SEQ v4+ Troubleshooting

dcam

So it turned turned out my core wasn’t pushed in tightly enough into the headers. I’m now getting a power signal, but I’m not getting a USB signal. It’s set to Flash memory, 5vin and the switch on the usb board is in the up position. (I’ve tried down too).

UPDATE: I got it working by unplugging the core cleaning one of the headers, and really firmly reinserting it. Feeling a little sheepish. 🙂

latigid on

Nice to hear! Often just “wiping” the contacts is good for data signals. Now onto the rest of the build 😀

dcam

I ran load seq_l and load seq_r in Midibox NG V1.036.

Seq_l results in three lights turning on and seq_r results in all of them. Is this correct? The switches don’t register any activity in debug mode, but the encoder does. According to the video I thought the switches turned the lights on and off.

latigid on

If you can get DIN events working, the issue is probably to do with the 595 shift registers?
SEQ_R is if your JA board comes after the lesMECs.
Feel free to upload clear photos and I’ll take a look.

dcam

Do you mean for the RH version?

latigid on

Thanks for the photos!

Even if you have an RH case, you should still test with SEQ_L, at least until the le mec boards are built as the registers are mapped accordingly.

From your photos, many of the joints look like “cold blobs”. You should try retouching these pins to to get the solder to wet properly to the pads. I recommend using some flux to help here.

Potentially there is a short between pins ⅞ of IC3?

dcam

I reflowed two 595’s and then the jumpers on the core and JA board and now I’m getting midi triggers for each switch. Hizzah!

When I start seq_l now, the same three leds still light up. (Record and two others) The record led changes intensity when I activate the switch but the other two white leds don’t. Are these reversed polarity or is there some other explanation?

latigid on

A common issue here is that one or more of the LEDs are broken or soldered in with the wrong polarity. Check each one with the multimeter diode tester. On the MEC switches, the encircled pin is the anode and the flat marking indicates the cathode. Don’t PANIC (Positive Anode, Negative Is Cathode).

dcam

latigid on

All the LED’s appear to be the correct polarity when measured with voltage. Only one of them lights up though. (The Red one)

The weird thing is, any positive point and any negative point in that row will light up the red led. The red led appears to have a lower voltage diode reading than the white leds, but so do the replacements.

latigid on

When LEDs are wired in parallel, the current follows the path of least resistance, which is through the red LED.

Do any LEDs correctly light up when their buttons are pressed (using the script)? If not, then the issue could be cabling or soldering on the Core board (IC 1B and J8/9).

dcam

I lied, two of the LED’s that were on in seq_l had reverse polarity. Reversing them turned off the record light. I also reversed one diode on the LeMEC board! Oops. All switches, encoders and LED’s are tested and ready to go. Onto the Matias switches!

dcam

Why are there two holes for the flux led A + B connections on the LE-MEC LH PCB board? I’m assuming I should use the hole the line points to.

latigid on

It’s so you could combine a few colours if you wanted. But use any RJ pads that work, just use the outside ones for each resistor.

dcam

I’m on to testing the superflux! On the LeMEC RH, columns 3 and 7 don’t display green on channel B. (SW3, SW11, SW7, SW15). They display red just fine, and the switch sends information. When in debug mode and clicking on the encoder, they light up red. They turn green when I trigger the diode with a multimeter, so the led is fine. I’ve checked the connections on pin 2 for each superflux, the diodes, and resistors in that column. I’m at a loss here though. Is it an IC pin that I need to reflow?

Thanks for all of your help!

latigid on

This could be T14 on the top of the board, R23 or IC4 pin 2. Or RJ3 might need checking?

dcam

As far as I can tell, R23, IC4 p2 and RJ3 are fine. I’m not sure how I can inspect T14 without taking everything apart? Would I have access if I removed SW7? Is there a trick to remove the switches?

latigid on

Of course if you have already soldered in the switches, it is much more difficult! You only have 3mm clearance between PCBs, so maybe you need to take all of them off. But you could try with SW7. If you heat up both pins at once it should hopefully work.

dcam

Maybe the Youtube video should be amended to run the fluxtest script before soldering in all the switches?

I don’t know what to do. Taking it apart doesn’t seem feasible…each switch is also physically locked in place in addition to being soldered isn’t it?

It seems I have 4 options:

Try to take it apart…seems very risky i’ll probably damage the superflux with the amount of heat required to remove the solder and probably damage components trying to remove the Matias switches.
Build a new board
Complete the build with 4 switches that don’t light up
Move on…

latigid on

dcam Maybe the Youtube video should be amended to run the fluxtest script before soldering in all the switches?

In the video, the MEC row is tested before soldering in the switches, so you might have noticed the issue there?

I don’t know what to do. Taking it apart doesn’t seem feasible…each switch is also physically locked in place in addition to being soldered isn’t it?

It is “only” held in by the friction of the PCB and the solder pins.

It seems I have 4 options:
Try to take it apart…seems very risky i’ll probably damage the superflux with the amount of heat required to remove the solder and probably damage components trying to remove the Matias switches.
Build a new board
Complete the build with 4 switches that don’t light up
Move on…

Don’t worry! Plenty of people have done something similar in the past. Desoldering can be done, even better if you have a desoldering pump or two soldering irons. It is also possible to “unclip” the sides of the switch, which will practically destroy it, but it works as you can separate the pins easily. We have switches available and you should even have some spare as part of your kit.

You could first check if the signal reaches the pins mentioned with the following (you should also be able to toggle the column with a switch).

Use MB_NG
set debug on
set dout 41 1
(or set dout 41 0 to turn “off” the transistor)

You can also toggle other rows on with the same switch to test, I think 41 is the right number (8 OUTs per chip, starting at 0).

dcam

latigid on In the video, the MEC row is tested before soldering in the switches, so you might have noticed the issue there?

All the switches worked and sent signals, so it passed the MEC row testing, the issue is that the path Superflux 2 (Green) to LightState B doesn’t work.

« Previous Page Next Page »