• State Resolved
  • Locked Locked
  • Replies 8 replies
  • Subscribers 18 subscribers
  • Views 871 views
  • Users 0 members are here
Support feedback
Options
Options
Related

This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

CC430, converting SMD based schematic to thruhole

Jayanth Acarya
Prodigy 120 points

Hi,

Let me start by admitting that what I am trying is probably foolish, since I am quite RF noob, and don't have the expensive tools (spec-analyzer, freq-gen etc.) common used (I believe) for RF experimentation/design troubleshooting. However, I'm going on a tiny bit of hope that this is probably doable and I might learnt a bit in turn.

I've got an ez430-Chronos (433MHz) and bunch of MSP430G2 valueline launchpads. What I would like to do is to use the ez430 schematics, and convert the SMD component to thruhole, to create a simple wireless mesh. I plan to pick up the equivalent thruhole components, corresponding to the SMD ones, but I am not sure equivalent thru-hold inductors for all of those used exist. So my question is --

1. Does someone think that this is a bad (/stupid) idea, and unlikely to work ? (e.g. breadboard itself might be too noisy for RF ?!)

2. Is someone aware of any online tool that helps convert BOM between SMD and thru-hole ? Any software that does such a thing ?

thanks,

Jay

  • 0
    TI__Guru**** 317180 points

    The inductance and capacitance used in the reference design for CC430 are small (in the range a few nH or pF) and need around 5% tolerance. With a breadbord you will have large inductance/ capacitance on the connector points which you don't have too much control over and hence not very suited for RF work. Without typical RF instruments to help in the design and debug phase the probability to get something up and running is best if you follow the reference designs closely. It is possible to use a CC1111 dongle as a simple spectrum, typical to check if you actually sends something on the frequency you want.

  • 0 in reply to TER
    Prodigy 120 points

    Thanks TER, and thanks Tor-inge (for verifying).

    Do you suppose taking a veroboard or a custom-made (toner-transfer type PCB) might have better board-RC properties, than the breadboard ?

    That way I can replicate the Chronos schematics by a slight proportional magnification, and attempt thru-hole. I've read elsewhere however, that in case of RF design, length, thinkness of traces and even distances between them play a crucial role. However, I wonder if 433MHz is 'high-enough' a frequency to warrant such care/caution, or those are expected in the GHz ranges only ?

    Anyhow, I think I will try out at least 1 to see how it goes. Worst case, might try with larger SMD devices. My limitations are a very basic & spartan home lab -- hobby grade.

  • 0 in reply to Jayanth Acarya
    TI__Guru** 106705 points

    Hi Jayanth,

    I would highly recommend trying custom-made (home made) PCB and smd components over breadboard or veroboard. You could try to replicate the reference design of the RF circuit as closely as possible, but use 0603 components instead of 0402. 0603 components will probably give you some RF loss, but you should at least be able to solder them by hand even without a microscope. It will also be way better than through hole components. You should get a very thin soldering iron tip, and use lots of flux when you solder.

    Good luck!

    /Fredrik

  • 0 in reply to Fredrik K
    Prodigy 120 points

    Thanks @Fredrik_K. Have done a small project with 0402 components in past, but I'd hoped that I'd not have to repeat it again. I've used a 10x lighted magnifier for soldering, and while it was painfully slow (very slow), the bigger hassle was finding the SMD components in small numbers, and getting the DIY board done. Could be a case of bad / cheap toner though.

    Just realised that, finding 26MHz and precise 32.768kHz crystals is no easy game, and found another thread which says that the crystals used are from MicroCrystal for the 32.768kHz part and Ansen Quartz for 26MHz ones.

    While I am trying to figure if they have distributors in India, but wondering if it is mandatory to use solely those crystals, or can other one be used ? For my limited volume use importing those in small quantities would be extremely expensive. Are there any known alternatives ? Are there known reasons why, working with local XTALs, which are possibly not top-notch, may be a bad idea ? Maybe I am pressing the limits of engineering sanity, but trying to make-do with what is easily available, else will need to fall back on plan-B.

     

  • 0 in reply to Jayanth Acarya
    TI__Expert 3595 points

    The exact crystal is not important, the specifications on ppm error, drift and the like are.  Find crystals that conform to those values and you are good to go. 

    http://articulationllc.home.comcast.net/~articulationllc/sm0402.htm Is a cool article with some hints for doing 0402 soldering.  I have had to do a lot of it here in the lab, the hard one is going to be the CC430 as the paddle (the large pad that is under the die) has to have a good attachment to ground vias or you will get very poor performance. Other things to note, PCB material and thickness is very important to keep the RF characteristics the same as for the reference design.  Again, you can go 0603, and make it a bit easier, and often you can ask manufacturers for some samples for very small runs, but the big distribution houses have all the components in stock.

    But as to through hole designing, not going to work.

    Chris

  • 0 in reply to Chris M
    Prodigy 120 points

    Thanks @Chris_M. That explanation is very helpful.

    Had completely overlooked the need to solder the paddle. Not aware of a way to do that with soldering iron, so I guess only way is reflow soldering. I was hoping to avoid stencils and the reflow-oven (very expensive proposition in my part of the world).

  • 0 in reply to Jayanth Acarya
    TI__Guru**** 317180 points

    To make a prototype you don't need a reflow oven but you need some type of hot plate ( something like this: http://www.testequipmentdepot.com/weller/accessories/whp1000.htm) Place a small block of aluminium on the hotplate and the PCB on top of that and you are able to solder the paddle. You need a model where you know the temperature to avoid thermal damage.

  • 0 in reply to Jayanth Acarya
    TI__Guru** 106705 points

    You could even do the soldering with a hot air soldering iron. No guarantees, but you should be ok :-), I've done it several times without destroying the chip.

    /Fredrik