FX-Sabers.com

The S.A.B.E.R. GUILD: Saber Manufacturers => Plecter Labs => Sound boards technical issues / questions / problems => Topic started by: mech65 on October 04, 2020, 05:02:08 PM

Title: CFX fw.dat
Post by: mech65 on October 04, 2020, 05:02:08 PM
Solved

Hello
I have installed the CFX board into a Heiland hilt, neo pixel setup. At first everything worked ok. I started to use edited blade profiles and the board acted strange. It would shut down or lock up, no buttons would work. New high amp battery always charged to over 4.1 VDC. Formatted sd card many times and used fresh sd files from Plector site. I have been trying to load firmware to the board via %u201CDocklight%u201D. The board goes into boot loader mode and the comport is selected and recognized. Activation switch is pressed and everything is ok until the end. Message says I/O comport error. I have used two laptops and two new micro USB-USB cables. Same message. This board seems to be possessed now after working perfectly. Any advice would be greatly appreciated.
Thanks
Mark
Title: Re: CFX fw.dat
Post by: erv on October 04, 2020, 11:24:22 PM
board reboots after updating and close the COM port
Title: Re: CFX fw.dat
Post by: mech65 on October 05, 2020, 03:36:08 PM
I have tried loading the fw.dat file without Docklight. Pulled the killkey while holding the act. and aux. buttons. Board goes into boot loader mode with slow pulsating led. Push the act. button and a few seconds later I get the boot sound. However, I can ignite the first font and from there the board locks up and does not respond to any buttons. I am at a loss here considering it worked fine after I first wired everything up. I will try an older fw.dat.
Title: Re: CFX fw.dat
Post by: mech65 on October 05, 2020, 06:33:02 PM
Hello again
I have loaded the firmware CFX V-10.00 4E and the sd default package V10.005A. My board will still not operate. It makes a boot sound and that is it. Am I using the proper firmware and sd default packages. I have adjusted all my configurations to what I  had before when the board worked properly. I am probably overlooking something but cannot pinpoint it.
Title: Re: CFX fw.dat
Post by: erv on October 05, 2020, 09:28:38 PM
Log the board with debug enabled and also check all the voltage map
Title: Re: CFX fw.dat
Post by: mech65 on October 06, 2020, 05:43:05 AM
I checked the voltages last night and some are out of whack. The main switch reads only .61v and the aux. switch is 2.75v. Beside the motion sensor there seems to be two resistors, I am guessing. The one on the left should read 14v but is only reading 8.86v. All the other voltage readings are good.
Title: Re: CFX fw.dat
Post by: erv on October 06, 2020, 06:03:03 AM
8.8 v is correct for the motion sensor (that was updated in the manual long ago).
your switches should be 3.3V
Title: Re: CFX fw.dat
Post by: mech65 on October 06, 2020, 06:08:50 AM
The auxiliary switch is 2.75vdc and the main switch is .61vdc. The version 1.07 manual shows 14volts. Do I have the latest version?
Title: Re: CFX fw.dat
Post by: erv on October 06, 2020, 06:17:22 AM
xxx, 1.07 is from february that's what I meant for long ago. Corrected in 1.08 (not published yet). Your switches must be 3.3V when not pressed
Title: Re: CFX fw.dat
Post by: mech65 on October 06, 2020, 06:21:58 AM
Thanks for getting back to me. What do I do about the low voltages on my switches? 2.75v on the aux. and .61v on the main. Can I repair this somehow? My board is non responsive to the switches.
Title: Re: CFX fw.dat
Post by: erv on October 06, 2020, 08:06:08 AM
remove switches and wiring, look for shorts. Otherwise, the pads could be internally damaged (ESD, induction from blade current)
Title: Re: CFX fw.dat
Post by: mech65 on October 06, 2020, 10:50:44 AM
Ok, to night I will remove switch wires and clean up pads with wick and resolder. How can I repair pads if damaged by induction?
Title: Re: CFX fw.dat
Post by: mech65 on October 06, 2020, 11:02:21 AM
I have just read on theCFX Facebook page that you can rescue your switch pads by adding a resistor in between the pad and switch. Is this viable if I do not find any shorts in my wiring?
Title: Re: CFX fw.dat
Post by: erv on October 06, 2020, 12:13:30 PM
you can try and use pudis=1
Title: Re: CFX fw.dat
Post by: mech65 on October 06, 2020, 01:02:22 PM
Hello Erv, what is pudis? I cannot find it in the configuration.
Mark
Title: Re: CFX fw.dat
Post by: mech65 on October 07, 2020, 05:42:49 AM
Last night I resoldered the switch and ground pads. Checked the wires and switches for momentary continuity. They look good. Also moved the positive wire for the blade as far away as I could from the switch wires. Replaced the sd card and pulled the kill key. The board boots up fine and everything works great. Then I plug the blade in and ignite it. The board locks up again and the bladeignites and retracts by itself before refusing to do anything else. What should I check next?
Title: Re: CFX fw.dat
Post by: erv on October 07, 2020, 06:12:46 AM
please post a video

if the blade locks up (like engaging aux to do a lockup) you're getting current induction into your switch circuit from the blade power. Make sure you use a dedicated low power ground / negative wire for the switch circuit
Title: Re: CFX fw.dat
Post by: mech65 on October 07, 2020, 06:34:04 AM
Sorry, what I meant by lock up was the board is non responsive. I have the ground wires going to the ground pad according to to the wiring schematic. Everything worked until I plugged the blade in. What is “pudis=1”
Title: Re: CFX fw.dat
Post by: erv on October 07, 2020, 06:42:04 AM
pullup disable
Title: Re: CFX fw.dat
Post by: mech65 on October 07, 2020, 07:13:48 AM
What does pull up disable mean? Where is it located?
Title: Re: CFX fw.dat
Post by: EXAR KUN on October 07, 2020, 09:48:13 AM
I think this is a good article on it:

Disable pull-up/pull-down resistors? - Getting Started - Particle (https://community.particle.io/t/disable-pull-up-pull-down-resistors/14088/3)

I'd also check your blade wiring and emitter PCB wiring. A stray wire hair can cause serious implications in Neopixel.
Title: Re: CFX fw.dat
Post by: mech65 on October 07, 2020, 02:22:28 PM
Thanks Exar Kun, I am getting down to completely desoldering everything and staring over. Can’t figure out why it worked great in the beginning and degraded over time to where I am now. I am reading the link you sent me, I will have to read a few more times to digest the info, I have been an Industrial Mechanic all my life and am lacking in electronics. Thanks again.
Title: Re: CFX fw.dat
Post by: mech65 on October 08, 2020, 05:20:18 AM
Last night I removed all the switch and ground wires from the board again. This time I used a jumper to go from the act. pad to the ground pad and the same for the aux. pad. With the blade plugged in everything is working great. Even though the wiring for the switches looked good something must be wrong to cause all the errors and spontaneous ignitions. I will replace all the switch wiring and the switches and go from there.
Title: Re: CFX fw.dat
Post by: erv on October 08, 2020, 05:26:49 AM
before doing so, measure the voltage on your (bare) switch pads
Title: Re: CFX fw.dat
Post by: mech65 on October 08, 2020, 01:06:29 PM
Sd card removed
Aux. 3.2vdc
Act. 3.14vdc

Sd card installed
Aux. 3.18vdc
Act. 3.17vdc
Title: Re: CFX fw.dat
Post by: scott on October 08, 2020, 07:57:38 PM
Can you send a picture of your board wiring?
Title: Re: CFX fw.dat
Post by: mech65 on October 09, 2020, 05:05:03 PM
(https://i.ibb.co/jhKMN2X/64-D038-BF-DE96-41-FB-A579-E48-F1-BB8-C0-A4.jpg) (https://ibb.co/jhKMN2X)
Title: Re: CFX fw.dat
Post by: scott on October 09, 2020, 06:27:09 PM
Can we see the other side as well?

Always tough to see in pics, but I have seen a lot of issues where flux is causing shorts. Tough to see the clean margins between pads, but def make sure there is clean clearance between all pads...looks close between speaker pads and +\-, although if they were connected you would have bigger issues.

Asked about the other side as I am not sure what you meant by jumping between switches and ground. If you just mean combining the ground wires from both switches, cool.
Title: Re: CFX fw.dat
Post by: EXAR KUN on October 09, 2020, 09:57:41 PM
Is your ground pad gone on this side?
Title: Re: CFX fw.dat
Post by: bombarta on October 10, 2020, 02:42:43 AM
Battery ground does not look right to me, far too much solder at either side of the data pad too. The data wire does not need to be so large as there's very little current running through it.
I take it the data wire or the blade has got the appropriate resistor fitted for NeoPixel? Sorry if you already answered that one.
Title: Re: CFX fw.dat
Post by: mech65 on October 10, 2020, 08:57:53 AM
I will take pictures of the other side. The ground pad is there, I just cleaned it up with solder wick. The blade and sound work great when touch a wire from act. switch to ground and aux. works also. I am going to run new wire and switches and try it again. I have a 330 ohm resistor in the data line.
Thanks for the responses
Title: Re: CFX fw.dat
Post by: mech65 on October 10, 2020, 04:35:31 PM
Hi bombarta, I checked continuity between the data pad and the grounds near it. All good. Smallest gauge wire I had was 26.  It probably has too much solder on the battery  pads but they are solid, I am getting full battery voltage when the kill key is pulled even on the other side of of the board.
Title: Re: CFX fw.dat
Post by: scott on October 10, 2020, 05:21:22 PM
Would still be good to see the other side of the board...we also have not talked about the neo pcb and blade. If you have more than one blade I assume you would have tried them. Possible that there is some issue or minor short on the pixel pcb?  Always good to check continuity from the wires on the board to the appropriate pins...and looking for 330 on the data wire to pin, not continuity.
Title: Re: CFX fw.dat
Post by: mech65 on October 11, 2020, 05:26:21 AM
I am using a four pin GX-12 connector for the blade. I printed an adapter that goes into the blade and houses the male side of the connector.  What I meant by continuity was there is none between the data pin and the grounds next to it telling me there is no short. I have checked the blade harness from the GX connector back to the board. 330 ohm for data and full battery power to the two grounds.
Title: Re: CFX fw.dat
Post by: mech65 on October 17, 2020, 07:15:47 PM
I have not been able to work on this project for a while. The last thing I did was remove the switch wires and switches. I then jumped the switch act. to ground and everything worked great including jumping the aux. pad to ground to change fonts and blaster bolts. I tinned the switch pads and ground pad and tried again but board boots up in font selection mode and then says it is going into deep sleep, nothing responds after that. Formatted sd card again, battery at 4.3vdc, same result. Scott, here is a picture of the back of the board. Still scratching my head.

(https://i.ibb.co/FBPYgvB/A02-D6-C5-D-9-AD5-4-EAE-ADAC-1677550816-B5.jpg) (https://ibb.co/FBPYgvB)
Title: Re: CFX fw.dat
Post by: scott on October 17, 2020, 08:11:00 PM
I know how frustrating it can be. And happy to help....unfortunately. When I asked to see the other side of the board, I was hoping to see how you had everything wired. Looks like you flipped switch wiring to the other side of the board.

Next time you are fully wired up, include pics of both sides...I am probably missing something but was confused when you talked about jumping act and aux. you want one pin of each switch going to the ground....bigger pad under all the accent led pads (4th one in after the 3 tiny ones) and the other pin from each switch going to the act and aux pads in the middle.  I know you mentioned testing but always good to test continuity on switches with your multimeter before attaching to the board.

Lastly. Since we are talking about switches, not sure what your saber is, but make sure you don’t have any shorts on the switch side (for instance, if the solder could hit the ousted of the hilt or something underneath.

Apologies if I missed this info/detail previously, just a bit of a moving target in the thread.
Title: Re: CFX fw.dat
Post by: erv on October 18, 2020, 12:45:03 AM
check your switch pad voltages again. If you get in the vocal menu, your aux switch (pad) is shorted to ground, or fried (internal pullup destroyed)
Title: Re: CFX fw.dat
Post by: mech65 on October 18, 2020, 04:56:35 PM
Hi Scott, yes the wiring is all on the top side. What I meant by jumping the switch pads to ground is I use a small length of wire trimmed and tinned at both ends. I then put one end on the ground pad and momentarily touch the act. switch to ignite the blade or do the same with aux. switch to change fonts or activate blaster blocks. However, this time around nothing is working. I have new wiring and switches in place but not soldered in, that’s why I was testing with the jumper wire. Hope that clears it up a bit. As for Erv question about the voltage on the pads. I have only .6 volts on the aux. pad and 2.76 volts on the act. pad. Erv is probably right about the aux. pad being fried. I am thinking about trying a pull-up resistor for the aux. pad.

(https://i.ibb.co/1QKXkpn/0-D3-E991-E-5733-4-DB5-88-E9-BE5-F731927-E5.png) (https://ibb.co/1QKXkpn)
Title: Re: CFX fw.dat
Post by: scott on October 18, 2020, 05:58:34 PM
If aux is toast, you should still be able to turn on and off. Would think.
Title: Re: CFX fw.dat
Post by: mech65 on October 18, 2020, 07:36:22 PM
It boots up only the vocal menu, does not let me do anything and then goes into deep sleep mode. No action after that. I am going to put a 3300 ohm resistor as per the drawing and hope for the best.
Title: Re: CFX fw.dat
Post by: erv on October 19, 2020, 01:14:50 AM
If aux is toast, you should still be able to turn on and off. Would think.

No, as it takes you to the menu and waits for aux to be released
Title: Re: CFX fw.dat
Post by: mech65 on October 19, 2020, 01:46:47 PM
I put a 3300 ohm resistor in as per my last post. Voltage on the aux. pad went from .76 volts to 1.79volts but the board boots and does not respond to any buttons. I am going to try a 1000 ohm resistor.(https://i.ibb.co/9WLPX2y/A691-A09-D-96-B1-41-DF-9-AD4-DFAE5-A7-E3-F5-E.jpg) (https://ibb.co/9WLPX2y)
Title: Re: CFX fw.dat
Post by: mech65 on October 19, 2020, 06:50:05 PM
I have just finished installing 1000  ohm pull-up resistors on the act. and aux. pads to the 3.3 volt pad. I had 2.63 volts on the aux. pad and 3.23 volts on the act. pad. I installed the sd card and voltages changed to 1.67 volts on the act. pad and .62 volts on the aux. pad. Boots up, red light goes out. Blade ignites for a few seconds and retracts. Nothing works after that. Very perplexing. After five builds I have never come across a situation as this. This is my first CFX build. Always trying new things. Anyone have any ideas? I am about to cut losses and move on to the  Proffie boards. There are always failures but I have put a lot of time into this board to make it work.
Title: Re: CFX fw.dat
Post by: erv on October 19, 2020, 09:16:35 PM
Your internal pullup is probably damaged (inside the mcu), dragging signal to the ground.

You can disable pullups (and use your external ones) by installing pudis=1 in the config just after debug.
Title: Re: CFX fw.dat
Post by: mech65 on October 20, 2020, 05:00:01 AM
Would you please give me instructions on how to install “pudis=1”. I was going to completely resoldered the board today and start over but I will give this a try . Do I just type this line in?
Title: Re: CFX fw.dat
Post by: erv on October 20, 2020, 05:12:31 AM
yes you simply put this in the config.txt, as early as possible (hence right after the debug line)
Title: Re: CFX fw.dat
Post by: mech65 on October 21, 2020, 04:42:06 PM
Thanks Erv. I have some life in this board now. I kept changing resistor sizes on the act. and aux switches trying to get the highest voltage on the pad. I have 2.6 volts on the aux. pad and 2.4 volts on the act. pad. I am using a 160 ohm resistor on the aux. pad and 100 ohm resistor on the act. pad. Any higher and the board locks up and won’ t respond to buttons. Any lower on the act. resistor and the blade ignites and retracts by itself randomly. Thanks to everyone’s help and Erv’s direction to “ pudis=1”.

Title: Re: CFX fw.dat
Post by: mech65 on October 21, 2020, 04:42:49 PM
(https://i.ibb.co/3WhrsFk/9-C178454-CE61-4635-B0-AB-F8-B80798-AAC2.jpg) (https://ibb.co/3WhrsFk)
Title: Re: CFX fw.dat
Post by: erv on October 22, 2020, 01:25:42 AM
glad you got it revived !