latigid on

Troubleshooting midiphy SEQ v4+

115 posts in this topic

seqv4plus2.thumb.JPG.b6a5ecd1bf511aae5d7

seqv4plus1.thumb.JPG.5b770e2e39d1437da10

Here you have a photo (or two). In the upper one, the culprit is in the second Mathias row, the sixth from the right (third from left)

2 hours ago, latigid on said:

Is the cathode pin correctly soldered?

Yes, as the blue LED will work flawlessly.

2 hours ago, latigid on said:

Do the DINs work? 

Yes, all buttons and encoders work as they should, on all boards.

2 hours ago, latigid on said:

When you say you've isolated the LED, you removed it?

Yes, that's right. Outside of the circuit, all three colors will light up when testing. Inside the circuit, I cannot get the red LED to light up with the multimeter.

2 hours ago, latigid on said:

Did you try one of the spare parts? 

Not yet, because that possibility does not seem logical to me, as the LED works as it should when removed from the PCB.

2 hours ago, latigid on said:

You could try to temporarily short to the other red LED in the column or the adjacent cathode in a row.

Ok let's say I do this and it works, what does that tell me?

In order to being able to do my own thinking as well, the schematics would be really helpful. Are these available somewhere?

Thanks, ilmenator

 

Share this post


Link to post
Share on other sites

Thanks for the photos. From the looks of it, you didn't use that much solder on the LEDs. Did you try reflowing them?

What is the forward voltage of the LED? If it is too high, it could indicate that the soldering temperature was too hot. As everything else works, I don't see how it could be anything else (maybe a broken trace or pad shorted to 0V). Just swap the LED in this case.

Hence: manually connect the column  of the upper red LED to the lower one. If the LED lights, then there is a missing connection between RJ6_1 and the LED. Also test for continuity between the red anode pad and 0V (ground).

Relevant parts of the schematics attached.

 

lemec_blm.JPG

lemec_col5.JPG

lemec_RJ6.JPG

Share this post


Link to post
Share on other sites

Connecting the upper red LED to this one did not help - it was still not lighting up in the fluxtest.

Replacing the LED helped, BUT: after unsoldering it, outside of the circuit, all three colors of the presumably faulty LED light up, and with the expected brightness.

I'm dead sure this was not a soldering issue. Question: what was it then?

 

Share this post


Link to post
Share on other sites
1 hour ago, ilmenator said:

 Question: what was it then?

 

There is a number of possible reasons, from electrostatic discharge, shock (dropping) it, to that one not being as resistant to heat, a slight manufacturing diversion, or soldering it for one second to long - maybe the soldering iron was over the central part while soldering...

We had some issues with the red ones in the current batch, that's why we test-burn all of them for a few hours in LeMEC conditions (assuming you are using 47R) and only ship good ones, Andy has built a cool testbed PCB for it - which could also be used for MIDIbox RGB pixel graphics arrays later on, we will add it to the shop, when we have a bit of time.

So, to answer, either this one was bad undetected, developed the fault after shipping, or the handling caused it, it is impossible to say now -  that's why we put 5 of them for free into every bag of 35, so you have replacements.

Can you measure the red forward voltage and compare with the other ones you have in reserve? If my theory holds, its Vf should be significantly higher than on the others, that's why it still lights up when using your DMM in diode test mode and does not light up in the LeMEC.

Many greets,
Peter

Share this post


Link to post
Share on other sites

hi all,

don't know if this is the right place. i have finally finished the AOUT NG and DOUT board to add some cv/gate/trigger to my seq v4+.

I'm a little confused about the interconnections and jumper settings of the pcbs. inside of the sequencer i have the LINETX connected to the core like shown in the video but since the pcbs are a new revision and the J-headers are named differently then on the LINERX/TX-page i'am afraid of damaging something..

basicly i would like to do the following:

mbhp_line_driver_usecase_direct.png
 
- could anyone please help me on how to do the connections and setup the jumpers?
-is it necessary to use an additional power supply when using 0-5v only?
-what is the pinout of the AOUT port of the LINERX?
 
many thanks
rbv2
 

Share this post


Link to post
Share on other sites

Hi,

Actually, your usecase is more like this:

mbhp_line_driver_usecase_mbseq.png

As no serial inputs (switches) are used, the J2_SI setting is not important. If you use an AOUT_NG, then JAOUT offers 1:1 pinning, but it overlaps with the LINE_RX J19. The pinout is as shown in the ucapps schem.

If you've already soldered J19, you can rearrange the ribbon cable as shown:

fetch.php?media=seppoman:aout_ng_end.jpe

 

Thereafter use the top row (opposite side to the notch in the header) to connect to AOUT_NG J1. The other side is connected 1:1. These days I would put the black cable on the other way, but it doesn't really matter.

 

For midiphy LINE_RX, I recommend to use a dedicated +5V supply and leave J3 unjumpered. The ICs use quite a lot of current (100 ohm terminators), which isn't transmitted well down a DB-25 cable.

 

Any other questions please ask.

 

 

Share this post


Link to post
Share on other sites

thanks a lot for your answer.

I haven't understood the power supply yet.

where should i connect the +5V power supply for the LINERX module, and do i need another power supply for the AOUTNG module?

 

thank you!

Edited by rbv2

Share this post


Link to post
Share on other sites

At first, maybe try to keep J3 jumpered and see how the performance is.

You can think of the J8/9 SRIO chain (or J19 SPI) like a power buss, so if there is a power supply connected to the cable, the chips on the line driver will receive power that way. But if you think of where that power comes from it goes all the way back to the Core USB connector. Thus there is the option to unjumper J3 and supply the MIDIbox buss internally (like in a Eurorack case). This was not thought through extensively before, which is why I've designed something much more suitable for racking in a case, with lots of different power options (still needs testing but should be available soon).

One option for a separate +5V is to put the J19 connector somewhere in the middle of your ribbon. The first four wires are 0V|0V|+5V|+5V, so if you have a PSU handy you could connect those there. J3 should be unjumpered in this case. AOUT_NG would share the same +5V. This buss-power regime is not optimal, but it seems to work okay. Better would be to "star wire" each PCB back to the source PSU; that way the return currents should interact less.

Share this post


Link to post
Share on other sites

thanks a lot.

got it now :) I'm gonna take the power out of my eurorack.

 

Share this post


Link to post
Share on other sites

Hmm scratching my head a bit here - over the weekend I reconnected the USB pcb and core pcb and was able to upload the seq v4 project.hex file successfully.

I just finished the JA pcb and went to upload the ng_v1_036 project.hex file for testing, but MIOS studio doesn't seem to recognize my core at all. MIOS Studio does see MIDI IN/MIDI OUT named MIDIBox SEQ V4+ (and the other 3 seq channels), but no response from the core.

Troubleshooting steps so far:

  • Removed SD card PCB from core.
  • Inspected the pcbs for shorts but don't see any. I reflowed a few joints just to be sure.Inspected 10 pin cable connecting the USB pcb and core pcb but it looked good.
  • Tried changing the device ID in MIOS Studio
  • Tried unplugging RME audio interface to remove other MIDI devices
  • STM has the three bottom jumpers removed, power from 5v, and BOOT CONFIG set to FLASH, the other three jumpers are still in place.

I looked through the MIDI troubleshooting guide but it seemed to be more focused on PIC troubleshooting, if I missed something apologies!

Edit: Image upload failed so I've posted them on imgur: https://imgur.com/a/j61CNJs

Raw MIOS Studio output: 

MIDI IN:

Scanning for MIDI Inputs...
[1] UFX Midi Port 1
[2] UFX Midi Port 2
[3] MIDIbox SEQ V4+ (*)
[4] MIDIIN2 (MIDIbox SEQ V4+)
[5] MIDIIN3 (MIDIbox SEQ V4+)
[6] MIDIIN4 (MIDIbox SEQ V4+)
MIDI Monitor ready.

MIDI OUT:

Scanning for MIDI Outputs...
[1] Microsoft GS Wavetable Synth
[2] UFX Midi Port 1
[3] UFX Midi Port 2
[4] MIDIbox SEQ V4+ (*)
[5] MIDIOUT2 (MIDIbox SEQ V4+)
[6] MIDIOUT3 (MIDIbox SEQ V4+)
[7] MIDIOUT4 (MIDIbox SEQ V4+)
MIDI Monitor ready.
[1075782.103] f0 00 00 7e 32 00 00 01 f7
[1075782.104] f0 00 00 7e 40 00 0d 02 00 00 00 00 00 00 01 00 00 00 00 f7
[1075782.320] f0 00 00 7e 40 00 02 00 00 00 01 10 11 12 13 14 15 16 17 18 19 1a f7
[1075782.533] f0 00 00 7e 40 00 02 00 00 00 01 10 11 12 13 14 15 16 17 18 19 1a f7
[1075782.747] f0 00 00 7e 40 00 02 00 00 00 01 10 11 12 13 14 15 16 17 18 19 1a f7
[1075782.960] f0 00 00 7e 40 00 02 00 00 00 01 10 11 12 13 14 15 16 17 18 19 1a f7
[1075783.172] f0 00 00 7e 40 00 02 00 00 00 01 10 11 12 13 14 15 16 17 18 19 1a f7

Device Log:

No response from MIOS8 or MIOS32 core!
Check MIDI IN/OUT connections
and Device ID!
For debugging see also
Help->MIDI Troubleshooting

 

Edited by synaptech

Share this post


Link to post
Share on other sites

Have you tried restarting MIOS Studio between updates?

Jumper settings appear correct.

You can also measure resistance between the USB B port (with power off) and PA11/PA12 on the socketed Waveshare MCU breakout, should be 22R. 

Share this post


Link to post
Share on other sites

Just for documentation purposes, here are synaptech's images:

U12d6JP.thumb.jpg.090f64a8e6a3c9767c8a6d

OGqJfm6.thumb.jpg.0b96227c52138dd79ecd91NMUOq7o.thumb.jpg.3f97f77be2f31f42170337MsWGH5C.thumb.jpg.dc75203197aa2a81a460d3lQiHIDV.thumb.jpg.b83a0a22318ace2760b7da1o7vt5l.thumb.png.270588a8548d6eb199d4f5

SwlmJ8k.jpg

Share this post


Link to post
Share on other sites
3 hours ago, latigid on said:

You can also measure resistance between the USB B port (with power off) and PA11/PA12 on the socketed Waveshare MCU breakout, should be 22R. 

This is not a Physical USB issue cause the bootloader seems to work otherwise the In/Out "MIDIbox SEQ V4+" port should not appear.

@synaptech
Disconnect all others USB, restart your  PC.
Put a jumper on JPA0,
Connect your Core and launch MIOS Studio,
and tell us what happens, please...

Best
Bruno

Share this post


Link to post
Share on other sites
6 hours ago, latigid on said:

Have you tried restarting MIOS Studio between updates?

Jumper settings appear correct.

You can also measure resistance between the USB B port (with power off) and PA11/PA12 on the socketed Waveshare MCU breakout, should be 22R. 

Wow - kicking myself here. Simply restarting MIOS Studio brought it back to life! I was able to get the core showing again and uploaded the MIDIBox NG app as well, though I did need to restart MIOS Studio again for it to show (running on Windows 10).

Appreciate the super fast responses!

Thanks for helping with the pictures ilmenator, not sure why I kept receiving -200 errors during file upload.

Share this post


Link to post
Share on other sites

Congrats on getting it working again! As you heard my grumbling passage in the video tutorial "i should really get a mac" - really meant that at that time, as Windows seems to have quite buggy MIDI-USB. Restarting MIOS Studio or even Windows can help quite often.

Many greets and have fun building,
Peter

Share this post


Link to post
Share on other sites
13 minutes ago, Hawkeye said:

As you heard my grumbling passage in the video tutorial "i should really get a mac"

I use both OS, It's less under mac but it also happens ;)

 

Share this post


Link to post
Share on other sites
1 hour ago, Hawkeye said:

Congrats on getting it working again! As you heard my grumbling passage in the video tutorial "i should really get a mac" - really meant that at that time, as Windows seems to have quite buggy MIDI-USB. Restarting MIOS Studio or even Windows can help quite often.

Many greets and have fun building,
Peter

Hah yes it's too true, I think I'd been skipping around the build videos so much I'd missed that you had restarted MIOS Studio to get the board showing up. I do actually have an macbook pro for work but it's a new one with the damn USB-C ports and I have yet to get an adapter. :rolleyes:

I got the JA board up and running this morning, all the buttons and encoder seem functional!

The LeMEC boards are a different story:

On the left LeMEC board it seems like all the buttons (encoder press and matias switches) don't register, turning encoders left and right behaves correctly.

On the right LeMEC board all of the buttons seem to work as expected, turning encoders works on all encoders except on encoder 16.

I haven't had a chance to dig into troubleshooting as I had to run to work, but if anyone has a guess as to why all the buttons aren't working on LeMEC_L I can focus my hunt at lunch/this evening. I'll update with pictures as soon as I'm able. Thanks again!

Share this post


Link to post
Share on other sites
38 minutes ago, synaptech said:

I got the JA board up and running this morning, all the buttons and encoder seem functional!

Very good!

 

38 minutes ago, synaptech said:

The LeMEC boards are a different story:

On the left LeMEC board it seems like all the buttons (encoder press and matias switches) don't register, turning encoders left and right behaves correctly.

At a guess, R26 is missing, or IC6 is installed without R17/C7. Presumably you haven't tested LED functions yet? 

I can provide more help when I see a picture of the board.

 

38 minutes ago, synaptech said:

On the right LeMEC board all of the buttons seem to work as expected, turning encoders works on all encoders except on encoder 16.

Do the encoder pins contact the pin header below? Try trimming the pins, also isolate with a bit of tape.

 

 

 

Share this post


Link to post
Share on other sites
2 hours ago, Hawkeye said:

 as Windows seems to have quite buggy MIDI-USB. 

Hi !

I dont say that as a excuse for Windows but : we have to keep in mind that before and after a firmware update, it is not the same USB device anymore. And MIOS don't let the old one go, and don't show the new one. But well, I think a patch to correct this specific usecase will never happen.

Share this post


Link to post
Share on other sites
2 hours ago, latigid on said:

At a guess, R26 is missing, or IC6 is installed without R17/C7. Presumably you haven't tested LED functions yet? 

I can provide more help when I see a picture of the board.

 

Do the encoder pins contact the pin header below? Try trimming the pins, also isolate with a bit of tape.

 

 

 

R26 is there, IC6 isn't installed so I think R17/C7 are ok being missing? I haven't tested the LED functions yet but have done a test with my multimeter to make sure all the superflux LEDs are still functional.

 

2 hours ago, latigid on said:

Do the encoder pins contact the pin header below? Try trimming the pins, also isolate with a bit of tape.

I inspected the two LeMEC boards and can't spot any shorts between the pins, I did trim the pins on the outside edges to avoid shorting against the header.

I also tried swapping the two SEQ_PLATE pcbs to see if it was a problem with the plate board but the behavior was the same so we should be able to rule those out.

Still having errors uploading directly to the forum, so they're uploaded to an imgur album again: https://imgur.com/a/naVxaRL

Thanks for the help!

Edited by synaptech

Share this post


Link to post
Share on other sites

Soldering looks clean and it's correct to omit IC6 and R17/C7.

Software-wise, how are you testing? Are the boards chained left to right or did you test each one separately? Using the _L configuration will not work if the first board is _R as the matrix is offset by three SRs. Hardware ideas: the transistor types are wrong or swapped, mixed up with a BAT diode etc.

For the missing encoder, on lemec with the plate removed, check J3, 8th pin from the left, to pin 5 on IC5, 9th pin to pin 6, respectively. The pins should have +5V on them. If you (carefully) short the pins to 0V, do the encoder/DIN events trigger? If not it could be a cold solder joint or a short to 0V (e.g. IC5, pin 8).

Now connect the plate PCB. You should have continuity back to IC5. If not, check that the pin header makes good contact with the through-board header.

 

Share this post


Link to post
Share on other sites
Just now, latigid on said:

Soldering looks clean and it's correct to omit IC6 and R17/C7.

Software-wise, how are you testing? Are the boards chained left to right or did you test each one separately? Using the _L configuration will not work if the first board is _R as the matrix is offset by three SRs. Hardware ideas: the transistor types are wrong or swapped, mixed up with a BAT diode etc.

For the missing encoder, on lemec with the plate removed, check J3, 8th pin from the left, to pin 5 on IC5, 9th pin to pin 6, respectively. The pins should have +5V on them. If you (carefully) short the pins to 0V, do the encoder/DIN events trigger? If not it could be a cold solder joint or a short to 0V (e.g. IC5, pin 8).

Now connect the plate PCB. You should have continuity back to IC5. If not, check that the pin header makes good contact with the through-board header.

 

Back at work so I'll have to follow up on the testing this evening - but for testing I had everything chained JA->lemec_L->lemec_R. Thanks for the quick responses!

Share this post


Link to post
Share on other sites

@synaptech one thing that happened once or twice at my side (built 6 lemec boards in total), was that the SMT IC soldering looks good as seen from above, but there was a little gap between IC leg and PCB, which can only seen from the side - no contact on a pin. This can happen when you "push on the IC", soldering a single leg first and the IC is not completely flat on the PCB.

Therefore, i'd recommend to use a magnifying glass and check the 595s and 165s pin for pin, especially looking at them from the side.

Looking at your pictures, on the LeMEC L, the upper right pins on IC2 might have suffered from such a fate - it should be easily corrected by re-soldering them with a fine tip and a tiny bit of solder.

Many greets,
Peter

Share this post


Link to post
Share on other sites

Good news, got the encoder working! I had a cold joint on IC5, the pad looked wet but the leg of the IC hadn't actually gotten any solder on it.

Still stuck on the buttons on LeMEC L, the transistors are all the same type from what I can see (marked with 6G) and the BAT54s (marked with 5A). Is there a specific IC that handles the button presses for the whole LeMEC board? It's odd that it all works on the LeMEC R but none of the buttons function on the LeMEC L.

Share this post


Link to post
Share on other sites

@synaptech - that's good news with the encoder! :cheers:

As written above, i'd recommend to thoroughly check every SMT IC pin on LeMEC L - i had more than once problems with solder connections i thought were ok. You can "reflow" them quite easily. Please check the top pins of IC2 on LeMEC L, can't really see it in the photo, but there might be some solder missing, or they might not "reach" the PCB - check from the side with a magnifying glass.

Good luck!
Peter

Share this post


Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!


Register a new account

Sign in

Already have an account? Sign in here.


Sign In Now