Continue to Site

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.

  • 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.

Display circuit description

Status
Not open for further replies.

MrDEB

Well-Known Member
Here is my proposed circuit for 2-5x10 led matrix displays.
Only 1 led is showen for clarity.
attached is a description and pic
any corrections etc would be welcomed
 

Attachments

  • tiltredo51908display.pdf
    29.1 KB · Views: 238
  • display operation.zip
    444 bytes · Views: 151
Hi Mr DEB,
Your "buffer" is actually the old bandpass filter from the Oriental circuit. I changed it into a buffer since you are using different bandpass filters.
Your transistors have too many resistors and one is missing.

Your zipped DOC will not open.
 

Attachments

  • Mr DEB again.PNG
    Mr DEB again.PNG
    21.4 KB · Views: 353
Thanks

The description is a word doc. Best way to post file?? Here is a copy/paste
Audio in from filter is isolated from clean audio by the op amp buffer.
Output from buffer is feed to a peak detector composed of diode and 1uf capacitor (D3, C27)
Transistor Q7 is turned on with a low from the cd4017 output.
The input of cd4049 is then high, turning on transistor Q200(1 of 5 colums)
Q200 enables the row of 10 leds which are turned on according to outputs from lm3916
The trim pot is to adjust intensity of different color leds. The 50 ohm resistor acts as a safeguard against excessive current.
Leds are fed a burst current of 150-200ma according to color (adjusted by trim pot). The leds are only on for a split second according to the adjustment of the cd4017 timing.
 
"Audio in from filter is isolated from clean audio by the op amp buffer."
Yes.

"Output from buffer is feed to a peak detector composed of diode and 1uf capacitor (D3, C27)"
A 100k resistor is supposed to be in parallel with the 1uF capacitor to slowly discharge it.

"Transistor Q7 is turned on with a low from the cd4017 output."
No. The output from the CD4017 is high which turns on Q7.

"The input of cd4049 is then high, turning on transistor Q200(1 of 5 colums)"
Yes.

"Q200 enables the row of 10 leds which are turned on according to outputs from lm3916"
No. Q200 turns on a column and the LM3916 shows up to 10 rows.

"The trim pot is to adjust intensity of different color leds. The 50 ohm resistor acts as a safeguard against excessive current."
You don't need two 50 ohms resistors. Use one 82 ohms instead. Use a trimpot if you want.

"Leds are fed a burst current of 150-200ma according to color (adjusted by trim pot). The leds are only on for a split second according to the adjustment of the cd4017 timing."
You had two 50 ohms resistors which limit the current in 2V red LEDs to only 85mA. One 82 ohm resistor will limit the current to 104mA so the LEDs will look like they have 20ma continuously.
 
Last edited:
oups I got the 4017 wrong

in your diagram where does the 100nf cap connect to?
In the oritental circuit he shows only the 22k resistor, the diode to groud.
In Tina simulation it shows only 360mv for bias (I need to remeasure but pretty sure 360mv is close??)
need to switch around transistors.
The oriental circuit has 2 transistors per frequency but this one has 3 (alo of transistors).
gots to be a better method??
 
in your diagram where does the 100nf cap connect to?
It is from the audio output of a filter.

In the oritental circuit he shows only the 22k resistor, the diode to ground.
In Tina simulation it shows only 360mv for bias (I need to remeasure but pretty sure 360mv is close??)
The diode will have about 700mV across it.

need to switch around transistors.
No. The transistors are fine.

The oriental circuit has 2 transistors per frequency but this one has 3 (alo of transistors).
got to be a better method??
The Oriental circuit has the LM3914 directly driving the rows of LEDs with a very low average current. Your circuit has transistors to give a high peak current for good brightness.

You could use many fewer parts if a separate LM3916 drives its own row of LEDs continuously without multiplexing. Then 60 transistors are not needed for the same brightness.
 
I contemplated several lm3916

but it could get expensive.
it would take lots of transistors to even come close to the cost of 1 lm3916.
transistors = the oriential circuit says C1815 and ??
I attached a pic of filter w/ display drivers.
I changed the resistors on the lm3916 to 50 and put a 100 ohm trim pot. I have several different colored leds(red, green, yellow, blue) all take a different current and intensity levels so to even things out I put in trim pots. (10 for $1)
I got TINA simulation working right so I know the filters work as predected.
 

Attachments

  • filter section528.GIF
    filter section528.GIF
    60.7 KB · Views: 239
I can buy an LM3916 for $1.00CAN which is $1.03US. I would use 5 of them and eliminate the 60 transistors and eliminate all the multiplexing parts.

You are making mistake after mistake:
1) Your opamp has a 100k resistor at its output that feeds the virtual ground inverting input of another opamp. There is no signal voltage there for the audio output and for the negative feedback.

2) there is no current-limiting resistor for the LED anymore. You are feeding a huge base current into the left transistor and a pretty high base current into the right transistor and the LED will smoke and burn (or the power supply will instead).
There is nothing to turn off the left transistor.

3) The diode is missing from the bias of an opamp so its output voltage will be too low for the peak detector.
 

Attachments

  • Mr DEB again and again.PNG
    Mr DEB again and again.PNG
    13.6 KB · Views: 259
Where can I purchase

LM3916 for $1 each?
Need 10 of them.
in one of your posts you suggested removing the transistor from the Q200?
maybe I made a mistake.
The LM3916's would sure eliminate alot of problems.
 
I moved the buffer

In simulation this works?
 

Attachments

  • filter buffer.GIF
    filter buffer.GIF
    95 KB · Views: 227
  • filter buffer screen.GIF
    filter buffer screen.GIF
    99.9 KB · Views: 222
LM3914 vers the LM3916

Still working on design of a stereo 5band Eq unit w/ 2-5x10 displays.
Looking for an easyway out of using LOTS of transistors etc.
The LM3916 is $4+ each
The LM3914 can be had for $1.75 each
Can one use the LM3916 for a display of the 5 frequencies?
 
I buy LM3914 and LM3915 for $1.00CAN from a surplus store.
The LM3916 is rarely used because its steps are at wide and narrow spacings. The LM3915 has equal spacings for its steps (3dB).

Your opamp U2 has a voltage gain of 10 so it will overload when the EQ is turned up. Change R7 to 100k for a gain of 1.
 
Will change gain and LM3914?

Looking better all the time-may go with the LM3914
Thanks
 
logarithmic amp

I find out that the data sheet for the LM3914 recommends a logarithmic amp for the lm3914 input.
Been trying different configurations in TINA but it ruins the sweep.
any suggestions
 
The LM3914 is a linear voltmeter. The LM3915 is a logarithmic sound level meter.
When an LM3914 is used for sound level indication then its steps at the bottom (low levels) are huge and the steps at the top (high levels) are crammed together.
The steps for an LM3915 are equal at 3dB for each increase or decrease.
 
Unless I can get LM3916's cheap

and seeing how the display is for show maybe just go with the LM3914's.
What would it take or name of surplus store you know of for LM3916's cheap or 3915's
 
Getting somewhere w/ circuit

After moving parts around I found where to insert the peak detector/log amp circuit.
Am assuming the log amp will improve the performance of the lm3914?
I forgot to remove the transistor/switch combo. Shouldn't make a difference
 

Attachments

  • sweep1.GIF
    sweep1.GIF
    64.4 KB · Views: 1,618
  • sweep schmatic.GIF
    sweep schmatic.GIF
    85.5 KB · Views: 207
Hi Mr DEB,
Your log amp has nothing to zero it so its output might be at +10.8V most of the time.
Its output should be at about -0.5V without a signal and be at -0.63V with a very strong signal. The tiny output swing of only 0.13V will not be seen on an LM3914 display.

If it is amplified then the log circuit makes a pretty good thermometer since its output changes with temperature.

The log amp will need exra parts to make it temperature-stable and its output needs an inverting amplifier.
 
I ordered 10 - LM3916's.

of the hassel, problems etc in trying to get the right transistor bias, buffers, log amps etc.
lots less parts and pcboard landscape to design.
just need to generate a BOM and order.
thinking of changing the power supply from 12v to 9v OR purchase a larger voltage center tap transformer.
I was going with a 7812 and a 7912 voltage regulators in parrell but need a higher voltage (duh wasn't thinking straight when I purchased)
 
Status
Not open for further replies.

Latest threads

New Articles From Microcontroller Tips

Back
Top