Mesmerizer - LED flickering but no modulation of sound

tdukes

Active member
I replaced my XR2206 with the ones from Jameco. Now the LED seems to flicker, but is very bright. I don't notice a modulation of the sound. The image of the output from the oscilloscope is stable too and does not seem to be modulating. Any suggestions welcome.

I have attached photos of the input, the output, and pin 2 of the XR2206.
mesin.jpg mesout.jpg mesxr2206pin2.jpg
 
I gave up on the 919. Got myself a diy signal generator that uses the same chip. I used the chip from the dit and slapped it on the mesmerizer. No oscillation. Built the diy project. Stuck signal generator out to pin 2 and it sorta worked like a tremolo! A tremolo! Experimented on triangle, square and sine wave. Now i’m thinking of slapping everything in an enclosure. Btw LED still dead. But sound is tremolo-ing or vibrato-ing. A0DD2CF3-7D9B-4EA1-97D5-80573CEAF6C1.jpeg
 
So it looks like you are also getting 2V peak to peak with about a 4.5 DC bias for a 1/250ms = 4Hz modulation at pin 2. And in the picture you are getting modulated sound? But you are using the triangle wave. Does it also "vibrato" on sine?
 
have you tried putting the 919 chip into the signal generator to see if it works? now you have a way to test that chip!
It doesn’t work on the signal generator. It came with HT1918 batch no.(?).

I also replaced the 2n5457s with J201s. Emitter and Collector switched. Works fine now. A little harsh at full depth but I can find some good sounds in this.
 
Where did you get your JFETs? Did you test them beforehand?
The 2n5485 are packaged like they came from tayda. I have 2 more 2n5457 that came from who knows where, probably eBay.

How would you test these? I have a multimeter that measures hFE, but these are JFETs right?

I have some SMD to through hole adapters, but it looks like the only SMD JFETs I have are CMPF4393 I got for the Bohemia I haven't built.
 
The 2n5485 are packaged like they came from tayda. I have 2 more 2n5457 that came from who knows where, probably eBay.

How would you test these? I have a multimeter that measures hFE, but these are JFETs right?

I have some SMD to through hole adapters, but it looks like the only SMD JFETs I have are CMPF4393 I got for the Bohemia I haven't built.

This thread has a good amount of info https://forum.pedalpcb.com/threads/jfets-matching-etc.1139/
 
How would you test these? I have a multimeter that measures hFE, but these are JFETs right?

You can test the FETs with a multimeter. In this circuit, the important parameters are Vp and Rdson.

First, we'll measure RDSon. Short the S & G leads together. Measure resistance from S to D. Polarity doesn't matter for this measurement. Set your DMM for the 2KOhm scale. anything under 200 Ohms should be fine.

To measure Vp, connect the + lead of your meter to the S lead. Connect the - lead of your meter to the G lead. Set the meter for the 2V scale. Connect the - side of a 9V battery or power source to the G lead. Connect the + side of a 9V battery or power source to the D lead. The meter will read Vp. If the meter reads overrange, switch to the 20V scale. I think Vp needs to be <1V.
 
I swapped out the 2n5457s with another pair of 2n5457 and got pretty much the same results. No modulation of sound.

I then swapped both for a pair or 2N3819 and get modulated sound.

I am not seeing much Red from the LED, but the green is very bright. I don't know where I got these from but they are labeled:
Color: Red&Green Common Cathode
Rated Voltage:
Red: DC 2.0-2.2V
Green: DC 3.0-3.2V

Since the circuit look symmetrical and the green is very bright. It is possible the red got burned out? How sensitive is the modulation to the 1K registors R16 and R17? I am thinking I can up those without affecting the modulating voltage noticeably. Does anyone disagree.

Thanks!
 
The advertised "rated voltage" for LEDs is pretty misleading. LED voltage depends on LED current and we tend to run the LED current pretty low on guitar pedals. For example, the maximum rated current for a typical LED is 30mA, but we don't run it anywhere near that current. Most pedals run the LEDs below 5mA; the ones on this site typically run them at 1.5mA. At those currents, red LEDs have a voltage drop of about 1.6V. Green LEDs are more like 2.7V. I listed some typical LED voltages in this post. The human eye is more sensitive to green light than red, so you need more current in a red LED to get the same apparent brightness. If an LED is stressed by too much current, it will either reduce it's efficiency or burn out.

You can reduce the values of R16 and R17 to get more brightness. I'd recommend not going below 330 Ohms. But first, make sure your LED is working ok. Use a 9V battery and a 1K resistor to test it. Be careful to get the polarity right. Hooking an LED up backwards can kill it. Don't ask me how I know that.
 
You can reduce the values of R16 and R17 to get more brightness. I'd recommend not going below 330 Ohms. But first, make sure your LED is working ok. Use a 9V battery and a 1K resistor to test it. Be careful to get the polarity right. Hooking an LED up backwards can kill it. Don't ask me how I know that.

Actually I was thinking of increasing R16 and R17 to get less brightness on the green. I think I burned a hole in my retina looking at it. But I think it would be a good idea to breadboard some different resistor values to see what looks good with the LEDs I have. I think the voltage swing I am seeing is 3.5 to 5.5 volts. One segment is at the peak when the other is in its valley.
 
According to the schematic, R17 controls the green LED current. Since the LED can be inserted either way and still work, the chances are 50/50 that R17 is the one you want to change. You could try removing R17 and see if only the red side of the LED is flashing. Then you'd know for sure.
 
I did some tests with my LEDs to see why I am not seeing much color change. I think unless I decide to get new LEDs I am just going to change the 1k on the green to 47k, leave the red at 1k, and call it fixed.

I wired according to attached schematic. I then adjusted the wiper so that I got pictures with (red@5.5V, green@3.5V), (red@5.5V, green@5.5V), (red@0, green@5.5), (red@5.5, green@0) The voltage was measured from the series resistor to ground, not just across the LED segment.

I used the voltages 3.5 and 5.5 because that is what I was seeing in my mesmerizer circuit.

The color change was more apparent in person than in the pictures to me. But not by much:
led_schematic.jpg

red0green5_5.jpg red3_5green5_5.jpg red5_5green0.jpg red5_5green3_5.jpg
 
Back
Top