weird electronics idea for desecrating legacy gear

[abbey road d enfer]

3914-Linear increase
3915- 3db log
3916- VU db
any pros please feel free to add/subtract to that, I am still hazy on the 15/16 optimization)
and it was clear that the 3916/15 were more like watching the needle for Gain Reduction.

Remember that a needle VU-meter is close to linear response, so 3914 would probably the closest IF you had retained the Wheatstone bridge circuit, but now, with the different configuration, it's very hard to tell. You have to experiment and opt for the one you like better.
Many designers (including me) have shunned the LM391x series in favor of the use of discrete (?) comparators and a custom resistor chain, which allows almost infinite tayloring of the response. It takes two 14-pin IC's, but it's not a big issue since the number of LED's is what will govern the size of the PCB.

 

[andyfromdenver]

Big update on gain reduction LED meter v2.0
Sorry no vid at present.
I have completely reconfigured the circuit so everything shown in the video and accompanying schematic for Gain reduction should be disregarded.  This has been a great learning exercise for me!

Ok, to summarize:  The photoresistor (1of2 in t4b) has been placed in the upper (or can do lower for a falling appearance) leg of a voltage divider with Vout as the input to ONLY an LM3914.  The Output meter benefits from the opamp buffering, but the GR does not require an opamp and even though it works like in video to show peaks, it does not "hold the needle" visually with a steady input tone. 
I have the Lm3914 in a "variable voltage indicator" configuration which uses very few parts, maybe 4. 
Now when inputing a steady voltage signal and adding compression the "needle" (led= dot mode, leds=bar mode) remains fixed until I stop the input signal.

My next step before final assembly off breadboard, is getting the calibration just right.

I took a bunch of measurements of when the photoresistor starts decreasing resistance for a given strong signal (I know that is all variable and that's what's making it tricky to calibrate).  I feel the LA2A starts adding compression with the knob around ~20, again I know that is dependent upon strength of input, so I am trying to ensure that that gets represented with a "blip" on the first LED.  However, I need to adjust the range because when I adjust for first LED light on at around 20 on the knob, I get to maximum level when I raise the compression knob to around 60.  So this is where I am fiddling

With the LM3914 in dot mode, it makes the visual of the deflection of the "needle" (by needle I mean led in dot mode) more like the front grill of Kitt which looks cooler and saves mA

tbc!

If I post all my new schematic in Drawing board with a title "How to add LED meters to LA2A" would someone pretty please chime in with improvements or tips/ range adjustment info for the LM3914?  Otherwise, I'll just keep plugging away and post a final result here later. Hopefully I don't need to switch to a multiple quads comparator circuit, to get a satisfying result, but I see that would allow very fine tuning...sigh...

Thanks!

2.0! (really much better)
Andy

P.S. I've got to say, simply viewing the Output LED meter and listening is really all the aide that is needed. Getting this GR display just right will be extra eye candy.

 

[abbey road d enfer]

I took a bunch of measurements of when the photoresistor starts decreasing resistance for a given strong signal

The correct procedure is much simpler. Input a steady-state signal, use the threshold and/or ratio controls to achieve whatever degree of GR you define as "onset of compression" (maybe 1dB, maybe 3, it's up to you), then measure the voltage. Do the same for what you think is the max GR you want to see or what the circuit can achieve, and again measure the voltage. Now you have teh range of voltage to which your display must react.
For example the dbx160 has its 1st LED at 1dB GR and the last at 40dB, although it can provide up to 60dB GR, but it's not a workable situation.

simply viewing the Output LED meter and listening is really all the aide that is needed. Getting this GR display just right will be extra eye candy. 

Not only. I find the output meter almost useless, since it is just a visual indication of the resulting signal, for which the ear is the ultimate judge. ("mix with your ears, not your eyes")
The GR indicator is much more useful since it displays an operating parameter that is not directly perceived by audition, except in extreme cases.

 

[andyfromdenver]

The correct procedure is much simpler. Input a steady-state signal, use the threshold and/or ratio controls to achieve whatever degree of GR you define as "onset of compression" (maybe 1dB, maybe 3, it's up to you), then measure the voltage. Do the same for what you think is the max GR you want to see or what the circuit can achieve, and again measure the voltage. Now you have teh range of voltage to which your display must react.
For example the dbx160 has its 1st LED at 1dB GR and the last at 40dB, although it can provide up to 60dB GR, but it's not a workable situation.
Not only. I find the output meter almost useless, since it is just a visual indication of the resulting signal, for which the ear is the ultimate judge. ("mix with your ears, not your eyes")
The GR indicator is much more useful since it displays an operating parameter that is not directly perceived by audition, except in extreme cases.

Thank you again!  Ok, I think I have all the tools now w/ your input to make the range adjustment from the data sheet.

I know what you mean about the output being superfluous, in the context of a board or DAW. I'm appreciating the additional reference. well, what the heck, I kind of have to now that the drilling is done!  : You're right though about the aide of the GR display, not just eye candy....eye vegetables.

Take care!

 

[andyfromdenver]

Ok good peoples
*and moderators, please feel free to move to The Lab if more appropriate*

Here is my V2.1 schematic for excellent results.
V+ =12V
photoresistor is from T4B for meter circuit.

The 3914 is not a good choice as the threshold voltage for LED 1 is highest
the 3915 has a first LED turn on signal voltage that jives very well with what my ears tell me as a gain reduction, and it scales nicely up to max compression.
This part is critical for me, because I know people want to see that baby smidge needle movement sometimes for super light compression.

I plan to modify the Velleman LED VU kit as I have already painstakingly bent those leds and mounted the circuit board.
It SHOULD (lol) be a simple matter of removing the input circuitry including the 741 buffer, and adjusting Rlo, Rhi, Ref out, Ref adj.

Once I complete that, I'll make a new vid.
Thanks for listening!

Andy

 

[Whoops]

Nice one Andy, actually I like it much better thant the sloppy meter in the originals, I like the 160A LED meters style, it works pretty well.
Also decent VU meters are quite expensive and it's nice to have other alternatives.

Well done

 

[andyfromdenver]

Nice one Andy, actually I like it much better thant the sloppy meter in the originals, I like the 160A LED meters style, it works pretty well.
Also decent VU meters are quite expensive and it's nice to have other alternatives.

Well done

Thank you so much!  Getting some positive responses from friends too, they like the visual.  I finally modified the Velleman kit and installed it back in the box, it's in my rack working   The schematic is slightly different than shown, so I'll update shortly and make a new video.  If a person made some boards or used ic friendly perfboards they could construct it plus the power supply for about $25.  But getting two Velleman  kits and power supply was about $60, so it pays to DIY (as I know we all know) 8)

I'm going waaaay aside here, but if there are any fans of the awesome late '90s psych rock band Olivia Tremor Control, I am living with Will Hart who is my musical hero.  Just got the studio (poctop U87s, Hairball 1176, scratch LA2A, CJs iron 6n1P DI, Neve Portico 2x preamps) set up in the new house and we are going to do some serious tracking  
We've got our Eno and Schmidt Oblique Strategies cards out, been listening to Steve Reich and 10cc for some inspiration  ;D

I'll update again!
andy

edit: I forgot I took this pic when i was testing the modified board before reassembly. The molex (if that's the right term) connector recieves the wires to the LDR in the T4B.