 Notice. New forum software under development. It's going to miss a few functions and look a bit ugly for a while, but I'm working on it full time now as the old forum was too unstable. Couple days, all good. If you notice any issues, please contact me. |
Forum Index : Microcontroller and PC projects : Cooking chips....
Page 1 of 2   |
|||||
| Author | Message | ||||
Grogster Admin Group  Joined: 31/12/2012 Location: New ZealandPosts: 9308 |
This has been touched on before in other threads, but when I got some 100-pin QFP chips for my own colour MM clone, they were vacuum sealed, and the label went on about having to bake them under certain circumstances. This is something that I have never had to do with other chips, and in fact the MicroMite chips that I also got in the same package from Microchip Direct, they make no mention of needing to bake them. So, how come some chips obviously need baking, and most of the others do not? Why is this needed at all in the first place? Why not make the chip so it does not NEED baking, if you see what I mean.... I am guessing that some wise member(s) here will enlighten me as to what it is all about.  Smoke makes things work. When the smoke gets out, it stops! |
||||
| WhiteWizzard Guru  Joined: 05/04/2013 Location: United KingdomPosts: 2817 |
I was just about to post the same topic - you beat me to it by a couple of hours! The other day I ordered some FT232RL (FTDI) chips from Mouser (cheaper bulk price from US than local suppliers) and guess what - they came in a 'humidity detector pack'. Instructions to bake them if any non-blue dots showing on a piece of card that was inserted in the pack too. If baking was required it said the shelf-life without baking was 168hrs maximum. 
I have used this particular FTDI chip in many projects over the years and NOT ONCE has it ever come with this warning before. Previous suppliers include RS, Farnell and other local UK suppliers - like I say, none warning that baking is required. Looking at the data sheet (FT232RL) I am now more confused than ever. It does mention 168hrs pre-baking but not sure (as it is a bit ambiguous) as to if it refers to this particular device I am using (data sheet covers various package options - I am using the SOIC) I had specifically chosen to use this chip in my 44-pin MicroMite module because of its robustness over the years, and of all the trouble free operation. IF you look at other FTDI chips in their USB-to-UART range they DO state that they need baking (too much effort as far as I am concerned!). Zonker will relate to this as the DFN FTDI chip he has incorporated into his design states 168hrs life pre-cooking (mentioned on his thread by myself and others). SO PLEASE is there anyone out there that knows the FACTS? Do we need to 'cook our chips' or not? 
Grogster: I have used the PIC32MX795F512L in all my MaxiMite clones WITHOUT baking AND without any troubles. My alarm clock (based on TFT MM) has certainly worked for more than 168hrs 
We await any experts out there to guide us further . . . . 
For everything Micromite visit micromite.org Direct Email: whitewizzard@micromite.o |
||||
BobD Guru Joined: 07/12/2011 Location: AustraliaPosts: 935 |
Here is a good answer. If you reflow solder them the moisture in the chip can explode. Hand solder is OK. |
||||
| WhiteWizzard Guru Joined: 05/04/2013 Location: United KingdomPosts: 2817 |
Hi Bob, Ok, that puts my mind at rest - Thanks. In summary, any moisture absorbed into the IC's plastic body/package (and different plastic have different amounts of absorption), then this moisture can 'explode' as it boils IF you were to use a reflow process during assembly. Reading the link you sent, there is one posted 'answer' that leaves me a bit confused still: Surely if you bake the board as suggested right at the end, the package will 'explode' during the baking process due to all the moisture absorbed while the unit was in the field?? So for all of you out there with one of my 44-pin MicroMite modules; DON'T put them in the oven or the FTDI chip on the underside may 'explode'! Likewise with Zonker's 28-pin MicroMite Module and his DFN chip. YOU HAVE BEEN WARNED
For everything Micromite visit micromite.org Direct Email: whitewizzard@micromite.o |
||||
MicroBlocks Guru Joined: 12/05/2012 Location: ThailandPosts: 2209 |
FTDI chips are know for this problem. It is one of the reasons why cheap chinese usb-serial converters are unreliable. It is not only when you reflow solder them also when soldering it with an iron it can damage the chip inside. It can go undetected for a long while and than that tiny crack gives way and result in a failure. Microchip does not have that problem. They use a better substrate that does not absorb moisture. I guess the FTDI factory does not have the right 'mix' from their packages. I refuse to use them, even when they are functionally very good. Often i buy parts to have them in stock and i don't want to worry about how long it can be exposed. As i am located in Thailand i need to take extra care as the moisture content of the air is often above 90%. Added to that i need to pick, pack and ship parts that can take anywhere between a few days to two weeks forcing the recipient to bake the chips. Unacceptable. Microblocks. Build with logic. |
||||
| WhiteWizzard Guru Joined: 05/04/2013 Location: United KingdomPosts: 2817 |
Hi TZA, Is the MCP2200 your preferred RELIABLE choice over the FTDI range? If not, have you any other recommendations? Thanks for your input . . . For everything Micromite visit micromite.org Direct Email: whitewizzard@micromite.o |
||||
Zonker Guru Joined: 18/08/2012 Location: United StatesPosts: 761 |
While I was fitting the DFN parts on the PCB, I was only soldering one pin on at a time. I think very little heat was getting to the whole chip. I was using liquid flux and I was bringing the iron with a small touch of molten solder on the tip to the pad. the total time for the reflow was about a tenth of a second, so I don't think the whole DFN package was getting hot at all... Just my 2c... |
||||
| robert.rozee Guru Joined: 31/12/2012 Location: New ZealandPosts: 2350 |
sorry, but someone has wildly misled you. most chips are supplied from the original manufacturer either in a vacuum sealed package and/or packed with a desiccant to absorb moisture. either way, if the chips have been exposed to the atmosphere (and hence moisture) for longer than a certain period of time (usually several days), it is prudent before putting through a reflow process to bake them at a certain temperature for a given period of time. this temperature is well below the reflow temperature, and definitely will not cause the chips to explode! the need for baking applies ONLY to devices that are to be then put through a reflow oven. if not baked there is a SMALL risk of the devices being damaged during reflow, but it is very much not a certainty. however, when you stack up the small probabilities across a product containing many devices, the chances of product failure can rapidly mount. also, remember that even a 1% failure rate at the end of a production line is a big problem that needs to be addressed. if you plan on removing a device from a PCB and wish to reuse it, it can be argued that you should following the same baking process to ensure any moisture is driven off if you intend to use a reflow process for removal. this is a situation that would very rarely arise. in any manufacturing or service environment most devices removed from a PCB would be discarded. reusing a device that had been removed would be considered very poor practice. baking parts has nothing to do with the quality of the components, quality if ingredients, or anything else that can be attributed to a particular parts manufacturer. in a manufacturing environment all chips that have been exposed to the atmosphere longer than a few days are routinely pre-baked before being put through any reflow assembly process. even a chinese manufacturer making cheap USB bridges will follow the above rules. these PCBs are populated by very expensive pick-and-place machines on very real production lines. the people who run such lines know how to handle the parts correctly. one must also bear in mind that these days the majority of electronic products are produced in china. brands that are traditionally considered 'high quality' such as sony, apple, nokia, etc. so, the bottom line is, if you don't have a reflow oven then you can ignore the warnings - they do not apply to you. and even if they did, the chances are that the sky will stay securely attached in place! rob :-) |
||||
| MicroBlocks Guru Joined: 12/05/2012 Location: ThailandPosts: 2209 |
Mmm, did i wildly misled myself...... :) Probably we have different experiences. The problem is when you have a small electronics parts shop, storage of these devices is very difficult. Sure the manufacturer packages them good. Once out of the package though it becomes a hassle. Repacking them one by one is a nightmare but the only way to store them long term, and even then the moisture creeps in. Time consuming handling compared to other parts. It is not a problem when after unpacking it is soldered on a pcb within a few days. When selling the FTDI chips (separate or in a kit) to hobbyist often the parts/kits are unpacked, the moisture warning often read after the fact(!), left to open air for prolonged times and finally mounted with not so wonderful soldering skills manage to kill proportionally a lot of them. Thailand's humidity is also a contributing factor to the failure rate. No such problems (99% less failures) with the MCP2200, which is a similar (often even cheaper) product. So having the choice between an FTDI or a MCP2200 it is pretty easy. Hope this answers your question @WhiteWizard, but again if the parts are put on the pcb within the timeframe there is nothing to worry about. My circumstances might not be yours. I need to stock them and order higher volumes (often 50-150) to get better prices. So in short, the sky was falling for USB-Serial parts until we changed to another manufacturer. With small profit margins paying attention to the MSL level pays off. Now we try to not stock parts with MSL 2a to 6. Microblocks. Build with logic. |
||||
| WhiteWizzard Guru Joined: 05/04/2013 Location: United KingdomPosts: 2817 |
Thanks for your replies gents. However . . . So all the FTDI chips that I have purchased from all suppliers APART from Mouser were supplied in 'non-protective' packaging (in terms of humidity). Therefore how do I know how long these have been 'exposed'? Yes I do put most parts on a PCB within days of receipt but going back to TZA's quote if the parts are put on the pcb within the timeframe there is nothing to worry about[/I} then do I have something to worry about? Physics tells me that water boils at 100C so I assume moisture is similar temperature. A soldering iron is hotter than 100C so basic logic says I have a problem?? Hope you follow this logic but please tell me if in your opinions where my logic falls down (assuming it does!) Sorry to drag this out but I don't yet have a clear, non-contradictory answer. Thanks again for helping me get my head round this bazaar situation . . . For everything Micromite visit micromite.org Direct Email: whitewizzard@micromite.o |
||||
| MicroBlocks Guru Joined: 12/05/2012 Location: ThailandPosts: 2209 |
[quote] So all the FTDI chips that I have purchased from all suppliers APART from Mouser were supplied in 'non-protective' packaging (in terms of humidity). Therefore how do I know how long these have been 'exposed'? [/quote] You bake them as you have no way of telling how long they were exposed to the open air and under which circumstances (humidity levels, temperature etc). [quote] Physics tells me that water boils at 100C so I assume moisture is similar temperature. A soldering iron is hotter than 100C so basic logic says I have a problem?? [/quote] Your logic is sound. Chips with many pins do get hotter then 100C when soldering by hand and as such can cause micro fractures. These fractures will happen. As you are soldering the leads chances are higher that those fractures will be around those leads, all the way back to the wires connecting the pins to the chip. With your soldering skill you might have no problem. The risk is a lot lower but is still there. For me one defective chip eats away profit for 5-10 chips. Something i can not afford. Hence my decision to not use them. All i can say is that that decision payed off. If FTDI used a better material then the MSL level would be 1, as are microchips products. Microblocks. Build with logic. |
||||
vasi Guru Joined: 23/03/2007 Location: RomaniaPosts: 1697 |
Hi WhiteWizzard, See also how many Maximite clones were handmade without any problem. This also demonstrate that Microchip make a great USB peripheral (at least as great as the ones from the 8bit/16bit microcontrollers) [quote=TZAdvantage]No such problems (99% less failures) with the MCP2200, which is a similar (often even cheaper) product. So having the choice between an FTDI or a MCP2200 it is pretty easy. [/quote] I can also say that a Microchip microcontroller with USB peripheral it is a reliable choice for an USB-to-Serial converter (you have access to the firmware sources and that is fantastic as you tune them exactly for your product). So, my logic says: a Microchip microcontroller is reliable from all points of view (soldering by hand, reflow, hardware functionality, firmware functionality, industrial usage, hobby usage which in my opinion is the most aggressive against the microcontroller). But I guess no one from here is able to offer you any kind of warranty. Hobbit name: Togo Toadfoot of Frogmorton Elvish name: Mablung Miriel Beyound Arduino Lang |
||||
| WhiteWizzard Guru Joined: 05/04/2013 Location: United KingdomPosts: 2817 |
We are getting there However, two points remain unsolved:
1: Grogster received his QFPs for his MaxiMite clone with instructions on how to bake his chips. Assuming he is using a PIC for his clone then why did he get the baking warning (if MicroChip supposedly use suitable chip plastic packaging)? 2: I thought the MCP2200 didn't require any drivers. Whether it does or doesn't, my Win7(64bit) doesn't see it when plugged in so initial reaction is they are not very user friendly (but I do understand all your pro arguments for using them from the above posts). Baking cakes is so much easier . . . For everything Micromite visit micromite.org Direct Email: whitewizzard@micromite.o |
||||
| MicroBlocks Guru Joined: 12/05/2012 Location: ThailandPosts: 2209 |
It depends on the packaging. Not everything from Microchip is MSL 1. The pic32mx150f128d TQFP is specified as a MSL 3 (don't have them in stock), The 28 pin soic and dips which i have in stock are MSL 1. I use element14 to quickly check the MSL level. They mention it with almost all chips. You can find it also on the Microchip website, but finding it for a particular packaging is difficult. Here is such a document: http://ww1.microchip.com/downloads/en/DeviceDoc/PCN_CYER-21G YVK501_Qual%20Report.pdf Just to have some comparison. The PIC32MX795F512L is MSL 3. Not much reports of failure when hand soldered. It probably is because of the size of the chip and the number of pins, there is more thermal mass. The FTDI chip, especially the QFN is very small and difficult to solder which results in longer soldering, higher temperature and higher failure rate. The PIC32MX795F512L 64 pin TQFP i bought here locally few months ago came in a plastic bag (was not a direct order from Microchip). So i had to bake them before use. It is all part of doing it the right way. Microblocks. Build with logic. |
||||
| MicroBlocks Guru Joined: 12/05/2012 Location: ThailandPosts: 2209 |
About the driver for the MCP2200. You can find that here: http://ww1.microchip.com/downloads/en/DeviceDoc/MCP2200%20Wi ndows%20Driver%202013-02-21.zip Also interesting when using this product is this: http://ww1.microchip.com/downloads/en/DeviceDoc/APPLICATION% 20FOR%20SUBLICENSE%20TO%20USB%20VID%20revised%2012110.pdf You can then have your own. "WhiteWizards MicroMite found!" :) Microblocks. Build with logic. |
||||
| WhiteWizzard Guru Joined: 05/04/2013 Location: United KingdomPosts: 2817 |
So does anyone have any tips for me as to how to 'bake chips' preferably based upon their own experiences & knowledge. Do you need a dedicated piece of equipment? Or can I use the kitchen oven? And what temperature do you bake at because my brain keeps telling me that the baking process will cause the boiling/'exploding' issue anyway ?!?
I really appreciate any help because I am just about to order my next batch of PCBs for the MicroMite Module. I am prepared to hold off so I can make a change from FTDI over to MCP2200 but only if I consider the 'exploding' risk actually exists. Thanks yet again for all responses . . . For everything Micromite visit micromite.org Direct Email: whitewizzard@micromite.o |
||||
| vasi Guru Joined: 23/03/2007 Location: RomaniaPosts: 1697 |
I will be very curious to know when Microchip started to deliver low quality microcontrollers... many things/changes happened in the last two/three years... Hobbit name: Togo Toadfoot of Frogmorton Elvish name: Mablung Miriel Beyound Arduino Lang |
||||
| Grogster Admin Group Joined: 31/12/2012 Location: New ZealandPosts: 9308 |
Crikey - lots of activity on this thread.
I thought I might get a couple of replies, but a lot more interest in this subject then I thought, so it would seem, and differences of opinion. That's all good, cos it gives us something to chat about. Being that all chips from the very early ones, to today all seem to use the same kind of plastic encapsulation, I was not aware of any absorbent nature of such plastic. I thought that plastic, by it's very nature, repelled moisture. Oh well. From this thread, I have come to understand and move forward with the following concepts: 1) Baking not required if you are hand-soldering, just don't let the chip heat up TOO much - give it a chance to cool off as you go. 2) Consider the cracking of packages to be a non-issue UNLESS you have experienced that situation yourself personally or you are wave-soldering. TZA mentions that he has, and also the humidity over there where he is. Things are pretty dry-ish around NZ most of the time, and Australia is probably even dryer depending on where you are. WW might have cause for concern unless he lives in a nice dry area of England! 
No, seriously, I don't imaging that WW will have issues either, and my HUMBLE opinion is that it is over-reaction to start swapping the FDTI chips at this point. WW says he has used lots of them, so if they WERE going to fail for him, one would imagine that they would have done so for him by now, generating faulty units and building up unreliability in his mind, of using FDTI chips. Question for WW - how many FTDI chips have you EVER used, and of those, how many have you found to have failed? Smoke makes things work. When the smoke gets out, it stops! |
||||
| WhiteWizzard Guru Joined: 05/04/2013 Location: United KingdomPosts: 2817 |
Hard to put an exact number but between 150 and 200 FT232RLs used on my own PCBs over the last 3 to 4 years. Not aware of any failures; yet! So am I right in my understanding that baking chips only dries out the plastic in preparation for high temps as used in reflow? Or put another way, baking does NOT seal the plastic package once and for all? For everything Micromite visit micromite.org Direct Email: whitewizzard@micromite.o |
||||
| WhiteWizzard Guru Joined: 05/04/2013 Location: United KingdomPosts: 2817 |
. . . and I still have the same questions as of yet unanswered: Do you need a dedicated piece of equipment? Or can I use the kitchen oven? And what temperature do you bake at because my brain keeps telling me that the baking process will cause the boiling/'exploding' issue anyway ?!? 
For everything Micromite visit micromite.org Direct Email: whitewizzard@micromite.o |
||||
| Page 1 of 2 |
|||||
 Print this page |
