Replicator 2/2x Button PCB IC??

[Scott Lichtsinn]

Does anyone know what the little 6 leg IC is on the back of the Button PCB?  Its labeled U1, i have been trying to source a replacement but can't figure out what it is.  All of mine seem to have various ID's on the top, 6J3, 6J4, 6J9, not a consistent marking on the top and i can't nail down what the heck they are.

What i do know is that i have a half dozen of these button pcb's that don't work at all, even with a brand new keypad on a known working front interface board.  But if i take this IC off a working button pcb and put it on one of my non-working ones then presto chango that button pcb now works fine.  So it appears that U1 is the reason these half dozen boards don't work, i just cant source a replacement for them.

Also i have been working on a fix for the keypads that turned out good, i found a supplier on eBay that sells 4mm conductive rubber pads for repairing rubber keypads, i carefully cut off the old pads on that back of a keypad that wasn't working and glued on the replacements and now that keypad works great.  So i'm going to order some more of these little rubber pads and fix all my dead keypads for spares finally!

[Rich Webb]

Have you traced out the board to make a new schematic, with the unknown U1 as a black box? Once it's all laid out it may become obvious what it's supposed to be. I could give it a go if you would post good pics of the front and back (the one up top is kind of fuzzy and overlaid with glare).

Naturally, MBI scrubbed the original from Thingiverse.

On Wednesday, April 4, 2018 at 4:12:21 AM UTC-4, Scott Lichtsinn wrote:

Does anyone know what the little 6 leg IC is on the back of the Button PCB?  Its labeled U1, i have been trying to source a replacement but can't figure out what it is.  All of mine seem to have various ID's on the top, 6J3, 6J4, 6J9, not a consistent marking on the top and i can't nail down what the heck they are.

What i do know is that i have a half dozen of these button pcb's that don't work at all, even with a brand new keypad on a known working front interface board.  But if i take this IC off a working button pcb and put it on one of my non-working ones then presto chango that button pcb now works fine.  So it appears that U1 is the reason these half dozen boards don't work, i just cant source a replacement for them.

[Jetguy]

No, they never posted the 2 and 2X series. Only open source was Replicator Dual AKA Replicator1

That said, it's most likely the serial to parallel shift register.
That has stayed the same in the design even to the 2 series- however different chips and layouts are used, but the logical function is the same.

The mainboard sends a serial data stream for the LCD. This serial data is converted to a 4 bit scheme in parallel supported by the character LCD module.
Technically, it's 8 bits, but 7 are used, 4 for actual "data" nibbles, and the other 3 are the control lines required to shift the data into the LCD module.

[Jetguy]

Again, just so we all get on the same page.

The ribbon cable from the mainboard is carrying a few signals.
#1 is the serial data stream for the character LCD
#2 is the SD card data lines for direct reading of the SD card through a voltage shifter (another ic on the board) because SD cards are 3.3V IO logic and the mainboard mega processor is 5V logic.
#3 is all the buttons are actually individual wires back to the mainboard.

What we have here is a voltage event likely blows the 2 or 3 main functions on the LCD module.
#1 it likely blows the shift register serial to parallel data chip so the LCD no longer works
#2 It can also blow the very LCD itself because it too won't tolerate the high voltage accident greater than 5V
#3 It can also blow this voltage shifter used on the SD card lines and so SD card reads may be broken or not work.

So even if you repaired other board just by replacing the shift register, you may uncover other damage and function issues with SD card.
Just saying, if in the repair business, check everything because a voltage event blowing one has every likel hood to blow the others too.

[Rich Webb]

Can't be the shift register -- too few pins.

The board has 5X (R-R-C) with the C pads unpopulated. That works for an interface to the five buttons.

A quick browse through Digikey doesn't turn up a likely candidate for a matching diode, resistor, or capacitor array. Too few pins for a small micro to read the buttons and yack on I2C.

And, of course, the honkin' big pads on the bottom. Unpopulated bypass caps?

Curious...

On Wednesday, April 4, 2018 at 9:39:50 AM UTC-4, Jetguy wrote:

[Jetguy]

Can't be. Like I said, I know for a fact the 5 button are not going i2C or anything else, at best, could be a TVS?
In fact, that's the most likely answer, a TVS diode for static protection. As such, the circuit would work without it.

[Gary Crowell]

They added a lot of TVS to the Rep2, so could be.  If that's the case, and there's nothing else wrong with the board, it should work with the TVS removed entirely.  Tracing the connections would reveal a TVS pretty easily.

On Wed, Apr 4, 2018 at 6:25 PM, Jetguy <vernon...@gmail.com> wrote:

Can't be. Like I said, I know for a fact the 5 button are going i2C or anything else, at best, could be a TVS?

On Wednesday, April 4, 2018 at 7:09:10 PM UTC-4, Rich Webb wrote:

Can't be the shift register -- too few pins.

The board has 5X (R-R-C) with the C pads unpopulated. That works for an interface to the five buttons.

A quick browse through Digikey doesn't turn up a likely candidate for a matching diode, resistor, or capacitor array. Too few pins for a small micro to read the buttons and yack on I2C.

And, of course, the honkin' big pads on the bottom. Unpopulated bypass caps?

Curious...

On Wednesday, April 4, 2018 at 9:39:50 AM UTC-4, Jetguy wrote:

No, they never posted the 2 and 2X series. Only open source was Replicator Dual AKA Replicator1

That said, it's most likely the serial to parallel shift register.

--

----------------------------------------------
Gary A. Crowell Sr., P.E., CID+

Linkedin  Elance  KE7FIZ  Things RocketryCNC

[Rich Webb]

Looks like Jetguy nailed it.

An SMF05C (multiple vendors) TVS looks like a perfect fit. (Some guesswork involved to the SIL connector.)

[Jetguy]

Don't give me too much credit, but yes, once I saw the picture, understood exactly what IC on what exact board, the I could narrow down the offending part type.
Rich, thanks for doing the further research as I simply did not have spare time to dig into this one.

Being dead honest, I don't know why when I first looked at this, I did not see the board  picture?? Classic google groups problem.

"What i do know is that i have a half dozen of these button pcb's that don't work at all, even with a brand new keypad on a known working front interface board.  But if i take this IC off a working button pcb and put it on one of my non-working ones then presto chango that button pcb now works fine.  So it appears that U1 is the reason these half dozen boards don't work, i just cant source a replacement for them"

So when I read that, and let's be honest, was probably tired, my first thought was that moving a working IC from a good board to a bad board, well if that was the shift register- then the LCD would work and the would say the board worked.
So really, my first reply was just me not doing my proper full look and investigation into the question.

FWIW, from an advice perspective:
The original replicator LCD boards didn't have TVS diodes. The mainboard didn't either. I've got dozens and dozens of examples of those surviving still today, working just fine. In fact, in all my reading ever, I've not come across an instance where a static shock was known to kill the mainboard because of no TVS on the button panel. I'm not saying it could not happen, but again, I think of all the forums I'm in, the fact FlashForge and Wanhao and CTC and others still sell and use this basic schematic sans any form of TVS tells me it's not an absolute "must have".
So IMO, while sure, having a good workign TVS diode on a control panel where a user could static shock it is not a bad idea. At the same time, this is repairing "dead" boards just trying to get them to work on the cheap anyway, so again, I'd remove the blown TVS and not blink and eye at replacing it (just leave it unsoldered).

[Scott Lichtsinn]

Wow thanks for all the info guys!!

Sorry for being so late to get back here and comment, its been crazy busy here since i posted this question.  

Yes, thats for sure what it is, a TVS Diode.  I wasn't thinking about that either.  And i can confirm, merely removing that part restored the function of the dead button board's that i have.  So bingo!  I will most likely order some replacements, because why not, cheap and easy to replace for me.

I don't see the Rep 2/2x having a major issue with a static discharge to the front panel, since the metal portion of the SD Card slot is far enough below the top surface of the plastic front of its enclosure that your not going to touch it or really get close enough for a discharge to it inserting the card, a discharge to the metal surround of the LCD panel would be more likely but again, i will agree with Jetguy, i haven't seen a failure on even the Rep 1's from this, i have personally zapped my Rep 1's a few times, usually the only thing that happens is the display goes garbled and you reboot the printer and it comes back, sometimes you can tap the center button and it will refresh the display and come back as well.  But no damage was caused.

I have since moved all my printers to a counter covered in ESD mat thats properly grounded, and its become habit to touch the mat before touching the printers to discharge myself, on top of that they live in a basement room with only concrete floor so it hasn't been a big issue but there is times i will build up a charge on the chair or before walking in there so i still self discharge before touching, its basically a habit for me anyway since i spend a lot of time at my rework station, always thinking ESD.

[Scott Lichtsinn]

You are correct Jetguy, the Carrier PCB that the LCD and Button board plug into also had damaged IC's on it, the LCD still functioned, but the button inputs did not even with a known working button board, also the SD card couldn't read.  I ended up replacing the two IC's under the button board, the 74AHC125D and the TPD6E001RSFR, total cost of repair per PCB was just under $1, that restored the SD card functions as well as the button input.  

These front panels came out of printers that all had failed Mightyboards, and this was before i got a hold of them, so i don't know details of the events that killed them.  I do know that i have 18 Rev H boards now that all had failed Heater and Fan MOSFETS, so i have become really good at 

swapping those little buggers.  I had two that had failed resistors in the Thermocouple inputs, i'm assuming a TC contacted a Heater and blew those, oddly it didn't kill the ADS IC, just the SMD resistors between it and the screw terminals for the TC's.  So that was a quick repair and the board 

went on happily printing.

So far the most common issues i have had, blown MOSFETS, blown TC inputs, and a few with blown 5v supplies.  I still haven't gotten an ID on the 5v regulator control chip, but swapping it from some other boards that had other failures but working 5v supplies brought those back to life as well.  

It's been a fun adventure into the Rev G and Rev H boards.  But i still have a stack of Rev E boards to fix for customers so i better get my butt back to work!

[Alex Gibson]

Just now in 2020 used this info to fix a Replicator 2 and get it back printing face shields for local doctors. Desoldered U1 to prevent phantom button-press loop which was stopping printing.

Massive thanks for this excellent information JetGuy and all contributors here. Is there still a community around the old MakerBots that I should follow?

Alex Gibson, Edumaker

[Troy McLucas]

There's not a whole lot of activity, but this is the only User Group I know of.  Once you adjust a couple of the little known flaws, the Replicator 2 is still a solid basic printer.  I'm still using mine weekly.  I printed a bunch of face mask ear savers last week for our primary care physician.

-Troy

[Brett K]

There is a Facebook group. Not super active but it is something. 

--
You received this message because you are subscribed to the Google Groups "Makerbot Users" group.
To unsubscribe from this group and stop receiving emails from it, send an email to makerbot-user...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/makerbot-users/e6ccbd42-43fe-4a3e-8d20-af3c139a988c%40googlegroups.com.