#4946444
Or try swapping your existing servos over temporarily - if the fault follows you know it's the servo itself (and worth obtaining another). If it doesn't, more likely cabling/elsewhere. Doesn't sound like code if it's only affecting the one tho.
#4947987
Same boat than you mate everything bought from sean too, but both servos are acting as crazy and i ve not the skill to recode. a bit sad as it was a big first for me in soldering and using arduino :( . If you got any solution... Thanks to Ecto labs by the way to take the time on Fb to speak with me :) !
#4947988
As noted in my electronics guide the servos twitching a little is fairly normal due to hardware/coding stuff which I don't understand. This was the setup that Jeremy Williams did for our original ghost trap. The servos going completely crazy is not normal and often caused by a crossed wire/short circuit somewhere which can be a total pain to narrow down. Make sure the mod done to the Music Maker board (cutting the trace) is nice and clean and complete. Try removing modules and try just the servos. If they act normal start adding modules back until something goes wrong. Apparently another problem is that there are many, many, many fake Tower Pro servos - not just replicating the design but even the sticker! I try to buy from reliable places and I test each servo before putting it in a kit but if you find that one servo is causing a problem let me know. Try hooking up one servo at a time to see if one is the problem.


Sean
#4947990
countspatula wrote: April 10th, 2021, 3:27 pm As noted in my electronics guide the servos twitching a little is fairly normal due to hardware/coding stuff which I don't understand. This was the setup that Jeremy Williams did for our original ghost trap. The servos going completely crazy is not normal and often caused by a crossed wire/short circuit somewhere which can be a total pain to narrow down. Make sure the mod done to the Music Maker board (cutting the trace) is nice and clean and complete. Try removing modules and try just the servos. If they act normal start adding modules back until something goes wrong. Apparently another problem is that there are many, many, many fake Tower Pro servos - not just replicating the design but even the sticker! I try to buy from reliable places and I test each servo before putting it in a kit but if you find that one servo is causing a problem let me know. Try hooking up one servo at a time to see if one is the problem.


Sean

hi sean, sent you a message on etsy about the problem, i really don t know what s happening i unplugged the light to be able to make a video. Motor moving crazy only when i push button and activate sequence. I followed your manual as i'm not an expert on electronic and even less on arduino like you, i worked step by step taking my time (around a week) to be sure i wouldn't do anything wrong and following all the tutorial, and did my best for a clean job. everything works great, just that servo problem sadly.
Last edited by dav59300 on April 11th, 2021, 1:07 am, edited 1 time in total.
#4947994
As noted a while ago in the thread, I too had a problem with the servos going crazy. I checked and rechecked every single thing a dozen times, had various other people look at it (including an electronics specialist from the animatronics studio I work at) and ended up replacing nearly every part. Nothing fixed it except switching over to Ectolabs' code and wiring, at which point it immediately worked. It's still a mystery that bugs me.
#4948001
is the wiring really different or easy access for a beginner ?i sthere any tutorial to explain it ? i 'm stuck on this one i guess. I was nearly ordering new servo and adafruit mp3 player today in case i did anything wrong. I also unplugged everything, tested one servo at a time etc... they still move a lot after pushing the button to open doors, and act like that until i close the doors. So would it be a servo problem ? wouldn t it be easier to have recode as commission from someone to make the servos stop after opening and rework before closing doors ? Sorry for the vocabulary as i'm not an expert but looking for a good solution before final asssembly ;)

edit spent the day checking with someone, no clue about what 's going on, we tested and retested. Here is a quick video of the servo :

https://www.facebook.com/daviddeessever ... 6078843772
#4948019
Yep, looks just like mine did. Ectolabs changed the code a lot, while also changing the lights to 3 neopixel jewels linked together, and attaching some of the parts to different pins. Changing it took 15 minutes maybe. I don't understand why, but the servos never worked properly in the original configuration, and then worked flawlessly with Ectolabs' mods.

Ectolabs definitely earned this: :crunch:
#4948025
All the information I used to mod it is on page 1 of this thread, where you'll also find Ectolabs' code, sounds, and the 2 libraries you need to include with the code. I can dig out my electronics later and take a picture of the wiring so you can see which parts are wired differently. It's only the lights and servo pins that get changed, and the lights are a really great upgrade. You'll need to purchase 3 neopixel jewels before using his code and wiring.
#4948088
Nuclearjungle wrote: April 11th, 2021, 8:26 am Yep, looks just like mine did. Ectolabs changed the code a lot, while also changing the lights to 3 neopixel jewels linked together, and attaching some of the parts to different pins. Changing it took 15 minutes maybe. I don't understand why, but the servos never worked properly in the original configuration, and then worked flawlessly with Ectolabs' mods.

Ectolabs definitely earned this: :crunch:
Thank you :-P

Interesting... I didn't realise my mods had fixed servos that are otherwise twitchy with the original code. Mine were okay, but maybe it's just luck of the draw. The edits I made were to ensure there was no interference when using NeoPixels instead of the individual LEDs the original code was meant for.

However, I'm thinking now that I may be able to create another modified version of the original code for the standard hardware, so no rewiring required. This would involve keeping everything the same, but incorporating my servo specific edits which adds a new function that makes sure the servos are only attached to the code for the short time needed for door opening and closing.

I will have a look at that as soon as I get the chance and update here. It might not work, but probably worth a try before buying replacement components.

Dave
#4948090
EctoLabs wrote: April 12th, 2021, 7:35 pm
Nuclearjungle wrote: April 11th, 2021, 8:26 am Yep, looks just like mine did. Ectolabs changed the code a lot, while also changing the lights to 3 neopixel jewels linked together, and attaching some of the parts to different pins. Changing it took 15 minutes maybe. I don't understand why, but the servos never worked properly in the original configuration, and then worked flawlessly with Ectolabs' mods.

Ectolabs definitely earned this: :crunch:
Thank you :-P

Interesting... I didn't realise my mods had fixed servos that are otherwise twitchy with the original code. Mine were okay, but maybe it's just luck of the draw. The edits I made were to ensure there was no interference when using NeoPixels instead of the individual LEDs the original code was meant for.

However, I'm thinking now that I may be able to create another modified version of the original code for the standard hardware, so no rewiring required. This would involve keeping everything the same, but incorporating my servo specific edits which adds a new function that makes sure the servos are only attached to the code for the short time needed for door opening and closing.

I will have a look at that as soon as I get the chance and update here. It might not work, but probably worth a try before buying replacement components.

Dave
that' s would be wonderfull :) ! From my side nearly spent 2 days on it again, learned a bit with arduino lol, still not convinced by the result, however, there are lot of people around there who had the same problem with original code, still have not the skill to work on code, but started to learn how to read it. A few person obverved that servo shouldnt have to work between opening trap sequence and closing trap sequence but the original code make the servo work all the way. they also spoke about conflicts and the fact that the music player shield could make interferences if i understand correctly.

Again sorry for my bad english (it isn't my nativ language). yesterday i decided also to buy 2 new servos as one isnt working now (became hot and stopped).

Everything else works for now so i unplugged all te other options and only working on thoses at the moment checking my soldering/resoldering if needed. Pretty sure it isn't the problem but need to check and verify each option.

If i find anything i ll post and let you know as i'm sure that topic could be usefull in the future :)
#4948108
Pasting my answer/update to ecto lab on facebook guys...hope it could help some of you :) !

"Hellooooooo there !!!
Well here is for the update... I was so desperate that i did damage the shield board that i ordered a new one plus some servos. I heard also that using nanopixel with changed code could help solving problems so i ordered some too. At last i have ordered new servo that arrived today. I checked again all my soldering, opened all my éléments and don t have any problem with soldered parts... After that i tried the new servos. Worked perfectly the first time (one servo is moving a bit at starting sequence but nothing crazy compare to what i had before). So problem come from the servos (both) . I know it s nobody fault but i never thought servos would be the problem, and thought it was my fault somewhere the way i built everything ( it s a big first for me). So i spent more money on new shield/ nano etc where i should ve only bought a new set of servo. I m just hoping the new ones LL be compatible with the structure of the trap now. Anyway be aware that type of mg90s can be a problem. Big thank you to you and the GB community for giving support really. Because of the nano arriving thrusday i may try the ectolab config if not too difficult... I keep in touch for sure but wanted to give update asap.
Again sorry for my bad english"

Edit : Sean did sent a partial refund straight after seeing my post, even before i ve been able to say anything. Thanks to him even if it wasn t the purpose of that post...what a journey, to be continued in the next few days on the build !?!
#4948212
A couple of people have asked me about how the NeoPixels in my mod should be wired up, so here is a little explainer...

Firstly, I used 3x Adafruit NeoPixel Jewels - these feature separate white LEDs in addition to the RGB and are available in different versions which have different white temperatures. I went for the 'Cool White ~6000K' version as this provides the brightest light - perfect for not looking into the trap ;)

Each Jewel is clearly labelled on the back so you can see where each connection is made:

Image

Luckily, NeoPixels can be chained together, so the wiring is nice and simple. Here's a diagram of what I did:

Image

Take a longer yellow wire and solder one end of the wire to the end of a 470ohm resistor (these have the colour bands: Yellow, Violet, Brown, Gold). It doesn't matter which end of the resistor you solder as they work in either direction. Then cut a small section of yellow wire and solder it to the other end of the resistor. I recommend covering the resistor and the soldered points with heat shrink tubing at this point. Now solder the loose end of the short yellow wire to the "Data Input" pin on the first Jewel. Feed the wire into the pin from the bottom of the Jewel and solder on the top.

Now cut two more yellow wires, each about 4" or so. Don't make them too short otherwise there wont be enough slack for them to wrap over the speakers inside the trap. I made this mistake originally. With one of these wires, connect the "Data Out" pin on the first Jewel, to the "Data Input" pin on the second Jewel. Repeat this with the second wire - "Data Out" on the second Jewel to the "Data Input" on the third Jewel.

With the red power wires, you need three sections together as the Jewels only have one power pin each. Make one long red wire and two shorter 4" wires and twist the bare ends as shown in the diagram. Solder the combined wires to the "5V DC Power" pins on each Jewel.

Similar story with the black ground wires, except you don't need to splice them together. Connect one end of the longer black wire to one of the "GND" pins on the first Jewel. Then solder one end of a 4" black wire to the other "GND" pin - connect this to one of the "GND" pins on the second Jewel. Repeat this to connect the third Jewel.

You should now have them wired up as you can see in the diagram. This leaves you with the three long wires connected to the first Jewel. The red is connected to the power rail on the repurposed breadboard (this is detailed in Sean's guide) and the black is connected to the ground rail on the breadboard.

The yellow wire needs to be connected directly to the Music Maker shield. This is where I had to deviate from the guide a bit. I had success by connecting the yellow wire from the Jewels to Digital I/O pin #5. I think Sean has one of the servos on pin #5, so this needs to be moved. I had success by moving the servos to pins #9 and #10:

Image

And here's it is inside the trap with the rest of the rat's nest using my modified LED holders. Rather than solder the three wires directly, I've cut the long wires and put a 3-pin JST connector in the middle so I can disconnect it at any time. Sean uses this technique for everything else, so you should have a spare connector that would have been used for the LEDs he originally intended.

Image

And that's it! You just need to upload my code (provided on page 1) to the trap, making sure you have the additional Arduino libraries installed and have the more accurate SFX files I've provided on your SD card.

Hope that helps :)

Dave
Last edited by EctoLabs on August 1st, 2022, 1:12 am, edited 2 times in total.
countspatula, Nighty80 liked this
#4951246
ImperialWalker wrote: July 1st, 2021, 11:51 am Just a quick question if you don’t mind. Do you recall the paint you used on the Banjo? Thanks.

It's a PlastiKote Metallic spray paint. Doesn't even mention 'brass' anywhere on the can (apart from the colour of the cap) but that's what it's listed as on various websites. It sits right above the real brass nut on the Hammond box and it's still a really good match. Not sure if it's available outside the UK though.
#4951259
EctoLabs wrote: July 1st, 2021, 12:14 pm
ImperialWalker wrote: July 1st, 2021, 11:51 am Just a quick question if you don’t mind. Do you recall the paint you used on the Banjo? Thanks.

It's a PlastiKote Metallic spray paint. Doesn't even mention 'brass' anywhere on the can (apart from the colour of the cap) but that's what it's listed as on various websites. It sits right above the real brass nut on the Hammond box and it's still a really good match. Not sure if it's available outside the UK though.

Perfect. I think they sell that in Canada. Thank you.
#4951973
Another question for you, Ecto.

You used a spray can Silver to chip back to. It appears to have worked.

What are your thoughts on using acrylic like Vallejo aluminum?

I'm at the stage where I need to put on the aluminum coat, but I am concerned that the acrylic might not stick to the primer as well as I would like. It is going to be a bit of a disaster if I go to chip the trap up and the metal coat comes off and it digs to the primer.
#4951986
Never used that, so hard to say. I had good results with Rustoleum Metallic silver which is a pretty inexpensive bog-standard silver rattle can. I don't think it's all that important to get the most realistic metallic looking finish as you're only going to see tiny little bits of it on the edges.

The key for me was not to scratch through the black top coat too hard. I used a blunt piece of scrap 3d printed plastic to gently scrape until the silver appeared. Less is definitely more when it comes to the weathering I think.

I guess if you're paranoid, you could do a clear coat over the silver before adding the black to give you an extra layer's buffer.
#4953156
Hey, I've started with the electronics, but I'm having some troubles.

Using the stock code, it feels like the light strobe goes longer than the door close. Is this intended?

The blue lightning sound is also not synced with the LEDs... hmm

Also, when I use this code (trying to randomize the capture sounds), the sounds work perfect, but the strobe is delayed until the sound finishes and then it goes whacky and stops.
Code: Select all
 if(sfx==true){
      musicPlayer.stopPlaying();
      musicPlayer.setVolume(30, 30);
    char trackName[ 13 ];
    int group = 4;  // can be 1 to 10 for your code
    sprintf( trackName, "trk%02d_%02d.mp3", group, random( 8 ) );
    Serial.println( trackName );
    //  output will have the form "trk09_07.mp3", assuming the random number was 7
        musicPlayer.playFullFile( trackName );
Any thoughts?
#4953159
I think I solved the one issue. The "playFullFile" made it play the entire thing before the LEDs could fire.

So that is solved. The end of the LED sequence and the blue lightning are still not correct though. They seem delayed by a few seconds meaning the strobes are still going when the door close sound happens and the lightning sound happens a few seconds before the blue LEDs fire.

And... it seems I have one green neopixel that turns on on startup... what the what?
#4958074
radi_jp wrote: October 24th, 2021, 9:00 pm I'm having the same problem.
Have you been able to improve it?
ImperialWalker wrote: August 2nd, 2021, 9:57 pm Using the stock code, it feels like the light strobe goes longer than the door close. Is this intended?

The blue lightning sound is also not synced with the LEDs... hmm
There are a lot of timing problems with the code at the moment. My mods were just written over the top of Jeremy's original so it's far from optimum. I have recently been writing code for my thrower build which has more things going on at once (lights, sound, motors, buttons, etc) but runs beautifully in sync with no delays. My plan is to do a complete rewrite for the trap soon which I will share here once done.
radi_jp liked this
#4958138
radi_jp wrote: October 25th, 2021, 6:33 am Thank you. I think I can make a great trap thanks to you. I am looking forward to the new version of the code.
Well, Sean and Jeremy did most of the work, so it's not all down to me, but that's nice to hear :)

I've just looked through my updated code, and, yes there are still a number of 'delay()' commands which are messing everything up. Lockdown afforded me a lot of time to learn better Arduino programming, so I can definitely do miles better.

I'm also working on a new quiet smoke system to eradicate the noise of the current one. Things are looking promising at the moment, so hopefully will be able to have smoke after the capture sequence as I originally wanted , without the noise ruining the effect. More details soon....
radi_jp liked this
#4958700
I ordered Sean's trap kit and I have all the wiring done. I am working on the programming and nothing is working. I turn it on and the speakers beep, the light bar flashes and that's it. I don't know if I messed something up while trying to program it or it's something else. Wondering if anyone else ran into these issues. Thanks for any help.
#4960952
@EctoLabs @countspatula

Hey Guys, Thanks so much for all the hard work, especially Sean and Jeremy.

I thought I would add a bit of my contribution for the 28 segment bargraph, I made an open source version for the wand last year and think it would work quite well on the trap as well!

https://github.com/MikeS11/GB_Bargraph_Sequence

Once I get everything printed im going to start playing around with the electronics :), I just ordered some stuff from Seans Etsy and I have a bunch of neopixel jewels hanging around from building my wand / pack. Though you should check out my more recent addition to the pack with the Afterlife lights :)https://github.com/MikeS11/ProtonPack

Cheers everyone!
Uniform Tips

It does rain frequently here in London, but not to[…]

Hasbro Ghostbusters

It could have been more accurate (barrel and[…]

The yellow parts are raw 3D prints, unsanded and u[…]

Sorry, I hadn't seen any of these replies. Either […]