Jump to content

Troubleshooting midiphy SEQ v4+


latigid on
 Share

Recommended Posts

Hmm, 2000 posts :angelnot::angelnot:

 

36 minutes ago, knochenfabrik said:

because i removed the pads (rings) on 3 pins of the resistor network (Pin 1 (dot) ,5 ,7)

How is pin 1 bodged? Do you measure +5V to it? I would bridge to the empty J89 socket, pins 3/4. You can see the wider trace to it in your image.

Link to comment
Share on other sites

On 6/23/2019 at 10:41 PM, latigid on said:

Hmm, 2000 posts :angelnot::angelnot:

 

How is pin 1 bodged? Do you measure +5V to it? I would bridge to the empty J89 socket, pins 3/4. You can see the wider trace to it in your image.

Hello there, I know this may not be according to the DIY spirit but I rebuild the RH Lemec from scratch and got it working :-). Will still try to bring the other board to life with your tips and will report on that so at least other people could profit.

I´ve already did all the assembly and tested everything in midibox NG. Also the Final Software seems to work fine. But I got one question regarding the LED-Matrix: I only see LED-Matrix working in a mode where I see the Volume of the Triggered Steps (like the peakmeter on a mixer) but I cant seem to got the other mode workingm where you can see steps relatively to their position (as shown in the build video). Do I just miss some configuration or is this maybe a hardware problem?

Thank you again for your help!

 

EDIT: Congrats to 2000 posts!!! Have a beer for celebration :D!

seqfertig.jpg

Edited by knochenfabrik
Link to comment
Share on other sites

Cheers! :cheers:

Don't worry about the other board, the main thing is the SEQ lives! Well done!

You can change the so-called TPD mode by navigating through Utility > Options #11. You can also test the hardware with SEQ_L.NG again; the CC 16 slider should light up the JA matrix. 

  • Like 1
Link to comment
Share on other sites

  • 2 weeks later...

Hey guys,

after succesfully building two Seq V4, I could not resist the V4+ version. In this context, I just populated the JA PCB, however, it unfortunately is not working as expected.

My problem is the following:

Using the NG application with SEQ_L.NG, I can trigger all LEDs individually by sending NoteOn commands from MIOS Studio. Yet, pressing any switch causes an irradic behaviour with several NoteOn being send (Ch3) and consequently several LEDs lighting up. Similarly, touching any solder point by hand elicits a similar random NoteOn and flickering response. Touching the SrOut connector fires multiple NotOn on Midi Ch1. Furthermore, sending CC16 from MIOS Studio does not light any led on the 16x8 matrix . In fact, during all my tests, I have not seen any of them lit :(

Naturally my first suspect for the multiple NotOn messages were the 100n cap of the 74HC165, the IC itself, or the resistor network. I do not know how often I have checked and/or re-done the corresponding solder joint. Nothing has helped so far. I further checked the orientation of the 165 and the resistor network, as well as their part number based on the mouser BOM.

For the 16x8 matrix, my tests were rather limited. I could not do a diode check before soldering, but the letters on the matrix side point to the bottom, and the parts are from Midiphy.

Since I am currently running out of ideas, any help would be appreciated.

Link to comment
Share on other sites

2 minutes ago, Karg said:

Using the NG application with SEQ_L.NG, I can trigger all LEDs individually by sending NoteOn commands from MIOS Studio.

suggests the 74HC595 parts are correctly soldered.

 

2 minutes ago, Karg said:

Yet, pressing any switch causes an irradic behaviour with several NoteOn being send (Ch3) and consequently several LEDs lighting up. Similarly, touching any solder point by hand elicits a similar random NoteOn and flickering response. Touching the SrOut connector fires multiple NotOn on Midi Ch1.

suggests floating digital inputs or similar.

 

2 minutes ago, Karg said:

Furthermore, sending CC16 from MIOS Studio does not light any led on the 16x8 matrix . In fact, during all my tests, I have not seen any of them lit :(

2 minutes ago, Karg said:

For the 16x8 matrix, my tests were rather limited. I could not do a diode check before soldering, but the letters on the matrix side point to the bottom, and the parts are from Midiphy.

Since I am currently running out of ideas, any help would be appreciated.

Are you testing the JA board by itself or hooked up to lemec_R? The shift registers are not located on the JA board.

 

2 minutes ago, Karg said:

Naturally my first suspect for the multiple NotOn messages were the 100n cap of the 74HC165, the IC itself, or the resistor network. I do not know how often I have checked and/or re-done the corresponding solder joint. Nothing has helped so far. I further checked the orientation of the 165 and the resistor network, as well as their part number based on the mouser BOM.

Check pin 16 of the RN is connected to +5V. All 165 digital inputs (pins 3-6, 11-14) and pin 10 should be at +5V, possibly excepting 5 and 6 connected to the encoder.

Check for continuity between DIN pins (they should typically not be connected).

What is the part number of the RN just to be sure?

 

Photos would help, and a common problem is that the cable is not assembled properly. Try again with another cable.

 

 

Link to comment
Share on other sites

8 minutes ago, latigid on said:

Check pin 16 of the RN is connected to +5V. All 165 digital inputs (pins 3-6, 11-14) and pin 10 should be at +5V, possibly excepting 5 and 6 connected to the encoder.

Check for continuity between DIN pins (they should typically not be connected).

What is the part number of the RN just to be sure?

 

Photos would help, and a common problem is that the cable is not assembled properly. Try again with another cable.

Wow, thanks for the prompt response!!!

RN pin 16 showes 5V. So does pin 1 and 16 of the 165. However, none of pins 3-6, 10, 11-14 is on 5V!

165 pin 7-8 have a resistance of 5MOhm (not sure if thats"normal), the rest is not shorted at all.

The RN has the writings 4816P LF 1-103 C184 on top. No pins seem connected using a multimeter.

Pictures are following.

 

Quote

Are you testing the JA board by itself or hooked up to lemec_R? The shift registers are not located on the JA board.

Actually, alone. That explains the matrix well. I was wondering already since the pins of the shift registers don't relly match up to the number of LEDs. Thought some fancy multiplexing is used :)

 

Link to comment
Share on other sites

9 minutes ago, Karg said:

Wow, thanks for the prompt response!!!

RN pin 16 showes 5V. So does pin 1 and 16 of the 165. However, none of pins 3-6, 10, 11-14 is on 5V!

165 pin 7-8 have a resistance of 5MOhm (not sure if thats"normal), the rest is not shorted at all.

The RN has the writings 4816P LF 1-103 C184 on top. No pins seem connected using a multimeter.

Hmm, I think that's the wrong part. The bussed network has a "2" prefix.

Cross-check the part here:

RN_JA.thumb.GIF.249bf0b31a473b3019b08b9b

You should have only one piece of 652-4816P-T2LF-10K and two of 652-4816P-1LF-10K .

 

Best,
Andy

 

Edited by latigid on
Link to comment
Share on other sites

Just now, latigid on said:

Hmm, I think that's the wrong part. The bussed network has a "2" prefix.

You should have only one piece of 652-4816P-T2LF-10K and two of 652-4816P-1LF-10K .

 

Aaaaah, this could have been so obvious. I will see that I change the ICs and then report back.

Thanks a lot already!!!

Link to comment
Share on other sites

Works like a charm now!

Still annoys me, that mistake :) Particularly since it is mentioned everywhere to check the parts number (and I actually confirm every single item). I guess mistakes happen. Thanks a lot again for the quick and competent help!!! A beer is coming your way :)

Actually, you don't happen to live in the Rhein-Main Area in Germany? Then we could have that one in person...

 

Link to comment
Share on other sites

Troubleshooting round two.

I am a lot closer to finishing the build, and just have performed all tests described in the build video 2 around 1:30-1:40. All tests were successful but one:

the MEC switch SW18 on one of the LeMEC boards (R, the larger one) does not respond. I took the switch out and shortened the pads to rule out a defective switch already. Further, I have traced the diode next to it to pin4 on IC3. Judging with my multimeter, this conenction seems good. My plan was to trace the other pins of SW18 of the working board, to then compare with the non-functional one, but I am failing to correctly trace them. Any hints towards the right direction?

 

 

Link to comment
Share on other sites

Hello, just on the road at the moment, but you might try to check the connection made with a MEC switch. It is marked on the silkscreen as 'sink bridged by switch' or similar. The problem would be the same for all SW17-20. Test for continuity over that line and insert a bridge wire or solder the switch in if needed. 

Are any other switches not functioning? 

Link to comment
Share on other sites

25 minutes ago, latigid on said:

Check the connection made with a MEC switch. It is marked on the silkscreen as 'sink bridged by switch' or similar. The problem would be the same for all SW17-20. Test for continuity over that line and insert a bridge wire or solder the switch in if needed. 

Are any other switches not functioning? 

I don't fully understand :(

However, if I use a wire to connect the "sink bridged by switch" with the pin in the right bottom corner of a MEC switch, I typically get the "keypressed" response in MIOS Studio.

However, with this test as well, I do not get this for SW18. All other swotches are working well (MEC and Matias).

Link to comment
Share on other sites

I can't offer an explanation that would have a problem isolated to SW18. If there was an issue with 165 ICs, then the whole column wouldn't work. If there was an issue with transistors on the sink side, then the whole row of four wouldn't work.

My immediate guess would be that the diode is soldered the wrong way around or with cold joints. Could also be a faulty diode, but it's very unlikely.

This is viewed from the top of the board. I've numbered the pins of SW18.

lelemec_sw17-20.thumb.GIF.dce99d5c920eb2

Pin 1 (+2) is the cathode pin (also used for the LED). The connection should be common to the other switches SW17/19/20. If the marked bridge on SW19 is not present (i.e. no switch or bridge wire), then none of those four switches will work.

Pin 3 is the DIN input. It should connect to the diode adjacent. The diode anode (i.e. non-striped end) should be common to other diode anodes in the column (i.e. on SW2/10).

 

Link to comment
Share on other sites

22 hours ago, latigid on said:

I can't offer an explanation that would have a problem isolated to SW18. If there was an issue with 165 ICs, then the whole column wouldn't work. If there was an issue with transistors on the sink side, then the whole row of four wouldn't work.

My immediate guess would be that the diode is soldered the wrong way around or with cold joints. Could also be a faulty diode, but it's very unlikely.

Thanks again! That hint pointed into the right direction. I exchanged the diode last night and then the switch worked :)

Unfortunately, I directly experience the next problem. In this context, my appologies for bothering you so much. I have not fully understood yet how the wiring works, thus I am constantly fishing in the dark with my troubleshooting attemts and eventually run out of reasonable ideas. Your help is greatly appreciated!!

The next (and hopefully) last problem is, after soldering the MEC/Leds the latter don't light up on both LeMEC boardsn when performing the l_seq.ng testing. I discovered it right after exchanging the diode last night, and have tried to isolate the exact problem since then. What I have done so far:

  • Confirmed orientation of major parts, including LEDs
  • Resoldered most connections
  • Testing in MIOS Studio: l_seq.ng:
  1. MEC swithes are all recognized
  2. full Jog section works, including 16x8 matrix
  3. ngr set led:XXXX 127 switches all Flux LEDs on (checked by bridging the cathodes)
Link to comment
Share on other sites

Pin 1 is correct for SJs. Unless you're using a different RJ config, such as green and blue LEDs? Picture would help to clarify. 

Then are you sure about LED orientation? Check with diode mode of your multimeter. Red lead to circle, black lead to line on the MEC switch silkscreen should illuminate. 

Some have had problems over-stretching LED legs causing them to break. It would be odd for all LEDs to break  though. 

 

Link to comment
Share on other sites

Makes sense! 

SJ pins correspond to the numbers on superflux LEDs: 1=red 2=green 3=blue.

As you're not using the red superflux LED, you should instead jumper SJs to pin 2. This connects the MEC LED to the A pin on the RJ, so the hardware configuration should stay the same. 

(this assumes that SJs A connect to 2 and B to 3)

Link to comment
Share on other sites

Hello Everyone, I've decided to try a hot air station for my v4+ build. So far so good. My question is, is it safe to use hot air on the superflux leds? Should I just use an iron for those?

thanks,

Luke

 

Edited by lukas412
Link to comment
Share on other sites

1 hour ago, lukas412 said:

Hello Everyone, I've decided to try a hot air station for my v4+ build. So far so good. My question is, is it safe to use hot air on the superflux leds? Should I just use an iron for those?

thanks,

Luke

 

I recommend to use a soldering iron to keep the heat away from the semiconductor dies/wire bonds. If you use hot air, perhaps bake the LEDs at 80°C overnight to lessen the chance of failures (steam microbubbles might cause issues).

Note that I haven't tried baking the LEDs, so YMMV.

Link to comment
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

Loading...
 Share

×
×
  • Create New...