Troubleshooting midiphy SEQ v4+

[mcmurray]

I got both OLEDs working!

It turns out one of the cables was bad, which just goes to show how careful you need to be when making them. They looked ok, but I've been building them with vice grips which isn't optimal. I need to get a bench clamp like Hawkeye uses.

I will put this thing together now, thank you latagid and lp1977 for your help! Hope everyone is staying safe :)

[latigid on]

Awesome to hear! Most of the time it's the cables ;-)

For seq_r.ngc testing, it's possible that .NGC file is buggy. I was baffled as to why the hardware would be fine in one config but not the other. It's something to check.

For now, it is of course fine just to test with the _l config file and swap boards around for the finale.

Sorry about the bugs and hope to see the SEQ running soon!

[latigid on]

If you can't get a vice, this is also a nice sort of tool:

https://www.dicksmith.com.au/da/buy/bourne-electronics-idc-plug-crimper-crimp-tool-t1540/

Vice is nicer though, as once the cable is in and the connector is in place, you can control the angle of the cable to avoid shorts and overlaps.

[mcmurray]

It's working, but I will pull it apart one more time (not until I've played with it for a while first!).

MIDI OUT1 port isn't working, probably the connector for that board needs re-seating. And some of the pushbuttons are not properly aligned. Both issues are my fault though.

[mcmurray]

Probably a stupid question BUT is it possible to remove the bottom of the case only, to get at the core and ribbon cables but leave the front panel interface intact?

 

Doesn't look like it but I thought I'd ask before I pull it apart completely.

Edited April 9, 2020 by mcmurray

[latigid on]

Sorry, the bottom of the case is welded by the corners.

Btw, the seq_r.NGC file was completely the wrong thing. https://github.com/midibox/mios32/pull/8  but still needs to be approvied by TK.

If you post a picture of the finished unit you get your serial number!

[momelq]

This is not directly a hardware issue, but I noticed that on my V4+ several buttons do not have a function assigned.

For example the push function of the dial buttons do nothing. 

Or the lower Matthias buttons could do something useful in these contexts:

• Instr -> set midi Ch
• mute -> solo

Can I define these things in MBSEQ_HW.V4 and did somebody try to define these buttons?

[latigid on]

You could add feature requests to

http://midibox.org/forums/topic/13137-midibox-seq-v4-release-feedback/?page=61

I'm not sure if MIDI channel makes sense because it's part of the track events/setup. But you can ask!
I think mute/solo was on the list for a while.

For button assignments, you can freely choose any function you see in the HWCFG file. I don't know if you can reassign the encoder presses, or if that's generally a good idea vs. leaving them reserved as accelerators.

[manu]

Dear Peter, Andy, Adrian and Thorsten and all others involved in UCApps and Midiphy, 

thank you for this amazing project. 

I am almost done with the SeqV4+ and it was a great experience. I enjoyed the two instructional videos from Peter more than any TV show [great music by the way], the hardware design is fantastic and I can’t wait until I finally can use it in combination with the legendary software from T.K.. 
Going through the building process made me very happy, so thank you for that. 

Unfortunately I have still two problems before officially asking for my serial number ; )
My knowledge of electronics is quite limited, so I am very thankful for any help. 

I searched through this topic and related. So I hope, that I don't post problem descriptions, that already existing close by. 

1. Left OLED display is not working properly

I probably broke the display itself while soldering, but before spending the money on a new display, I wondered if anyone could spot an obvious mistake. 

While powering SeqV4 the content is displayed fine

  but then while running, the right half of the display stays completely "filled".

The left half is otherwise working as expected.

I checked: 

1) On the core on J15_S the displays are jumpered to 3.3V and R33D on the core is 560R as suggested by Peter. 

2) I swapped the display cables on the core (J15A/B). The defective display behaviour is unchanged.

3) I swapped the cables from the working display and defective display. The defective display behaviour is unchanged.

4) I checked the shielding against the capacitor as suggested by CJ55. It was already nicely shielded on the OLED PCB with a strip of extra tape. Thanks for that : )

5) I double checked the soldering points on the OLED connector.

I probably damaged the OLED while soldering, right? The only thing that makes me doubt is, that the display seems to work while powering up SeqV4. 

2. LED matrix does not work anymore

After I finished the Le-Mec boards I tested the LED Matrix through Midibox NG and CC16 and it worked fine. Now that I have assembled the PCBs into the case, the LED Matrix unfortunately does not work anymore. Neither with SEQV4 nor Midibox NG - no life sign. When I test the LED Matrix with a Multimeter as suggested by Peter in the video, on the top right pin and the 4th from the left pin, the corresponding LED lights up. What could be the next meaningful component to test?

Thank you for your time. 

Best Regards, 
Manu

Edited April 19, 2020 by manu

[latigid on]

Hi Manu,

Great that you're nearly done! Thanks a lot for looking through the other troubleshooting ideas first.

For this one:

Just now, manu said:

3) I swapped the cables from the working display and defective display. The defective display behaviour is unchanged.

 

Does it mean that you removed the cables from the OLED end, and the known-good J15 port and cable also showed the same behaviour with the "right-hand" OLED? Normally I would expect that the cable had a missing connector or shorted cable. It might be worth building a brand-new cable if you have the connectors and ribbon handy. You can also check the Core module IC2 and resolder the pins.

Do you notice any different behaviour if you just connect one OLED at a time? Also, how are you powering up? You might try a phone charger USB or hub with more power available.

 

Just now, manu said:

2. LED matrix does not work anymore

 

Maybe a cable got loose? There was one IDC8 connector that didn't fit well. Otherwise, I'm not sure what to suggest if everything else is running. Did you check with the correct _L/_R NGC and the correct SEQ HWCFG?

Best,
Andy

 

 

[manu]

Hi Andy, 

thank you for your fast response.

 

19 minutes ago, latigid on said:

1. Left OLED display is not working properly

Does it mean that you removed the cables from the OLED end, and the known-good J15 port and cable also showed the same behaviour with the "right-hand" OLED? Normally I would expect that the cable had a missing connector or shorted cable. It might be worth building a brand-new cable if you have the connectors and ribbon handy. You can also check the Core module IC2 and resolder the pins.

Do you notice any different behaviour if you just connect one OLED at a time? Also, how are you powering up? You might try a phone charger USB or hub with more power available.

Thanks for the suggestions. So I tried to connect one OLED at a time and also to power the unit through a phone USB charger. Unfortunately no difference. The faulty physical display always had the same defective behaviour, regardless of all the described swapping. So probably I damaged that one display.

19 minutes ago, latigid on said:

2. LED matrix does not work anymore

Maybe a cable got loose? There was one IDC8 connector that didn't fit well. Otherwise, I'm not sure what to suggest if everything else is running. Did you check with the correct _L/_R NGC and the correct SEQ HWCFG?

Yes, that was my hope as well. I tried already 3 times to follow the steps again, from Peters video for testing the LED Matrix, but no success. Based on your suggestion on the previous page, I will try to set the pins directly to hopefully get some kind of a life sign. 

 

Thank you for your time. 

[manu]

I am very happy, that I was able to fix both issues. 

Quote

1. Left OLED display is not working properly

A new display fixed the issue. An expensive soldering mistake, though I only soldered with 280 degrees Celsius but probably I touched the pin for too long..? 

Quote

2. LED matrix does not work anymore

Maybe a cable got loose? [...]

 @latigid on was right! It was the cable, I ordered parts for a new one and it worked. Thank you so much @latigid on and @Hawkeye for your support. I will now assemble the unit and request a serial number within the next days. Jippppiiieee! 

[latigid on]

Awesome work, thanks for the update! Your serial number is waiting :)

[Hawkeye]

Cheers, @manu, glad that it is sorted out now, hope you enjoy your new sequencer!
We're already working hard on the ultimate user interface extension for it! :-)
Best regards,
Peter

[manu]

Yes, I'm totally in love with my new SEQ V4+. Amazing work from everyone involved .

I'm very much looking forward to learning more details about the new extension you are working on. The SEQ V4+ Eurorack Modules are already on my list : )   

Edited April 30, 2020 by manu

[mcmurray]

On 09/04/2020 at 2:56 PM, latigid on said:

Sorry, the bottom of the case is welded by the corners.

Btw, the seq_r.NGC file was completely the wrong thing. https://github.com/midibox/mios32/pull/8  but still needs to be approvied by TK.

If you post a picture of the finished unit you get your serial number!

Apologies for the late reply (been busy af with work) but here's mine for a serial number - I hope this is the right place to get one, if not let me know where I should post it. This was such an enjoyable build I'm considering a Loopa :)

Edited May 12, 2020 by mcmurray

[latigid on]

Green and gold!

We're keeping track here:
http://midibox.org/forums/topic/20885-midiphy-seq-v4/?page=23

 

[slo]

I'm trying to understand why my SeqV4+ OTG does not work. There is no activity on USB1 in the midi monitor
With the switch is in the up position I can flash the core.  With the controller plugged into USB A port and the switch down, no midi activity
The jumper (J8) is in place and my controller is powered
The seq settings are correct and if I plug the controller into the midi port IN1 it's working and recording from the controller

I'm powering the Seq from a powered USB hub connected to the USB B port and a computer and interestingly the midi data is being passed from the USB A port (controller) to USB B (hub) and you can view the controller midi data in MIOS Studio.

I have tried 3 different controllers, no midi activity on USB.

[latigid on]

Hi George,

I'm not 100% on how it all works and could be wrong, but to my knowledge you shouldn't see any MIDI data on the computer from the slaved device plugged into USBA. So my first guess is that you should unplug the computer; you can't (or shouldn't be able to) have the SEQ as both host and slave at the same time. You could then check the SEQ MIDI monitor and route it out a different DIN port.

Second thought is that the controller (which ones, so we can consider making a whitelist?) does not initialise properly.

Best,
Andy

[slo]

Hi Andy
Here's the controllers I'm testing with. Oxygen 8 V2, Samson Graphite M25, Keystation 61es. The first 2 worked in my V4 setup, I installed a switch that put the core into host mode by grounding the sense pin on the core micro USB, and I could flash the core with updates switching from host to slave through the same USB port.
Seeing and selecting a controller in MIOS could be normal. Things are still the same with the computer out of the picture though, the problem remains that the V4+ does not go into host mode.

[latigid on]

The v4+ does nothing different to your previous method: the ID pin is grounded when in OTG mode and floating otherwise. Maybe check pin PA10 is correctly grounded on the wCore PCB/breakout? Cabling issue or missing contact through the 2x25 pinheader? Perhaps the value for R3 on the USB board is too conservative and 100R would be better.

 

[slo]

Ok Thanks Andy, I'll dig into it later this week and report back

[latigid on]

One other idea: you could try plugging directly into the 407v board (mini USB cable). Here is the schematic, maybe some of the jumpers are helpful?
https://www.waveshare.com/w/upload/5/58/CorexxxV-Schematic.pdf

 

[Antichambre]

Id pin(PA10) is floating externally but pulled up internally(cpu), connected to ground externally for host mode. Core must be restarted on each Dev/Host switching. There's no jumper or other option on PA10 it's directly connected to CPU pin.
So check continuity between R3(Dev/Host switch) and the Core, this line should be broken you just have to find where... R3 is not really necessary and can be replaced by a piece of wire.

 

Edited May 18, 2020 by Antichambre

[slo]

Fixed my OTG problem by using a "charge only" USB cable to power the V4+ from a USB hub. I can only speculate that computers, hubs etc. have some transmissions going on that corrupts the OTG operations.