Please note that LTspice reads "2Meg2" as "2Meg". You need to write 2.2Meg.dogears said:
It can make a significant difference for 1k8.
Mixer internal level indicator
- Thread starter JAY X
- Start date
Help Support GroupDIY Audio Forum:
It can make a significant difference for 1k8.
Is that a "true VU" meter (expensive), or a mA meter labeled with VU scale? (I'm sorry if this has already been discussed).JAY X said:
Have you breadboarded that meter driver to confirm frequency response and accuracy?
JR
Hi!
The vumeter is of the cheap ones from JLM audio.
The vu buffer is not tested, sorry, I will try this week ASAP.
The other circuits, the signal present/peak, I have simulated them and seem to work.
Also I have simulated the peak led driver next to the vumeter drivers.
Nothing built yet... :
Jay x
The vumeter is of the cheap ones from JLM audio.
The vu buffer is not tested, sorry, I will try this week ASAP.
The other circuits, the signal present/peak, I have simulated them and seem to work.
Also I have simulated the peak led driver next to the vumeter drivers.
Nothing built yet... :
Jay x
There are two issues 1) ballistics and 2) frequency response.JAY X said:
True VU ballistics (IMO) are over rated, and academic so let's focus on #2.
From observation I'd be nervous about low level HF response. That topology of full wave rectification will work fine at low frequency, but at every zero crossing transition the output of op amp has to swing from -2 diode drops to + 2 diode drops before it can conduct. At low level and HF even a decent op amp like tl07x runs out out of open loop gain. This will look like a level dependent low pass filter.
For a wiggly VU looking meter, probably not a concern, but there are other precisions rectifier that behave better.
Before you take the advice of a tired old man who has designed too many meters, breadboard this one up... if it looks OK it is.
I like the Peak LEDs in addition (I had peak LEDs on my big VU meter consoles).
JR
The other circuits, the signal present/peak, I have simulated them and seem to work.
Also I have simulated the peak led driver next to the vumeter drivers.
Nothing built yet... :
Jay x
No. The NFB loop is completed via the diodes and meter.mskeete said:
Hi!
I have breadboarded the circuits. The Vu buffer with peak led, works well, although i opted for a different design with the diodes attached at the vumeter, and not in the feedback path.
The signal present/peak circuit doesn't work. With a TLO72, its output seems to latch up, and the leds remain light up....
At the output of the rectifier i measured -2.7v dc, while playing some music through.
I replaced the TLO72 for an OP275, the leds didn't light up with or without signal.
Maybe the rectifier design is not adequate. If there is another one i can build, please let me know...
Both IC are feed with +17v for LM339, and +/-17v for TL072.
I changed the values for the Lm339 resistor divider: 20k /620R/130R. to get the threshold voltages:
0.109vdc for -20dbu
and 0.615vdc for -2dbu.
Jay x
I have breadboarded the circuits. The Vu buffer with peak led, works well, although i opted for a different design with the diodes attached at the vumeter, and not in the feedback path.
The signal present/peak circuit doesn't work. With a TLO72, its output seems to latch up, and the leds remain light up....
At the output of the rectifier i measured -2.7v dc, while playing some music through.
I replaced the TLO72 for an OP275, the leds didn't light up with or without signal.
Maybe the rectifier design is not adequate. If there is another one i can build, please let me know...
Both IC are feed with +17v for LM339, and +/-17v for TL072.
I changed the values for the Lm339 resistor divider: 20k /620R/130R. to get the threshold voltages:
0.109vdc for -20dbu
and 0.615vdc for -2dbu.
Jay x
If you have changed the design perhaps share a schematic of your current design.JAY X said:Hi!
I have breadboarded the circuits. The Vu buffer with peak led, works well, although i opted for a different design with the diodes attached at the vumeter, and not in the feedback path.
The signal present/peak circuit doesn't work. With a TLO72, its output seems to latch up, and the leds remain light up....
At the output of the rectifier i measured -2.7v dc, while playing some music through.
I replaced the TLO72 for an OP275, the leds didn't light up with or without signal.
Maybe the rectifier design is not adequate. If there is another one i can build, please let me know...
Both IC are feed with +17v for LM339, and +/-17v for TL072.
I changed the values for the Lm339 resistor divider: 20k /620R/130R. to get the threshold voltages:
0.109vdc for -20dbu
and 0.615vdc for -2dbu.
Jay x
JR
Don't see an input ground connection for audio inputs.JAY X said:
Looks like typical precision half wave rectifier (positive output for negative AC input swing).
22k in series with inputs to the LM339 seems large but the worst case input bias current is like 400 nA so only single digit mV into 22k.
That should be pretty easy to debug... do you see changing DC voltage on C192 and 194 with changing input AC voltage.
Check voltage on op amp pins for valid output state operation.
JR
Which circuit are you talking about? If it's the first on top of the page, it should work. If you have negative voltage at the rectifier's output, you may have reverse-connected the diodes...JAY X said:
It should be easy to troubleshoot with a VOM...JAY X said:
The precision half wave rectifier output should never swing negative. Feedback diodes D8 and D44 will clamp the output at roughly -0.5V, reverse biasing the series output diodes,D15, and D45. Further the - input of both op amps should be held at 0V by negative feedback.
To confirm proper operation you can connect a DC voltage to the input. A positive DC voltage input will peg the op amp output at -0.5V with 0V at the hold capacitor. A negative DC voltage input will cause the op amp to swing positive with voltage gain.
JR
Its always the last thing you checked....JAY X said:Hi!,
Finally It works! A power connection at opamp was missing... :
Nerves are bad companions...
Now i have to adjust threshold voltages.
Jay x
JR
Hi!
Despite the circuits worked on the test board, after building the pcb i had some problems. Most are solved, but i still have problems with the two transistor peak led circuit. The led lights up without signal being present. (the leds are in the correct orientation).
I Made some measurements.:
At R1 (4k7), after the 2u2 C1 cap, I measured -0.85v dc and -16.5v at the base of Q1.
At the leds: -0.54v Anode, -0.74v Cathode. with respect to gnd.
The input resistors values are: 4k7/470R. this gives a ratio of : 10 x 0.6 = 6volts.
With other ratios the behaviour is the same.: 4k7/2k2 : 2.13 x 0.6 = 1.28volts
I checked the board for any trace or bad solder...
I also pulled out the 2u2 input capacitors, then the leds did not light up.
Jay x
Despite the circuits worked on the test board, after building the pcb i had some problems. Most are solved, but i still have problems with the two transistor peak led circuit. The led lights up without signal being present. (the leds are in the correct orientation).
I Made some measurements.:
At R1 (4k7), after the 2u2 C1 cap, I measured -0.85v dc and -16.5v at the base of Q1.
At the leds: -0.54v Anode, -0.74v Cathode. with respect to gnd.
The input resistors values are: 4k7/470R. this gives a ratio of : 10 x 0.6 = 6volts.
With other ratios the behaviour is the same.: 4k7/2k2 : 2.13 x 0.6 = 1.28volts
I checked the board for any trace or bad solder...
I also pulled out the 2u2 input capacitors, then the leds did not light up.
Jay x
If your measurements are accurate and everything is connected per the schematic, Ohms law instructs us that 3.3mA of current is flowing in R1.JAY X said:
I am not familiar with the capacitor notation, is C1 backwards? Even then 3mA + seems like a lot for capacitor leakage.
not sure which LEDs...
3mA of base drive should force Q1 hard on..., which cuts Q2 full off, so LED D1 should be full on with roughly 2V across it.
That should be enough to chew on first...
JR
Similar threads
