07 AM »
Oooooooohhhhhh, I'm all excited

RuudNL

Re: 500 series Level-Loc project
« Reply #7 on: February 04, 2014, 10:40:19 AM »
I notice that the FET only gets a control voltage.
Usually a part of the audio signal is superimposed on the control voltage, to reduce the distortion.
Did you measure the distortion?
As long as the signal on the drain is low enough, distortion is probably not a problem, but a low level usually means: not the optimal signal to noise ratio.
There is a solution for every problem!

http://www.vansteenisaudio.nl

Re: 500 series Level-Loc project
« Reply #8 on: February 04, 2014, 01:07:18 PM »
I notice that the FET only gets a control voltage.
Usually a part of the audio signal is superimposed on the control voltage, to reduce the distortion.
Did you measure the distortion?
As long as the signal on the drain is low enough, distortion is probably not a problem, but a low level usually means: not the optimal signal to noise ratio.

Hi, thanks for the comment.  In all honesty I have just copied the schematic wholesale from the previous 'Sherloc' circuit which in turn was a slightly tweaked version of the original.

The only changes I have made to the FET is to use a 2n5457 instead of the original - so if there are better alternatives (I used a 2n5457 as I have a load here, I'm happy to use others (J112?) as I believe the 5457's are now discontinued.) or if there are tweaks to improve things - i'm all ears!

Ian
 :)

Re: 500 series Level-Loc project
« Reply #9 on: February 04, 2014, 04:58:57 PM »
Ploughing on - I've stuffed the v1.11 board.  Tested the basic circuit OK.  Am now going to try out the 'mods'.





Switches are weirdly mounted at the minute as - unsurprisingly - the cardboard aint that strong.  The board needs a couple of revisions (more space behind switches, align pots, maybe move bypass switch forward to allow for thicker front panels) and I'm thinking about how I could route the I/O if I don't want the daughter board.

Ian
 :)


mjrippe

Re: 500 series Level-Loc project
« Reply #10 on: February 04, 2014, 06:56:18 PM »
Looking good Ian!  Thank you for sharing this as I'm rather enjoying watching the whole process.

RuudNL

Re: 500 series Level-Loc project
« Reply #11 on: February 05, 2014, 06:04:01 AM »
About the distortion reduction: most of the time a principle like this is used:



A portion of the (controlled and in-phase) audio is superimposed on the control voltage.
By adjusting R16x in this example, you adjust the distortion to a minimum.
(Principle shown: 1176 compressor)
There is a solution for every problem!

http://www.vansteenisaudio.nl

Re: 500 series Level-Loc project
« Reply #12 on: February 05, 2014, 10:22:14 AM »
A portion of the (controlled and in-phase) audio is superimposed on the control voltage.
By adjusting R16x in this example, you adjust the distortion to a minimum.
(Principle shown: 1176 compressor)

Thanks for the information - so in the case of the schematic I posted above I am looking at modifying R3 (currently 33K)? If there are any starting suggestions and/or useful references for me to look at please let me know.

Ian.
 :)


Niketouille

Re: 500 series Level-Loc project
« Reply #13 on: February 05, 2014, 10:54:20 AM »
Why do you want to reduce distortion guys ? I always thought the levelor was used to add distortion ? Anyway, is there is a way to increase distortion using the same principe ?

Re: 500 series Level-Loc project
« Reply #14 on: February 05, 2014, 10:59:38 AM »
Why do you want to reduce distortion guys ? I always thought the levelor was used to add distortion ? Anyway, is there is a way to increase distortion using the same principe ?

I'm just asking from an interest point - I'd give it a try and if it took anything away I'd return it to its original value.  The reasons for me using the Level-Loc are not based on fidelity!

The signal going into the Level-Loc is small (mic level) so I'd imagine that this element is not 'struggling' anyway?

Ian.
 :)

 

RuudNL

Re: 500 series Level-Loc project
« Reply #15 on: February 05, 2014, 11:40:50 AM »
With a FET as control element, you will get distortion anyway...
The advantage of distortion reduction, is that you can use much higher levels, and thus get an improved signal to noise ratio.
(Without distortion reduction, the maximum Uds will be in the order of 30 mV, without getting serious distortion.)

(I suppose you could take the signal from the emitter of Q3 and superimpose this,- with or without amplification-, on the control voltage. But if you like the distortion, leave it as it is!)
« Last Edit: February 05, 2014, 11:45:37 AM by RuudNL »
There is a solution for every problem!

http://www.vansteenisaudio.nl

Re: 500 series Level-Loc project
« Reply #16 on: February 05, 2014, 11:56:54 AM »
With a FET as control element, you will get distortion anyway...
The advantage of distortion reduction, is that you can use much higher levels, and thus get an improved signal to noise ratio.
(Without distortion reduction, the maximum Uds will be in the order of 30 mV, without getting serious distortion.)

Thanks again - all advice is welcome.  From a complete layman's perspective - if the FET was always distorting could this be part of its 'sound'?  If it is should the design parameter for the FET - assuming we want a reproduction of the original - be to maintain the contribution at that point?

I should reiterate my level of working electronics knowledge - not a lot - so any advice is welcome.

I should also point out my assumption that mic vs line is about -40dB, the transformer on the input of the Level Loc was providing 20dB gain so I have assumed the LVL_LOK-500 would be working at line level -20dB.

As the compression threshold is fixed according to the original values the input level needs to be set accordingly and thus the fet would only be seeing what would be normal at the Level Loc input - so any high signal levels would be too hot for the compressor to function 'normally'.

Not sure if that makes sense?

Ian
 :)

Biasrocks

Re: 500 series Level-Loc project
« Reply #17 on: February 06, 2014, 11:02:58 AM »
It's a Leveloc, distortion is part of the charm.

I say leave it original, but if you must offer a redesigned version; put it on a switch.

Regards,
Mark
http://SharktankPro.com

"I'd rather use an SPX90 than a UA plugin....." Joe Barresi

Re: 500 series Level-Loc project
« Reply #18 on: February 06, 2014, 11:41:02 AM »
Good timing...  :)

Managed to get the off-board wiring done:



Now - the honesty part... The HPF and LPF don't behave as I expected.  I'm guessing I've screwed up somewhere - BUT - following the above discussion about the Level-Loc I've also been thinking about what I want the unit for, what I expect from it and how I'd use it. 

I've come to the conclusion that I don't actually want those things!?  :-X

I have also redone the layout to get rid of the daughter-board...  Too many jumpers for my liking but I'm struggling to work my way around that without the daughter-board.



But will now probably redo it without the LPF and HPF sections.  I'm toying with the idea to remove the PAD option but I think it is just too darn useful to not have it there!

It'll also clear up some real estate on the front panel and allow me to get some big-ol-knobs on it like I originally wanted to! 23mm knobs...




Ian.
 :)

« Last Edit: February 06, 2014, 12:06:31 PM by irfrench »

Re: 500 series Level-Loc project
« Reply #19 on: February 07, 2014, 11:50:19 AM »
Yet another lunch hour dedicated to the cause.  I decided to remove the LPF and HPF controls (easy to do as switches are on-off-on) and try the revised front panel.  If you're interested its printed on paper and glued to a re-purposed DVD case.  Real classy.  ;D  The eventual panel will be black with white writing.







The new front panel feels a lot better to use, the work flow feels more natural too.  I guess that is lesson learnt about over-egging the pudding?

I've decided to order a prototype front panel based on the above layout and will make one last (hopefully) prototype so I am sure it will work 'nicely'.  The revised layout with LPF and HPF removed looks a lot like this:



I'll do some JFET testing too as I've been using a 2n5457 so far which may be unnecessarily difficult to get hold of...

Ian.
 :)