Why mod.. when you can make?

User Tools

Site Tools


blog:spoke_to_soon_....._so_maybe_back_to_z180

Spoke to soon ..... so maybe back to Z180 :(

The first eZ80 adaptor I made took awhile as I was being thorough and am not particularly well acquainted with solder paste and aluminum solder paste templates… but it seemed to work well enough. Blew it up.

The second and third ones, I did using classic drag soldering .. tack the IC down, drop a small bead of solder at the end, and medium-fast drag it across the pins; chances are good you'll get 75% of the pins soldered nicely, and the remainder will bridge.. but not much; hit the pins with flux, lay the solder wick across, and wick out the bridges… a very fast and generally successful approach. (Also handy if you have a hot air rework station so can recover from darned near any situation ;)

Blew those up too.

Right now, I'm not sure whats what. The eZ80 F93 has 8 pairs of power (Vdd/Vss) around it. I put together the (Expensive :/) adaptors, and tested continuity.. no bridges. Place onto breadboard, with just the VddVss (not even crystal on first try) … it went to pull higher amps, so immediately powered down; check continuity and then whathwat - all Vdd/Vss are shorted together. So spent way too much time looking for bridges, cleaning up anything remotely looking funny.. nada. Finally gave up, and made the second (and then third) adaptor.. second took about 20 mins, third took about 10. I'm pretty good at drag soldering.

Same result for all 3 … turn power on, and now they're showing shorts across Vdd/Vss. I'm really not sure what to think … I've not seen before where a short magicly develops, unless you completely fry the sucker. My power supply is set capped to 3.3V (and a couple amps, which is too much :/), so its not over likely they blew up from power spikes or weird stuff. And its certainly weird all 3 had the same cycle of seemign fine, then going to bad Vdd/Vss. This pretty much clears up any risk of bad solder since they would be differently broken in that case.

I checked the Vdd/Vss pins on my Breadboard probably 100 times, and pretty sure its okay.

I've got 1 more eZ80 F93 handy, and only sqare adaptor boards (not good for breadboard.) I do have the hot air station so can pop the guys off and re-use the adaptor pcbs themselves, but right today I'm at total loss.. no point in continuing right now, as I have no idea whats gone wrong. The 3 chips are now flagged 'suspect'. $15 down the tube :/ Not really liking the idea of soldering up the square-adaptor, and then in turn soldering that to a perfboard, and that into the breadboard.. a fair amount of effort (couple of hours really, all told) when no idea what went wrong. Chances are it'd just go again. Bad batch of ICs? bad adapter board? can't be an error in the datasheet for Vdd/Vss … so what could I have done wrong? (I doubt the other things, it has to be my breadboard.) FWIW, the Zilog USB Smartcable blew up at the same time.

Maybe its a hint .. LQFP100 is rather a pain to work with .. not too bad for me since I've got the right gear now, but hard for anyone else who'd want the kit, anyway. So I've always steered away from the eZ80 for this reason, looking instead to classic Z80 or the Z180. But now I've been going through the eZ80 datasheet, and I really want to use it .. those tasty GPIOs and modern peripherals (I2C etc), would reduce chipcount and make for some nice options; the built in flash is cheating a little, but does provide for some handy options.

For now, I'm going to backburner eZ80 .. I can always come back to it later, and maybe I'll get some ideas. But for short term (as I want to get Zikzak schematic nailed down, and start working on board design.. its been long enough, about time to knock out some pcbs!) I think I'll do some serial tests of Z180 to atmega644 (as I/O chip, since Z80/Z180 can't do it themselves like an eZ80 can :/) – this will require nailing down the Z80 toolchain, getting the cart wired in, and adding RAM .. all easy stuff - so I can make a basic echo server say. Once the avr can talk to/from the Z180, can pretty much call it a day and start designing the pcb.

But man, now I really really want to do eZ80 .. wth did I do wrong there :/

EDIT:

OKay, the likely culprit is a combination of things..

  1. I 'recently' (few weeks back) started using a new bench power supply; when I checked its outputs, it seemed ok
  2. normally (but not this time) I run power supply through a regulator circuit
  3. normally (but this time I forgot) I put caps across power pairs (Vdd/Vss in this case)

Someone suggested that some (cheap?) power supplies spike when powered up after _awhile_ off, but do not spike when poiwer cycled recently. Turns out mine _does this_ .. so when set to 3V out, it spikes to 10V on power cycle after being off for 15mins or more.

So it would seem likely that when I went up for lunch and checked if the circuit worked after.. poof, got popped right there. A regulator likely would have solved that, and decoupling caps across power pins may have caught it; so normally my boards protect themselves, and also used a different (perhaps better) power supply in the past.

Well, sheeeeeeit. LESSON LEARNED. Do not trust the power supply!

Discussion

Enter your comment. Wiki syntax is allowed:
 
/home/skeezix/public_html/zikzak.ca/zikzak.ca/dokuwiki/__data/pages/blog/spoke_to_soon_....._so_maybe_back_to_z180.txt · Last modified: 2017/11/11 02:58 (external edit)