• Welcome to our site! Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

3 channel spectrum analyser questions

Status
Not open for further replies.

Neil Groves

New Member
Hi Guys...i am building the circuit that i have supplied a link to but can't find any info about how it works, setting up or testing, so far i have built the input stage and the top filter (u1a and u1b) respectively, on feeding the input with a gorgeous sinewave, everything looks good untill the signal comes out of pin 9 where is is ringing badly? i don't know what to expect at the output, i am assuming that as the frequency changes, the voltage at the output rises and falls respectively as it does on mine, i expected a clean rising and falling D.C voltage at though instead of a peaky waveform at the Cathode of D1 though, can anyone give any insight to this circuit please?

I am aiming to be able to hook up my signal generator and sweep though 40Hz-16Khz and seeing the leds sweep through from bottom to top as the frequency rises.....or am i mistaken about what this circuit does?

Neil.

http://www.aaroncake.net/circuits/3chspec.asp

http://www.aaroncake.net/forum/topic.asp?TOPIC_ID=8847&
 

audioguru

Well-Known Member
Most Helpful Member
The schematic was wrong for many years but has recently been corrected.
It uses obsolete and hard to find ICs. Ordinary ICs will not work unless the circuit is re-designed.
 

Neil Groves

New Member
i tracked down the LM3900N without much trouble, and i thought i might use LM3914/5....i guess i need to learn some analogue electronics lol

Neil.
 

audioguru

Well-Known Member
Most Helpful Member
The LM3900 was a very strange amplifier and the only one like it. Today thare are many opamps that can be designed into a similar circuit.
The LM3915 is a standard world-wide LED VU meter for up to 10 LEDs. The Japanese smaller one is (was) very rare.
 
Last edited:

Neil Groves

New Member
AudioGuru, did you ever build this circuit? do you know what should be present on the cathode of D1? i was expecting a steady D.C voltage that rose as the frequency at the input of the circuit increased? but i am getting a rising D.C voltage with what looks like a waveform attached to it, or it might just be crud, i'm not sure. i'm not quite ready to give up on this circuit yet, but i have seen others that i want to try, my end requirement is a circuit that will flash lights to music divided into 3 channels, bass, middle and treble.

Terry.
 

Reloadron

Well-Known Member
Most Helpful Member
my end requirement is a circuit that will flash lights to music divided into 3 channels, bass, middle and treble.
Hi Ya Neil

Sounds like what you are after is a color organ circuit. If you run a Google of "Color Organ" and try "Color Organ Circuit" you should get countless hits including some kits from Jameco. You will see variations for 120 VAC and 12 VDC as well as driving incandescent and LED light sources. Must have been about 35 years ago I built one using a then Radio Shack kit that would drive 300 watts per channel. I had a really cool Christmas Tree for a few years. :)

Here is an example of a 4 channel version designed using an LM324.

If you look at most of the circuits of this type you will see their similarities. Also, many of the older designs floating around out there use the LM3900 including the version you posted.

Ron
 
Last edited:

audioguru

Well-Known Member
Most Helpful Member
AudioGuru, did you ever build this circuit? do you know what should be present on the cathode of D1?
No, I don't build crappy old circuits full of errors that use obsolete parts.

I used an online calculator to show its mid-frequency filter response and found another error:
R13, R15 and R17 are listed to be 580k which is not a standard resistor value. Maybe they should be 680k which makes almost the same results.

The mid-frequency filter peaks at 628Hz which is too low for music but OK for telephone quality speech.

Of course the filter rings like a bell as shown in the Step-Response I calculated, its Q is way too high at 15. I think with only 3 filters, they should have a flat response in their bandpass, not a narrow peak.
 

Attachments

Last edited:

Neil Groves

New Member
ok guys, you got me convinced, i'm going to scrap this circuit and find a better one

thankyou all for your time.

Neil.
 
Status
Not open for further replies.

Latest threads

EE World Online Articles

Loading

 
Top