Prizm 5.1 Neoxpixel install fails, weird beep sound and then nothing happens.

Hello,
I am new here and last week i exitedly opened my package with my prizm 5.1 in it to start my install for a photon neopixel build.

After soldering for a test-wiring my prizm just makes some beep sounds as soon as i take out the kill key but thats all for it, nothing more to happen..
if i put the killkey back in and pull it out again i will always get that same sound without anymore functions. I use a saber essential speaker, a full-charged panasonic 3400mah 3,7V Li-Ion-Battery, two momentary switches and the tcss pcb connectors for the led strips ws2812b.
I bridged the resistor pads on LS pad with a 0603 smd.
SD Card shows no issues whatsoever when i read it with the cardreader, didnt solder or work with the board with sd card in it, worked on a antistatic mat and always wear an antistatic wrist band. I dont know what the issue could be, pls anyone help to get me that neopixel build i want so badly.

https://ibb.co/c5tP3z
https://ibb.co/hexLAe
https://ibb.co/mJH6GK

Is the battery protection circuit tripping? Have you measured voltage on the supply pads? Is the battery fully charged? Do you have the right configs loaded? Photos look OK from a brief perusal of your soldering.

Tom

Quote:

---

Originally Posted by Tom Tilmon

Is the battery protection circuit tripping? Have you measured voltage on the supply pads? Is the battery fully charged? Do you have the right configs loaded? Photos look OK from a brief perusal of your soldering.

Tom

---

I checked everything twice. At least i found the solution. Was quite simple but hard to recognize. My data on the SD-Card was all in a Folder called Prizm5. As soon as i deleted the folder and shifted the data directly to the diskfolder everything worked fine.

yes, there should only be folders entitled: bank1, bank2, bank3, bank4, bank5, and bank6. glad you got it working.

Well, unfortunately problems are continueing..:(

My double strip only lights up red instead of blue (1st bank) or green (2nd bank). On the bank for green the red is flickering quite strong. There is some crackling on the speakers when lit, on the green setting the crackling is even louder. Foc and blaster work absolutely fine in white, thats why i dont understand why i cant adress blue or green colours. Battery is fully charged. Resistor 330 ohm is on dataline. I already snipped the first two pixels of the strings away and resoldered it to make sure its not a fried first pixel. Still the same..
Does anyone know what the issue could be?

so the strips you got are RGB, instead of the GRB. Most strips are GRB, but every now and then I get a pair that is RGB. On the boards I use, you can change the color order profile. I'll read the prism manual and see if color order is adjustable. The speaker crackling is typical of a struggling battery. Do you have room in the hilt for a bigger battery? Also, is that battery fully charged when the speaker crackles?

I'll check the prism manual to see if you can adapt it to those RGB strips.

Tom

yes thats what i thought too so i charged the battery fully before i tried again to make sure. Its a 3400mah NCR fully charged, though i still will try to modify my chassis for a 26650 with a handmill. the space in the hilt should be enough. Unfortunately it doesnt seem to be the issue here. I set my banks for 1st bank 0,0,1023 2nd bank 0,1023,0, 3rd bank 1023,0,0 but all of them light up red, 2nd bank hard flickering with very loud crackling. The strips are actually stated as RGB. Could it be just the no-brand cheap WS2182b strips itself that wont handle the data? Maybe i should by the real adafruit, i read about so many problems with cheap strips, but you had never any issues with the skinny adafruits right? so maybe worth the money but they are so freaking expensive..

You cannot remap the strip controller codes. They are set in stone. You can however, change your color configs in the config for each bank, as well as the main override file (where I think all the color profiles are stored for color changing functions).

Here is paste from Prism P. 34:

Quote:

---

Ledstrip color definition - Color Order
For a ledstrip, the color of the blade and FoC are defined by the same keywords and technique as high-power LEDs. Color profiles are set with a list of 3 numbers defining the color mix of R, G and B using the keyword color and fcolor (FoC). For standard blade color definition (config.txt) the same color mix is defined by the keywords drive and fdrive.
The color order is however specfic in the case of a ledstrip as you cannot remap the way the strip controller codes colors. We've kept the classic RGB order which works for all the pixels we've tested except the PL family which will be GRB.
For instance, a red blade using regular Neopixels would be defined by :
drive=1023,0,0 <per font color definition>
or
color0=1023,0,0 <color profile 0>

---

Tom

cheap strips can be spotty. I have bought good ones ane bad ones. I really wish we could get standard strips that are trustworthy, but right now its a crapshoot. I"ve been buying from Fleabay and here lately have mostly had good luck. Don't toss your prism out, most of the problems might be those strips. Yeah this stuf is expensive. I built 2 neopixel Kylos and the one guy was like is it done yet? I had to look at him as Anakin looked at the Tusken Village, but was able to gain control again. My frustration was I ran into bad strips back to back, and then the intelliblade I bought didn't work either. So I had spent about 400 bucks in stuff that didn't work.

There are decent strips out there. You can buy the adafruit ones that will work for sure, or take another stab at some more Chinese strips. I'd post a link but this is TCSS site. NO can do. I hope you get better strips. PS, the way you say they are acting, you could have a bad pixel on one of your stirps. Is the weird behavior on both sides of the strips or only on one side. Does it start in a specific spot, or is the whole entire length of the strip involved?

Tom

Ok I tried various drive settings, especially 0, 1023, 1023/ 0, 1023, 0/ 0, 0, 1023/ 1023, 0, 0. They all light up red. As I tried to set various foccolours i tried fdrive 0, 1023, 0/ 1023, 1023, 1023/ 1023, 0, 0/ 0, 0, 1023. All of them light up white. Furthermore i noticed that at pulling the kill key theres one short initial flicker of white over all the strip for a short half second or sth.

Yes thats what i meant, i guess its the chinese strips itself that cant handle the data well. They light up from the first pixel to the very end of the strip evenly in red with a slight flickering, getting worse on the green setting. On both strips its the very same behaviour. I already clipped the first two pixels of both strips to make sure they arent fried or defect. But i think i will buy adafruits now instead of trying more cheap ones.. thats frustrating...

and btw.. thanks so much for these prompt answers, i appreciate that so much

PS, another thought about the weird behavior: I use 24 or 26 ga wire on the strips to carry the amps. If you're using too thin of a wire gauge, the full power needed isn't reaching the strips, that can cause unusual behavior too. To test this, with the prism on, measure voltage coming out of prism with multimeter. Then measure voltage the first pads of your strips. Then measure voltage at the tip end of your strips. They should all be relatively close, and if not, you're losing juice somewhere (like in thin wires).

Check all 3 connections at board and on strips (positive, negative, and data) for good connections. Flow them again, make sure that the joints are shiny after you hit them. Cold solder joints can cause weird behavior. If you have assorted resistors lying around, you can try different resistors on the data line. I have found great success using 470 ohm, but have heard of people running anywhere from 100-500, and a few that don't use data line resistors at all (I wouldn't do that).

PS, thin gauge wire is fine for the data line, but not for the power and return to the strips.

Tom

I have now built 20 or so neopixel blades, so I know the frustration. Seem to me you have total crap strips. Get the Adafruit if this is your personal saber, and you want to stop throwing money away. However, you can find better vendors. I'm paying 18-20 bucks a meter for most of my strips now, and 90% of the time, they work well; however, every now and then I get sent a meter or two that are either total crap that I have to throw away, or lots of crap where I have to cut out pixels in the middle of the strips, and rejoin, reflow each pixel connections and transistors. A bad strip can be a lot of work to fix for sure. The Adafruit strips work. If you have problems with them, then you can start looking at your wiring on the prism.

PS, is your data line bridge solid on the prism?

Tom

I was re checking your board photos. On your negatives to the neopixels you are running them all through only one of the pads. That is bad. You need to use at least 2 of those mosfets. You may have fried that one mosfet that both of your strips are connected to. Attach wires to the other two mosfets, the ones you haven't used already. Double strips need to divide their ampload across more mosfets. You might have hosed that single mosfet you were using.

The white flash upon boot up is normal. You can get rid of that by adding a capacitor to the stirps, but I don't. That white flash is no biggie.

Tom

Quote:

---

Originally Posted by Tom Tilmon

I was re checking your board photos. On your negatives to the neopixels you are running them all through only one of the pads. That is bad. You need to use at least 2 of those mosfets. You may have fried that one mosfet that both of your strips are connected to. Attach wires to the other two mosfets, the ones you haven't used already. Double strips need to divide their ampload across more mosfets. You might have hosed that single mosfet you were using.

The white flash upon boot up is normal. You can get rid of that by adding a capacitor to the stirps, but I don't. That white flash is no biggie.

Tom

---

Uhm theres a second white wire on the first mosfet straight in the outer edge too, its just not to be seen on the photo, sry for this. I used 26 gauge wires for the connections. Dataline is a nice solid soldered connection, resoldered it for nice shiny joints. Thought about a capacitor too, do you think the colour behaviour could be solved by the capacitor too? I really guess its the strips..

No, once juice is flowing and the strips light up, that capacitor has 0 influence. It just doesn't allow that initial pulse to hit the blade when you pull the kill key. I think you have bad strips if you're running off both mosfets then.

well, today my adafruits arrived and unfortunately the issue still lasts on. I wired them up to a NBV4 of a friend of mine and nearly everthing worked fine. Also tried to use a brand new SD-Card to make sure it cant be that. So i guess i fried either the mosfet of the Dataline or the nearby resistorpads by soldering too long on them...

But there is one issue with the NBV4 i still cant handle. I tried various parameters for qon qoff between 100 and 500, lsfadeon and lsfadeoff are set to 1 while lightstick is 0 but still i get no ignition and retraction effect at all. The led strip lights up instantly and instantly shuts down as well.

did you change all the config settings for neopixels? Including how many are in your strip?

Quote:

---

Originally Posted by Tom Tilmon

did you change all the config settings for neopixels? Including how many are in your strip?

---

Yes of course, 144 as they are on the strips themselves. Everything works fine, including any colourmixing, just ignition and retraction effect fails for no apparent reason

I'll get that occasionally if there is a stray short between data and ground. (on the strips and on both adpaters. Check microscopically for any errant wire threads crossing over).