jab Posted January 22, 2014 Report Share Posted January 22, 2014 (edited) Hello from Taiwan - Not sure if this is the right forum, but here goes: Had a MBHP_Core board made up by a local pcb prototyped and had an LPC1769 sent over from singapore. Started putting things together and noticed that some of the pins on the LPC don't line up with the board, see attached. Is this a known issue with the Rev C LPC's? Update 2014.10.16: Embedded Artists did document the change: To Our Valued Customers, The LPCXpresso LPC1769 board has changed revision to 'rev C' (from 'rev B'). The change was made during October 2013. In 'rev C' the 19 holes in the middle-right position of the board (see picture below) have moved 50 mil in the direction of the arrow. Kind Regards, Embedded Artists AB January 29, 2014 Link: Product Change Note - EA-XPR-003 Edited October 16, 2014 by jab Quote Link to comment Share on other sites More sharing options...
ilmenator Posted January 22, 2014 Report Share Posted January 22, 2014 (edited) Oops, that doesn't look good - glad we are moving on to the next generation ARM cores... You can fix that by using wires, but I wonder how they can just change the layout from rev to rev? Also, I'd double check that the pins have actually just moved physically, and don't carry different functionality altogether (like in: they moved the pins to protect the user from bad mistakes...). Edited January 22, 2014 by ilmenator Quote Link to comment Share on other sites More sharing options...
Shuriken Posted January 22, 2014 Report Share Posted January 22, 2014 I looked on the Embedded Artists Site there is no mention there of a Rev C. The link on the page still goes to Rev B schematic. Maybe it's worth sending them an e-mail. Quote Link to comment Share on other sites More sharing options...
TK. Posted January 22, 2014 Report Share Posted January 22, 2014 Wow! No information can be found about this board in the internet. It seems that it has been produced in the same year like Rev B (2010), but nether went into mass production (I guess for good reasons... due to the incompatible pinning). Please let us know about the comment from Embedded Artists Best Regards, Thorsten. Quote Link to comment Share on other sites More sharing options...
smashtv Posted January 22, 2014 Report Share Posted January 22, 2014 Grrrrr....... Rev B does fit: I will add a compatibility warning to the shop. Best regards Tim Quote Link to comment Share on other sites More sharing options...
Hawkeye Posted January 22, 2014 Report Share Posted January 22, 2014 Hmm, Could it have been a cheap clone? Which shop did you order it from? Many greets, Peter Quote Link to comment Share on other sites More sharing options...
jab Posted January 23, 2014 Author Report Share Posted January 23, 2014 Thanks for the replies. I've been looking for information on the Rev C board for a few days now. I thought that surely this is a known issue... I didn't buy from embedded artists, shipping to Taiwan is 32 euro which is a little pricy and doesn't include customs fees. So I checked the product page and distributor list at NXP.com, and they listed Element14 as a distributor for Asia. I noticed that they have some connection to Farnell so all seemed legit. Any tips on how check the authenticity of the board? What's interesting is that the latest data sheet from NXP doesn't even list "C" in the board revisions... link I'll send a message to embedded artists and what they say. I can take some pictures tonight when I'm back from work. Quote Link to comment Share on other sites More sharing options...
SMIDIRIN Posted January 23, 2014 Report Share Posted January 23, 2014 Hey i ended up with 2 of these after a long wait for my backorder.. first one i had already soldered the female headers so it always wants to pop out! The second one i attached all the headers together and soldered at an angle so it doesnt look perfect but works fine.. pinout seems ok. Two new working seqv4. If it had not worked i would've taken pics and posted sooner! Mine were from mouser. Quote Link to comment Share on other sites More sharing options...
jab Posted January 23, 2014 Author Report Share Posted January 23, 2014 Hey i ended up with 2 of these after a long wait for my backorder.. first one i had already soldered the female headers so it always wants to pop out! The second one i attached all the headers together and soldered at an angle so it doesnt look perfect but works fine.. pinout seems ok. Two new working seqv4. If it had not worked i would've taken pics and posted sooner! Mine were from mouser. So I'm not the only one. Thanks for chiming in. I figure i'll end up doing something similar, but it's so weird that there isn't documentation of this change... I registered at embedded artists, but it seems like they basically send you to nxp for support. I put up a post there as well. Quote Link to comment Share on other sites More sharing options...
jab Posted January 25, 2014 Author Report Share Posted January 25, 2014 Rev C change confirmed by Embedded Artists: (From lpcware.com forum) Hi, The LPCXpresso boards undergo some revisions during their lifetime. The LPC1769 board is now at rev C. The difference between rev B and C is that the position of the holes in between the 27-pos edge holes has changed slightly. This was to fit the board with the LPCXpresso Motor Control board rev B. Kind Regards, Anders @ EA team (the producer of the LPCXpresso boards) Seems like a completely undocumented change... I requested additional details, but not sure if they will come back with anything... Quote Link to comment Share on other sites More sharing options...
Shuriken Posted January 25, 2014 Report Share Posted January 25, 2014 Jeez, that sucks. Quote Link to comment Share on other sites More sharing options...
sneakthief Posted January 25, 2014 Report Share Posted January 25, 2014 Another nail in the coffin for the LPC17 - all hail STM32F4 Discovery :P Quote Link to comment Share on other sites More sharing options...
TK. Posted January 25, 2014 Report Share Posted January 25, 2014 If EA did this to sabotage the high volume sales caused by MIDIbox folks, they achieved a worst case marketing effect -> customers just switch to the competitor! ;-) Best Regards, Thorsten. Quote Link to comment Share on other sites More sharing options...
Hawkeye Posted January 25, 2014 Report Share Posted January 25, 2014 My left small toe somehow tells me, that MIDIbox would have switched to the competitor anyways, but at least now we can blame the original manufacturer :-D. / what on earth were they smoking to make such a redesign decision? :-) Quote Link to comment Share on other sites More sharing options...
jab Posted January 27, 2014 Author Report Share Posted January 27, 2014 My left small toe somehow tells me, that MIDIbox would have switched to the competitor anyways, but at least now we can blame the original manufacturer :-D. / what on earth were they smoking to make such a redesign decision? :-) Very frustrating, I decided to go with the LPC17 version since it looked like a quicker path to a completed project. In hindsight, I wish I went the STM32F4 route... Quote Link to comment Share on other sites More sharing options...
borfo Posted March 22, 2014 Report Share Posted March 22, 2014 (edited) I got a rev c from Mouser recently too for a V4l that I built... I soldered the header pins on the LPC before I realized that they didn't line up properly with the core lpc17 pcb holes. Had to bend and force the pins into their spots. I've been having some trouble with the build - some LEDs aren't working properly, and I'm having trouble getting ethernet to work. I'm wondering whether either issue might have something to do with the LPCXpresso not fitting right, and maybe some of the pins aren't making good contact. I think that if I had fit the LPCXpresso to the PCB before soldering either side of the header pins, I could have made it fit, but I didn't realize there was an issue until I had already soldered them onto the LPC1769. Is a CORE STM32 PCB expected anytime soon? I didn't want to perfboard my first midibox project, but maybe I'll give that a shot. ...it's a drag too, since I bought 2 LPC1769 boards, planning to build a seq v4 once I built the v4l. But I guess I can just build STM32 boards if I have trouble with the CORE LPC17s, since the frontpanels are separate anyway. Edited March 22, 2014 by borfo Quote Link to comment Share on other sites More sharing options...
ilmenator Posted March 22, 2014 Report Share Posted March 22, 2014 Actually, I don't think the situation is that bad for most users: if you are not planning to move the LPC module to a different core board ever (and most will probably never do that), then you can just use the header pins for the two parallel pin rows as described in the instructions, and use wires (cut resistor legs) soldered in a one-to-one fashion between the remaining pins (those that don't physically match). Quote Link to comment Share on other sites More sharing options...
borfo Posted March 22, 2014 Report Share Posted March 22, 2014 The fit is not too far off. It's tough to make it fit if you solder to the LPC before fitting the pins into the PCB, but I think you could make it fit easily enough by just fitting the header pins into both the PCB and the LPC before soldering either side. I just ordered my V4 parts the other day, when I get those parts I'll try that and let you guys know if it works. Quote Link to comment Share on other sites More sharing options...
ilmenator Posted March 22, 2014 Report Share Posted March 22, 2014 The fit is not too far off. It's tough to make it fit if you solder to the LPC before fitting the pins into the PCB, but I think you could make it fit easily enough by just fitting the header pins into both the PCB and the LPC before soldering either side. That isn't any better than what you ended up doing. Male and female headers fit very tightly together, at least the tolerance is much smaller than 1,27mm (which seems to be the offset?). Quote Link to comment Share on other sites More sharing options...
borfo Posted March 22, 2014 Report Share Posted March 22, 2014 (edited) The row of holes has not moved that far on the rev c board. Header pins fit pretty tightly, but there is always a little bit of play before they're soldered - i think there would be enough play to allow the LPC to fit on the pcb. It wasn't easy to bend all the pins in the moved row to fit exactly after soldering, and the connections are not great. I think it would be a lot easier to fit the parts together before soldering - if anyone's reading this before building a Core module with the new LPC, try fitting the parts before soldering. Edited March 22, 2014 by borfo Quote Link to comment Share on other sites More sharing options...
ilmenator Posted March 22, 2014 Report Share Posted March 22, 2014 The row of holes has not moved that far on the rev c board. Well that's strange because from the picture that jab uploaded in the first post of this thread, it looks like we have an offset of exactly 1,27mm for pins 1-19: Looking at this, I think you are not actually talking about these pins, but that you soldered the two rows of headers onto LPC and core board separately and tried to stack them afterwards. That is not recommended practice anyways. So it seems like you are referring to something different than what this thread was originally about? Quote Link to comment Share on other sites More sharing options...
borfo Posted March 22, 2014 Report Share Posted March 22, 2014 (edited) No, I'm talking about the same thing, but I had assumed that the row of pins along the top of the LPC were the pins that were moved, not pins 1-19. But obviously looking at the picture, it's pins 1-19 that moved. It's hard to tell looking at my board, since it's all soldered together at this point. Regardless, having just assembled the board with the rev c LPC, I think that anyone trying to assemble with a rev c should try putting the pieces together before soldering, and then soldering them in place. They won't be able to fit the header pins all the way into the holes, but they will fit and the connections will work without a bunch of rewiring. It would be easier to fit them that way than it was trying to fit them after they were soldered into the LPC. Certainly, there's no harm in trying that anyway. If they don't fit, then you can try another approach, since you won't have soldered anything. You're welcome to disagree, of course, if you like, and your wire-with-cut-resistor-legs method would also work. Edited March 22, 2014 by borfo Quote Link to comment Share on other sites More sharing options...
ilmenator Posted March 22, 2014 Report Share Posted March 22, 2014 Regardless, having just assembled the board with the rev c LPC, I think that anyone trying to assemble with a rev c should try putting the pieces together before soldering, and then soldering them in place. Yes, that is the recommended method anyways, even for the older rev B. You're welcome to disagree, of course, if you like, and your wire-with-cut-resistor-legs method would also work. Nothing to disagree here! Quote Link to comment Share on other sites More sharing options...
borfo Posted March 23, 2014 Report Share Posted March 23, 2014 Yes, that is the recommended method anyways, even for the older rev B. Yeah, I should have tried fitting them before soldering, but after building a bunch of projects like this it's easy to get lazy and just assume things will fit. Hopefully other people will notice this thread before putting their Core boards together and they won't make the same mistake. In other news, the v4l is a great little sequencer. I'm looking forward to getting the V4 frontpanel PCB whenever they're back in stock and putting that together... Quote Link to comment Share on other sites More sharing options...
jab Posted April 13, 2014 Author Report Share Posted April 13, 2014 (edited) Hey guys, since there's been more discussion about this (and maybe for my own closure) I thought I'd show my solution to this, it's similar to what's been previously mentioned: I don't plan on using this hardware with anything else, but I still wanted to make the parts removable. I tried to find a method to put a bend in the male headers, but I couldn't make consistent bends, so I scrapped that idea. Then I noticed that the pins on the female headers are thinner, and could be formed pretty easily - just pressed the leads at a 30ish degree angle against my workbench, and decided to see if I could make this work So I soldered in male headers to the core module (a few pins only to start), then fit the female headers to the LPC and tried to line everything up - which took a little time but was possible.. There's a gap between the header and the pcb, but in my case the leads were long enough that I still had enough to solder to. I soldered a few pins on the LPC, and then tried removing the LPC with the headers attached - it's a tight fit, but possible. So then I plugged it back in and soldered the rest. Continuity checked out ok for all pins. I haven't tested all the functionality of my core module yet, but so far I haven't found any connection issues. LCD 1/2, DIN/DOUT, and SD card all are working well. Anyways, I'm sure my EE professors would cringe if they saw this, but I'm an ME, so I'm going to use that as my excuse ;) Not the cleanest solution, but so far it's working out, and while I don't make a point of removing the LPC, it's nice to have the option available. Hope this is some help for folks that find themselves with a rev C LPC. Edited April 13, 2014 by jab Quote Link to comment Share on other sites More sharing options...
Recommended Posts
Join the conversation
You can post now and register later. If you have an account, sign in now to post with your account.