Universal Bargraph driver - led meter for SSL9k, GSSL

GroupDIY Audio Forum

Help Support GroupDIY Audio Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.

Ephrem

Well-known member
Joined
Apr 29, 2007
Messages
52
Location
France
Hello everyone,

Please find here gerbers files for Universal Bargraph driver SSL9k :wink: .

This is a precision full wave peak detector.
I did not find them on the forum, therefore i've created them

http://puzzlemind.free.fr/Ephrem/Gerbers.zip

pcb.JPG


The original design is from Keith & lm3915 datasheet : http://www.beatbazar.com/guests/ssltech/universal-bargraph/bargraph.pdf

I'm not a professionnal designer. I 'm gonna test this pcb soon.
I made that circuit with veroboard. that works perfectly. Really accurate.

The PCB :
ledmeter.JPG


You can download the PDF here :
http://puzzlemind.free.fr/Ephrem/LEDMETERpcb.pdf

On this forum, I see many preamps without bargraph or peak led.

From my recording experience, You have to check preamp output level (0dB at least) before AD conversion :!:


Regards
Ephrem
 
Usually the line input device the preamp is feeding has some sort of metering; a peak / clipping LED is a very handy thing to have. My First preamps had metering, but I just never really found them to be of any useful
purpose. The meters only read true if you are driving a level calibrated line input.
 
[quote author="BYacey"]Usually the line input device the preamp is feeding has some sort of metering; a peak / clipping LED is a very handy thing to have. My First preamps had metering, but I just never really found them to be of any useful
purpose. The meters only read true if you are driving a level calibrated line input.[/quote]
I understand what you're saying. Life can be simple (say just one gain-dial and metering at the recorder), but when one additional knob/setting is added then the additional requirements build up.

Example a preamp I'm toying around with now (FWIW, an Altec 1589B)... can have up to 77dB of gain, yet has just a simple single volume-control per channel (after the first 33 dB of gain) and there's no metering.
It fits that 'check level at destination'.

But as soon as a gain-control for the first stage is added one would want to know the optimal combination of gain- & volume-control: amplify enough w.r.t. noise, but don't make the first-stage clip.

By ear is still always possible of course, but that's not the point of this thread.

Regards,

Peter
 
The general idea is to keep the material of interest somewhere above the noise floor and below clipping on peaks. The setting for maximum dynamic range falls halfway between these two points and is dependent on what the audio source is providing. I tend to err on the side of higher gain settings because I would rather have an ocassional clip on high peaks rather than the constant veil of noise covering up the material of interest.

I'm sure others will have a different view and method according to what they are recording.
 
For example if you chain a ssl9k + gssl+ sound card, you need to check level signal between ssl9k and gssl.
However, both diy design do not include level meter.

I know you can bypass the gssl to check ssl9k audio level, but its easier and more instinctive to visually check output level of each rack.

Ephrem
 
Ephrem,

After I designed it, I wished that I'd used the unused half of the first TL072 as a noninverting, unity-gain buffer...

Basically taking the input signal to pin 3 and taking the output from pins 1&2 instread of having them do nothing... it would make the input less impedance-sensitive, is the (slight) advantage.

Keith
 
Just an idea, but maybe consider turn the LEDs, so it will be easier to bend them in case you want the PCB at a 90° angle from the front pannel.
 
Good work!

Another thought, and Roger may wish to share his thoughts on this:

I included the resistors (R12-R21) to reduce the heat load in the IC.

I think that Roger's use of the current regulator may take less baord space and be more compact... although the regulator cost ma or may not be higher than ten resistors...

Me, I have no particular preference, but I also had the idea of "dim-lighting" the unlit LEDs, so that in total darkness you could still get a sense of when the bargraph "runs out" if you know what I mean... it takes ten more resistors, that's all..

Any interest?

Keith
 
hi all,

Would anyone be willing to send me a copy of the pdf file? the link does not seem to work.

 
Sorry Guys! :-[ I ment this pdf for the component values...

http://www.beatbazar.com/guests/ssltech/universal-bargraph/bargraph.pdf
 
Off in the distance... is that a choir I hear singing! the clouds seem to be parting overhead, with a sunbeam shining through! Thanks So much Sredna! You are truly the next best thing to a Modified Delorean!  8)
 

Latest posts

Back
Top