I have a flourescent VU meter board, and it has BA658s onboard the unit. This meter requires an external amplifier/rectifier time constant to drive it.
I remember the LED VU chip BA6124 had a built in high gain rectifier. So i built a circuit around this IC and the amplifier actually drives this VFD nice. problem is, i only have one IC. and im actually wasting it, as im not using the bult in 5-LED scale. just using the amp/rect section.
Does anyone have an equivalent circuit of the amp portion of this IC?
I tried building my own LM324 driver/rectifier and i couldnt get it to work at all. I need something that works on single supply.
any ideas? thanks.
I remember the LED VU chip BA6124 had a built in high gain rectifier. So i built a circuit around this IC and the amplifier actually drives this VFD nice. problem is, i only have one IC. and im actually wasting it, as im not using the bult in 5-LED scale. just using the amp/rect section.
Does anyone have an equivalent circuit of the amp portion of this IC?
I tried building my own LM324 driver/rectifier and i couldnt get it to work at all. I need something that works on single supply.
any ideas? thanks.
The problem is tying the op-amp + input directly to one of the power supply rails. No op amp really likes this.
The easiest fix - without changing your basic circuit - is to create an artificial 'ground' reference for the op amp at half the power supply voltage (R5,R6,C5 on my diagram) and then reference both op-amp inputs to that voltage (R7, R8). This also means you need to DC isolate the op amp circuit from the rest of the circuit on both sides by means of two 47uF capacitors (C2, C3).
This diagram ought to illustrate the point.
(Don't pay any attention to me mistakenly labelling R2 as 47k)
The easiest fix - without changing your basic circuit - is to create an artificial 'ground' reference for the op amp at half the power supply voltage (R5,R6,C5 on my diagram) and then reference both op-amp inputs to that voltage (R7, R8). This also means you need to DC isolate the op amp circuit from the rest of the circuit on both sides by means of two 47uF capacitors (C2, C3).
This diagram ought to illustrate the point.
(Don't pay any attention to me mistakenly labelling R2 as 47k)
Attachments
Well the ones on the output are the ones of concern; the input ones are loaded by the op-amp, it's bias- and feedback resistors - which is a load slightly less than 100k.
The output is loaded by *about 1k*, so using the formula:
f = 1/(2Pi.R.C) gives a turnover frequency of around 3.3Hz.
Since electrolytic caps are only specified to around +/- 50% tolerance or worse, it's pointless trying to be TOO fussy with the maths.
My schematic drawing software is Orcad V3. Old, yes, but then I prefer my Garrard 301 over an Ipod, so don't try to talk me into something 'modern' that wants 2GB of memory.
Yay. that circuit worked like a charm. if i only knew exactly how it worked. i know your biasing the differential around half of VCC but i still dont understand too well. its been forever since ive screwed with op-amps.
anyway... also the display was too fast. i had to change my constant to 100uf and 10k. still a bit too fast. 100uf and 1meg is fast at the top but too slow of a drop rate.
anyway... also the display was too fast. i had to change my constant to 100uf and 10k. still a bit too fast. 100uf and 1meg is fast at the top but too slow of a drop rate.
Got both channels hooked up now. theres a problem somewhere. One channel seems to bounce around the 0 mark. depends on volume, but its rather fast. but the other channel is linear. but its got way more gain than the first channel. hmmmm.....
hehe nevermind, i had a shorted diode. both channels are working alright. but its too touchy. meaning if the set knob is just below a point, theres nothign on the vu. but if i touch it just above the point, then the VU pegs. hmmm...
hehe nevermind, i had a shorted diode. both channels are working alright. but its too touchy. meaning if the set knob is just below a point, theres nothign on the vu. but if i touch it just above the point, then the VU pegs. hmmm...
A resistor/capacitor combo like that has an exponential decay, so it WILL drop off fast initially, and then more & more slowly. Different behavious needs a different circuit.
100uF cap there is remarkably high; it makes me think the rectifier diode isn't operating correctly.
Ideally the op-amp likes to work at exactly half its supply. With the divider at 5V, you may be clipping the bottom off the sinewaves withing the op-amp, which means the DC bias on C3 varies with signal level, and that DC might be showing up on the VU meter.
How much gain do you need before the diode? Maybe more amplification is in order.
Strictly speaking, you don't have a VU meter, because the dynamics are wrong for it to be a true VU meter.
The LM3915, and 3916 datashheets have pe-conditioning circuits, rectifier circuits, etc. with the right dynamics for VU, PMM etc. They are just preconditioner circuits, suitable for ANY bargraph driver IC. BUT from what I recall, they are also general purpose op-amp designs that assume symmetrical supply voltages.
The LM3915, and 3916 datashheets have pe-conditioning circuits, rectifier circuits, etc. with the right dynamics for VU, PMM etc. They are just preconditioner circuits, suitable for ANY bargraph driver IC. BUT from what I recall, they are also general purpose op-amp designs that assume symmetrical supply voltages.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.