1103 replies to this topic

[Onevox]

Greyed button. Good info. I'll use that as indicator. (Edit: I've moved this to the software support thread, since this doesn't seem to be an expansion board issue.)

Edited by Onevox, 21 August 2017 - 12:38 PM.

[marie]

Hi mjr,

 

From my pinscape board my (flashers) output 12 and 13 are not working (TLC5940#1 OUT11 and TLC5940#1 OUT12)... Thus blue from flasher 4 and red from flasher 5...

I looked but the soldering on the Pinscape Board (especially the TLC and flasher part), but it looks okay (First time I did some real soldering btw)... 

 

Can I try to recheck/ resolder some other specific points on my Pinscape board? 

 

Or should I start thinking of a workaround (remap the outputs for the flashers: and use fx output 10, 11 and 14 for one flasher and the other flasher through the power-board - 40ma for the small leds is too much as I put 2.2k  as R5)....

[mjr]

From my pinscape board my (flashers) output 12 and 13 are not working (TLC5940#1 OUT11 and TLC5940#1 OUT12)... Thus blue from flasher 4 and red from flasher 5...

I looked but the soldering on the Pinscape Board (especially the TLC and flasher part), but it looks okay (First time I did some real soldering btw)... 

 

Can I try to recheck/ resolder some other specific points on my Pinscape board? 

 

Definitely check the connections - hopefully it's just a solder joint that didn't quite connect and you can fix it by re-melting the solder there or adding a bit more.  There are so many soldering points on these boards that it's hard to get them all right the first time.  I had to re-solder a couple of points on my own on the first attempt.

 

It can be difficult or impossible to spot a bad joint visually - sometimes the problem is hidden underneath the solder and you just can't see it.  So the best way to test is with a multimeter set to continuity testing mode.  Most meters have a mode that makes an audible beep or chirp when it sense continuity - if you have a mode like that, use it.  If not, you can set it to resistance/Ohms mode.  Good connections should read close to 0 ohms (usually something like 0.1 or so), and bad connections will read "infinity" or something similar.

 

The way to test for bad solder joints is to touch the meter probes to the IC leads at each end of the wire run you want to test.  Don't test at the solder pads, since you can sometimes have situations where the solder is connected to the pad but not to the IC lead, so you can get a false "good" reading if you test at the pad.  If you always test at the IC leads, you can be pretty sure you're testing the circuit all the way through.

 

So let's see...  for flasher output 12: test the following pairs:

 

IC1 pin 11  to   OK3 pin 8

OK3 pin 7   to   JP7 "+5V"

OK3 pin 10  to  JP10 "+5V"

OK3 pin 9  to  IC7 pin 14

IC7  pin 16  to  JP11 pin 7 ("4B" printed on the board)

IC7 pins 4, 5, 12, 13 to  JP10 "GND"

 

Similar procedure for flasher output 13:

 

IC1 pin 12  to  OK4 pin 2

OK4 pin 1   to  JP7 "+5V"

OK4 pin 16  to  JP10 "+5V"

OK4 pin 15 to IC8 pin 3

IC8 pin 2  to  JP11 pin 9 ("5R" printed on the board)

IC8 pins 4, 5, 12, 13   to   JP10 "GND"

 

 

Hopefully it'll just be a couple of bad solder joints and you'll be able to track it down with a little testing.  The other possibility is that one of the IC chips is bad, but I'd think the odds of that would be higher if both connections were on the same ULN2064 or PC847, and in this case they're not.  However, if you don't have any luck finding a bad solder connection, and you installed the chips in sockets, you could try this:

 

- Swap OK3 and OK4 and see if anything changes

- If not, swap IC7 and IC8 and see if anything changes

 

If it's a chip problem, one or both of these swaps should change which ports aren't working.  If you do this test, test each swap independently and write down which ports are working and not working in each case.  We should be able to deduce from that exactly which chips are broken, in which case the fix will simply be to replace those chips.  Or, if you have extras of each chip, you can simply try replacing each of those four chips one at a time and see if the symptoms change.

 

The other possibility is that the TLC5940 chip IC1 has a couple of bad ports.  A couple of people have had bad TLC5940's, but I don't remember a case where just a couple of ports weren't working - my recollection from those cases was total failure where none of the ports on the chip worked.  Most of your ports are working, so I think that at least makes it unlikely that the TLC5940 is at fault.  Doesn't rule it out completely, though, so if all else fails, you can try swapping in a new TLC5940 for IC1.  (Again, I'm assuming you socketed the chip; you can still unsolder it if you didn't, but it's a lot harder, so you can't just casually try a quick swap as a debugging test.)

 

 

Or should I start thinking of a workaround (remap the outputs for the flashers: and use fx output 10, 11 and 14 for one flasher and the other flasher through the power-board - 40ma for the small leds is too much as I put 2.2k  as R5)....

 

Yeah, in the worst case, you can always connect the non-working flashers to free power board outputs.  Hopefully you'll get the main ones working and that won't be necessary.

Edited by mjr, 27 August 2017 - 09:51 PM.

[marie]

Thank you for your extensive and clear reply (again!). I did the measurement, but could not spot any strange behavior. Nevertheless I re-soldered all points of OK3 OK4 IC7 IC8 and IC1, just to be on the safeside. But no solution to my flasher outputs 12 and 13, still dead

 

I did not use brackets, therefor I cannot do the swapping of OK3 and OK4 nor IC7 and IC8. I think that will leave only the last option and connect the flashers otherwise...

 

So for RGB flasher 4 -now flasher output 10, 11, 12- I believe I should change in the Pinscape setup into port 10, 11, 14, no changes in the DOF-config needed

And the RGB flasher 5 I should set in the DOFconfig to port 34, 35, 36 (JP6-1, JP6-2, JP6-3 of the power board) and in the Pinscape config enable 'use gamma correction' for these ports (34, 35, 36)

 

Is this line of thinking correct?

[mjr]

Thank you for your extensive and clear reply (again!). I did the measurement, but could not spot any strange behavior. Nevertheless I re-soldered all points of OK3 OK4 IC7 IC8 and IC1, just to be on the safeside. But no solution to my flasher outputs 12 and 13, still dead

 

I did not use brackets, therefor I cannot do the swapping of OK3 and OK4 nor IC7 and IC8. I think that will leave only the last option and connect the flashers otherwise...

 

That's too bad!  It sounds like you probably do have a couple of bad chips after all given that all of the connections seem good.  So I think you're right that the easiest solution is to just move the dead flashers to power board ports.

 

 

So for RGB flasher 4 -now flasher output 10, 11, 12- I believe I should change in the Pinscape setup into port 10, 11, 14, no changes in the DOF-config needed

And the RGB flasher 5 I should set in the DOFconfig to port 34, 35, 36 (JP6-1, JP6-2, JP6-3 of the power board) and in the Pinscape config enable 'use gamma correction' for these ports (34, 35, 36)

 

Is this line of thinking correct?

 

That sounds like a good plan to me.

[marie]

Redirecting RGB flasher 4 in the Pinscape did the trick for that one. 4 flashers fully operational 1 to go...

 

Noow that every works fine on my mainboard I continued with the next step and connected the power board:

- pins from JP6 Main board to JP2 Power board (SIN to SOUT, SCLK to SCLK  etc).

- connected the PC PSU and 2nd PSU to the power board (I tried theses connectors with my main board as well to make sure they are working) 

 

In Pinscape set up I added 1 power board and (for testing purposes) set all the outputs 34 - 49 to 'use gamma correction' 

 

Because I will not using a lot of high power outputs I only soldered the left part of power board (everything for the outputs for JP6 i believe). Somewhere in this thread I read this was possible I think..

In other words, the right part of the power board is totally empty except for R1, (and C12, IC 12 C11 and C6 next to the PC PSU)

 

When I connect all my leds to JP6 (for testing purposes) none of them is working when I test these ports with the Pinscape configuration tool.

 

 

I had some difficulties with R1 (broke off) So perhaps there is an issue?

Do I perhaps need to solder some parts in the left part?

Or perhaps al faulty TLC940 (which is on a bracket btw )

 

Is there a way to localize the problem by measuring or other means? 

 

Thanks!

[mjr]

 

Redirecting RGB flasher 4 in the Pinscape did the trick for that one. 4 flashers fully operational 1 to go...

 

Great!

 

 

 

Because I will not using a lot of high power outputs I only soldered the left part of power board (everything for the outputs for JP6 i believe). Somewhere in this thread I read this was possible I think..

In other words, the right part of the power board is totally empty except for R1, (and C12, IC 12 C11 and C6 next to the PC PSU)

 

When you say you left the "right half" of the board empty, do you mean you left the IC1 side empty?  If so, that's probably the problem.  The two TLC5940 chips are arranged in a daisy chain, so if you leave IC1 empty, the input data signal to IC2 isn't connected, because the signal travels through IC1 on its way to IC2.

 

There are two fairly easy solutions.

 

The first is that you could install the second TLC5940, along with R2 and C1, so that IC1 is working properly.  You can still leave all of the MOSFETs and PC847 chips on that side of the board empty.

 

The second, even easier, is that you could install a simple piece of wire from IC1 pin 17 to pin 26.  That will directly connect the data signal without a need to install the chip.  If you're certain you're never going to want to install IC1, this is the simpler approach, but I wouldn't recommend it if you want to keep the option open, since you'd have to remove the wire and the excess solder to get the chip socket in later.

 

 

For anyone coming across this in the future, if you want to populate half the board like this (for 16 outputs rather than 32), it's easier to install everything on the IC1 side and leave the IC2 side empty, because of the daisy-chaining order.  If you do that, you won't need the extra jumper wire.

 

It's also okay to omit any of the MOSFETs (and the two resistors that are grouped with the MOSFET) individually.  Each MOSFET controls one output port, so leaving out a MOSFET will simply leave the corresponding output port unconnected.  

Edited by mjr, 02 September 2017 - 05:32 PM.

[marie]

Oops I made the wrong decision then with choosing the IC2 side... Will choose the IC1 side for my next cabinet.

 

Just soldered a bracket on IC1 and connected pin 17 and 26 with a dupont wire, because I don't have another TLC9540 lying around... But I can put in another TLC if necessary. 

 

Because I was not sure what pin 17 and 26 was I tried both:

- just counting: pin3 from the bottom and pin3 from the top on the right side of TLC9540,

- following the schematic you made which starts with number 28 as out0: connecting pin2 from the bottom and pin2 from the top on the right side of TLC9540

 

But still no luck...

 

Is there something I can try next?

[mjr]

Just soldered a bracket on IC1 and connected pin 17 and 26 with a dupont wire, because I don't have another TLC9540 lying around... But I can put in another TLC if necessary. 

 

That's a great solution - makes it easy to put a chip in later if you decide to.

 

 

Because I was not sure what pin 17 and 26 was I tried both:

- just counting: pin3 from the bottom and pin3 from the top on the right side of TLC9540,

- following the schematic you made which starts with number 28 as out0: connecting pin2 from the bottom and pin2 from the top on the right side of TLC9540

 

Here's a diagram:

 

 

The red lines show the two pins to connect by bypass the chip.  For reference, all IC chips with two rows of pins like this have the same numbering scheme.  Look for the little "half moon" shaped notch at one edge, or a little white dot painted on near a corner.  If you look at the chip with the notch or dot at the LEFT side, pin 1 is the pin at the bottom left.  The pin numbers increase to the right along the bottom edge - pin 1, 2, 3, ... 14 in in the case of a TLC5940.  When you get to the last (rightmost) pin in the bottom row, the next pin is the one directly above it in the top row.  Then the numbers continue from right to left across the top row - 15, 16, ... 28.  Put another way, the pins are numbered counter-clockwise around the chip starting with #1 at the lower left.

 

If that looks right, the next thing to try would be to look at the voltages on the TLC5940 pins.  To test voltages, you'll need the power on, of course, so be careful not to short anything when you're probing.  As long as you're only using the PC power supply, though, it's all low voltages that won't be a danger to you.

 

Let's see...  Multimeter in DC voltage mode, black probe to any pin on the OUTSIDE row of the "PWM OUT" connector (nearest the edge of the board).  Red probe on IC2 pin 21 ("left side" of capacitor C2) should read +3.3V, and pin 22 ("right side" of C2) should read 0V.  

 

The data lines are high-frequency signals, so it's hard to get a meaningful reading with a DC meter, but you should see some non-zero voltage on pins 26, 25, 23, and maybe 24 (it'll probably read lower than the others).  

 

You might also want to double-check that the TLC5940 is inserted in the right direction, with the little notch in the chip matching the notch drawn on the board silkscreen.  If you did have a second TLC5940 on hand, I'd try swapping it (since you socketed it) as a quick test for a dead chip.

[marie]

My first option was like your clarifying diagram. So I take that as the base situation.

 

I double checked the direction of the TLC5940, and it was mounted correctly.

 

And then the results of the measurement:

 

On IC2 pin 21 I got 3.27V and on IC2 pin 22 0V... That is the good news

 

But the 26, 25, 24 pins gave 0.00V. pin 23 gave 0.01V...  To me that is o zero voltage

 

I got some TLC5940's from a friend of mine and tried swapping the TLC5940's but no change, also I put the extra TLC5940 in IC1 instead of the wiring, but no succes either...

 

I don't know if it might anything to do with my problem, but I still have some doubts whether R1 is connected correctly, because of the replacement of a broken resistor-leg. Is it possible to do a measurement to exclude my doubt on that (if relevant of course)?

Edited by marie, 03 September 2017 - 03:20 PM.

[mjr]

It sounds like you have something disconnected if you're getting 0V everywhere.  

 

Try this.  Power everything off and disconnect the USB.  But do leave the connection in place between the power board and the main board (the connection from main board JP6/PWM OUT to power board JP2/PWM IN).

 

Now test continuity between the following points:

 

Main board IC2 pin 17  <->   power board IC1 pin 26 - your jumper wire - IC2 pin pin 26

 

Main board IC2 pin 25  <->   power board IC2 pin 25

 

Main board IC2 pin 24  <->   power board IC2 pin 24

 

Main board IC2 pin 23  <->   power board IC2 pin 23

 

Main board IC2 pin 18  <->   power board IC2 pin 18

 

As for R1, you can test if it's okay by measuring the resistance between IC2 pins 23 and 21 - you should see about 10K ohms.

[marie]

Houston we have  a lift-off!

 

The IC2 pin 18 to IC2 pin 18 did the trick.... Did some resoldering and now  it works!

 

Thanks a million for your help and patience!

 

Now it is only hooking up the toys

 

The last part of my cabinet journey is connecting addressable LED's through a Wemos D1 instead of a Teensy. This solution needs some adaption of the DOF R3+ code. At the French forum they are helping me with this. Would be nice when the Pinscape - Wemos combination works.

 

Now time for some testing.

[mjr]

Houston we have  a lift-off!

 

The IC2 pin 18 to IC2 pin 18 did the trick.... Did some resoldering and now  it works!

 

Thanks a million for your help and patience!

 

Now it is only hooking up the toys

 

The last part of my cabinet journey is connecting addressable LED's through a Wemos D1 instead of a Teensy. This solution needs some adaption of the DOF R3+ code. At the French forum they are helping me with this. Would be nice when the Pinscape - Wemos combination works.

 

Now time for some testing.

 

Hooray!  I'm glad to hear it wasn't another bad chip - it's always nice when a little resoldering is all that's needed.

 

I'd be happy to try to merge the extra Wemos code into my R3+ build when it's ready.  In principle, changes for different devices should be easy to merge together since they usually only affect unrelated code.  When you/they have something in working order, send me a pointer to the updates and I'll see if it can be readily merged into my version.

[Onevox]

I'm ready to install IR to my PF and BG TVs, both TVs use the same IR code, and I'm now sure which header pins to connect to.

 

is the proper hookup ...

                                                              IR Emitter             IR Emitter2

JP4 Pin1 (at arrow)_______27R_______/            \  _______/          /

JP4 Pin 2  ________________________________________  __/

 

Thanks.

Edited by Onevox, 19 September 2017 - 04:43 PM.

[mjr]

I'm ready to install IR to my PF and BG TVs, both TVs use the same IR code, and I'm now sure which header pins to connect to.

 

is the proper hookup ...

                                                              IR Emitter             IR Emitter2

JP4 Pin1 (at arrow)_______27R_______/            \  _______/          /

JP4 Pin 2  ________________________________________  __/

 

Right circuit, but you actually want JP4 pins 5 and 6.  Pin 5 is the one labeled "IR+" on the silkscreen (tiny print above the header).  Here's a picture in context for orientation:

 

 

Note that the resistor is already on the circuit board - R11 - so you don't need another external resistor.  Just connect the IR LEDs directly.  Your wiring diagram is correct with the emitters in series.  The polarity matters, so make sure the emitters are connected like this:

 

JP4 pin 5 "IR+"  ------   (+) IR Emitter 1 (-)  -------- (+) IR emitter 2 (-) -------- JP4 pin 6

[Onevox]

Once again, you make it simple. And I find a way to overthink it and make it complicated. Thanks again. Yep, I got that resistor there. But bought 10 more. Can't wait to put away my remotes.

[Fusionwerks]

hey.. my first actual forum post....

I just finished all the soldering on my 2 boards, with the exception of R5 and R11. I have read the notes on the build guide, but...

R5 - I'm not planning to use RGB flippers, so should i leave it out, or just install the 2.2k?

R11 - I don't have my displays yet and i don't know if i will even need the IR feature, so same question, can i just leave it out?

[mjr]

hey.. my first actual forum post....

I just finished all the soldering on my 2 boards, with the exception of R5 and R11. I have read the notes on the build guide, but...

R5 - I'm not planning to use RGB flippers, so should i leave it out, or just install the 2.2k?

R11 - I don't have my displays yet and i don't know if i will even need the IR feature, so same question, can i just leave it out?

 

Congratulations on getting the soldering done - it's a big project, I know!

 

You should probably install something for R5, since I think the TLC5940 chips might not be happy without it.  I'd probably install a 680 ohm or 780 ohm there.  That will give you a 60mA or 50mA limit (respectively) on those outputs.  I like having a higher limit there if you're not currently using them, for flexibility in case you do ever decide to add something there.  The 2.2K is fine, too, but it'll limit you to 20mA on those outputs if you do ever decide to use them.

 

R11 can be omitted entirely if you're not using the IR emitter.  And my recommendation would be to omit it, again for flexibility if you do decide to enable it in the future - that way you can select an appropriate resistor size according to how you actually plan to use it.

[Fusionwerks]

Ok, so here comes the NOOB questions that i cant seem to find the answers to...

 

I haven't set up any software yet so i dont know what any of it looks like. I'm planning on doing it this weekend and doing some testing with the buttons and contactors.

 

1. Is there a "standardized" numbering for which button should be on which pin? Or does the DOF tool just take whatever pin you assigned to and put it in the right place? In other words, if i put right flipper on pin 1 and tell  the DOF tool that's where its at, the tables will just work accordingly?

 

2. Contactors... I'm using 24v contactors with a separate 24v PS. The way i understand it is this:

 

All the negative/0v leads from the 24v PS and the PS2 are connected together.

The positive 24v is connected directly to the contactor (or any other toy with the same voltage for that matter)

The negative side of the contactor is connected to the pin on the power board.

When the output on that pin is triggered, the power board actually just completes the circuit and the contactor fires.

 

Is this correct?

 

And again, is there a standardized pinout like above or does it work the same way as the buttons when you set it up in the DOF tool?

 

3. Is there a pin-out of the expansion / power board numbers? I cant seem to find any other than the button numbers on the main board.

 

I promise you guys im really not as dense as i may come off, but im just getting started.  

Edited by Fusionwerks, 29 September 2017 - 09:28 PM.

[mjr]

1. Is there a "standardized" numbering for which button should be on which pin? Or does the DOF tool just take whatever pin you assigned to and put it in the right place? In other words, if i put right flipper on pin 1 and tell  the DOF tool that's where its at, the tables will just work accordingly?

 

Numbering is completely arbitrary.  There's no special golden numbering system that you have to follow, and indeed there's not even such a thing as a recommended numbering.  You're exactly right that you just wire it up in whatever order, and then enter the mapping between port numbers and specific device types into the DOF config tool.

 

The whole point of DOF, really, is to make all of this abstract, so that none of the software even knows *what kind* of hardware you're using, let alone how many output ports it has, or how those ports are numbered, or how the ports are connected to devices.

 

 

2. Contactors... I'm using 24v contactors with a separate 24v PS. The way i understand it is this:

 

All the negative/0v leads from the 24v PS and the PS2 are connected together.

The positive 24v is connected directly to the contactor (or any other toy with the same voltage for that matter)

The negative side of the contactor is connected to the pin on the power board.

When the output on that pin is triggered, the power board actually just completes the circuit and the contactor fires.

 

Is this correct?

 

Yep, you've got it!  That's all correct.