Free PCB the other way around

IMG_20141120_153706-600

Last week Sjaak got a free PCB and a small SMD antenna from a user called Claude on the esp8266.com forum . Claude made the board using Dirtypcbs.com. As Sjaak is a huge fan of QFN chips he went straight to his mancave and started to assemble the board.

esp8266_1_600

Close-up of the soldered QFN chip under a USB microscope.

There were a couple of issues (as always with a first revision of a PCB) so a couple of patchup wires on the  bottom:

IMG_20141126_151655-600

Unfortunately he didn’t have a FTDI2232 (an other QFN chip!!) to solder so dig his old and trusty buspirate to poke around the UART interface.

IMG_20141126_154554-600

Here are the steps to select a baudrate of 74880:

HiZ>m 3
Set serial port speed: (bps)
1. 300
2. 1200
3. 2400
4. 4800
5. 9600
6. 19200
7. 38400
8. 57600
9. 115200
10. BRG raw value

(1)>10
Raw value for BRG (MIDI=127)

(34)>51
Data bits and parity:
1. 8, NONE *default
2. 8, EVEN
3. 8, ODD
4. 9, NONE
(1)>1
Stop bits:
1. 1 *default
2. 2
(1)>
Receive polarity:
1. Idle 1 *default
2. Idle 0
(1)>
Select output type:
1. Open drain (H=Hi-Z, L=GND)
2. Normal (H=3.3V, L=GND)

(1)>2
Ready
UART>
UART>
UART>(1)
UART bridge
Reset to exit
Are you sure? y

Then the bootloader blabla:

 ets Jan  8 2013,rst cause:2, boot mode:(3,7)

load 0x40100000, len 24236, room 16
tail 12
chksum 0xb7
ho 0 tail 12 room 4
load 0x3ffe8000, len 3008, room 12
tail 4
chksum 0x2c
load 0x3ffe8bc0, len 4816, room 4
tail 12
chksum 0x46
csum 0x46

Thanks to Claude for his free PCB ;) More info about his board in Topic on ESP8266.com.

Get yourself a buspirate here.

Join the Conversation

10 Comments

  1. ” As Sjaak is a huge fan of QFN chips” – Proof that Sjaak is really an alien, come to spy on Earth before the invasion – What man doesn’t quiver at the site of QFN? lol ;)

    1. I’m sorry it is relaxing for me to solder those chips ;) Once you know how to do it it is rather easy:

      1) hot air and a medium sized soldertip. Apply a bit of solder using the soldering iron on the exposed pad, use generous flux on the pad, then heat the exposed pad on the pcb and the chip with the hot air till the solder is molten, place the chip and keep it warm using the hot air. then let it cool down. It should be aligned ok by the surface tensiomn of the solder. flux the chip even more. If everything is aligned ok drag solder using the solderiron and a big blog of solder (make sure not to damage the pads). the result should be like the pcitures above.

      2) solder paste, stencil and an oven (or hotair). apply a solderpaste using a stencil and reflow it using an oven. IMHO it is cheating but work reasonable well. ;)

      1. LOL – I was just joking, poking a bit of fun. I haven’t done QFN yet, but I regularly do approx 250 pin QFP with a 0.5mm pitch. I’d think they would probably be harder than QFN, as the leads will bend if you accidentally touch and drag the leads while dragging, doesn’t take much to bend them.
        About time I got a microscope I think, maybe one of those Chinese Binocular ones.

      2. Sjaak, when you say “on the exposed pad” are you talking about a the center, thermal pad, or just one of the pin pads ? Either way I’m having trouble figuring out how the chip aligns to the footprint. If you’re talking about the thermal pad, it seems like it could align at any arbitrary rotational angle. If you’re talking about a single edge/corner pad, I don’t know why the part wouldn’t rotate around that pad. I could see the surface tension of two or more pads keeping it from spinning around a single fixed solderpoint.

        Once you get ready to slap on the chip once it’s molten, do you keep the hot air going on it the whole time? Are you really coordinated enough to do this with just two hands or do you make use of one of your third or fourth alien hands? Are you looking for proper alignment from just the top, the side, or both (with your alien telescoping eye ganglia) ?

        You can’t drag-solder the BSTs if they don’t wrap-around the sides of the package… right?

        After you’ve drag soldered the pins, do you ever reflow the assembly, just to make sure that all of the solder joints are hot (that is, not cold).

        Do/must you only use tacky flux for this, or can you get by with the more watery stuff?

        Have an eBay-like source for the tacky flux (if you find that it’s critical for some of your work). I’ve only found tacky flux through regular distributors from ChipQuik, and it’s a little pricey.

        ++thx

      3. @sleepwalker3: no problem. Although I have an alien stigma now :/

        @ewertz: i ment the center pad (which is called exposed pad in most datasheets). The flux i use is the one I scored in akihabara ( http://dangerousprototypes.com/2011/12/13/sjaaks-shop-a-holic-trip-to-akihabara-electric-town/ ). Basicly a rosin with acetone/IPA. You place the pcb on the bench, heat it with the hotair till it melts. Use your other hand to place (first heat it up a little) the QFN chip. keep it heated a bit and then cool it down. So no alien hands involved :D

        If interested i can make a vid (if I can find a suitable camera)

  2. Do you float the QFN on enough solder to get it up high enough off of the board so that when you drag solder, it wicks under, or do you think that you’re just attaching to the terminations on the side of the package?

    My few attempts to do this usually don’t result in the chip being planar with the board. I have enough solder under it so that it seems like it could wick under. But if that’s unnecessary, I certainly won’t bother because it takes a fair amount of solder to bridge the gap from the lands to the terminations, which is problematic (enough solder that the chances of bridging go up, and very difficult to solderwick back out because of the low clearance).

    Here we often say,”video, or it didn’t happen” — but I’ll spare you our obnoxious Earth humor, our alien soldering overlord.

    1. @ewertz – The video is above in the text under “huge fan” just under the top pic.
      Note that almost all of the tacky paste flux from ebay/dx/ etc. is fake. http://ultrakeet.com.au/write-ups/fluxInfo It probably works OK, but I prefer the real stuff.

      @Sjaak – Well all us DP insiders knew you were an alien anyway ;)
      I’m sure it will be forgotten soon. I think I might rather solder a QFN than a fine pitched TQFP, as least there is no leads to bend, though I expect it’s harder to detect (see) shorts under the package. I might give it a try sometime.

      1. @Sleepwalker3 Thanks for pointing out the pointer to the vid, which I had seen when it came out, but had forgotten about. Watching it again reminded me that a post-drag-soldering reflow step is, in fact, A Good Thing(tm).

        Contrary to current belief, Sjaak’s BGA soldering exercise fail at the end of the video proves that he’s not an alien. I’m not sure if I’m disappointed knowing this now or not — I found the possibility intriguing.

  3. Any thoughts as to whether or not backside heating to reflow is a better approach than topside, even though it takes longer? Is there not much risk of toasting the chip before the board gets sufficiently (heat) soaked? Or is the expectation that these are hack prototypes and we don’t care too much?

    1. I dont have had any problems with my approach. I can garantee it wont damage the chip, but for prototype it is ok IMHO. If you want to take no risk please read the datasheet carefully for the maximum ratings. There should be a value for max solder time or a reflow profile.

      When concerned invest in a reflow oven. :)

Leave a comment

Your email address will not be published. Required fields are marked *

Notify me of followup comments via e-mail. You can also subscribe without commenting.