next topic »

[Reply #30 on:]

Just like Lakes says, you need to use the pin labelled DTR on the CP2102 to wire to the reset pin (pin 1 on the ATMEGA328P). All the CP2102s I've been passing on to people are exactly the same as Lakes, and have a right-angled male pin header manually soldered in place there. This really has to be detailed better on our website, sorry.

There's a picture of a minimal Shrimp on a breadboard which uses a modded CP2102 and matches Lakes' schematic exactly (apart from the LED) in the Shrimping Safari article at http://blog.safaribooksonline.com/2012/07/03/time-traveling-with-old-laptops-and-arduino-compatibles/ It looks like this...

   

However, there's a bunch of commentary in this forum thread which suggests some additional protective components which should be added for stability if you're trying to make a general purpose Arduino-compatible on a breadboard. The argument is that the components are so cheap that they're not worth trying to eliminate them, which we'd have to agree with, given our bill of materials at http://shrimping.it/blog/bill-of-materials/

We've tried to incorporate all of these insights in the latest layout which looks like this...


Alternatively, if you haven't access to a soldering iron, get the timing really good on hitting the reset button (which you need to wire with a pull-up resistor as per the http://shrimping.it schematic). Hit the reset button just after Compiling and just before Uploading. You have about a microsecond Seriously, though, it's not that tight. I can generally get it every time.

[Reply #31 on:]

If anyone wants to send me a (really clear) picture of their busted breadboard layout I can try to figure out what's wrong. Equally if there's anything else I can help with by way of parts or what have you, then get in touch especially if you're in the North of England. There's bound to be some way I can swap some bits with you for sanity checking whether things are roasted.

Contact me via this forum or http://shrimping.it If you're happy for me to share your pictures it would be good to have a resource of examples which need various fixes.

I'm relying on Bill Westfield's totally amazing Optiloader for flashing chips from Mouser. It does have a weird behaviour after I've flashed them. If I try to reflash them it seems to report they're busted, but they're actually fully functional. I think it's just a counterintuitive error message. I should file that on the Optiloader github when I get a chance.

I did have a DOA CP2102 but only one out of the 50 or so which I've bought, and I don't know what people did with it when workshopping, so they could have really zapped it somehow. Put it this way, I'm not going to try and get my money back

[Reply #32 on:]

I have that exact same one, mine set-up in very similar way, just put some headers instead of cutting and rerouting the DTR. Still doesnt work for me thou!

Have you tried doing a Loopback test with it?
Also its easy to get the Tx and Rx wires swapped around when using it with a breadboard...

[Reply #33 on:]

@bromatt when you say "it doesn't work for me" what exactly are you experiencing.

If you can't reprogram it even manually with a (well-timed) reset button push then I'd stop investigating the DTR pin and look at the rest of the circuit, especially the TX vs RX as Lakes points out.

That one (inverted serial pins) foxed me with an entire workshop full of people when I'd done the diagram they were all working to the wrong way round [facepalm].

[Reply #34 on:]

Does anyone have a view on the stability implications of placing the large electrolytic cap on the opposite side of the circuit. I'm guessing it's only going to respond over a relatively long timescale anyway compared to the smaller caps (which are there for rapid response).

Having it on the other side would be really useful given the left hand side is really packed when using just the 5 rails of a breadboard/stripboard...


However, it's not worth doing if it defeats the object of the cap altogether.

[Reply #35 on:]

@bromatt when you say "it doesn't work for me" what exactly are you experiencing.

If you can't reprogram it even manually with a (well-timed) reset button push then I'd stop investigating the DTR pin and look at the rest of the circuit, especially the TX vs RX as Lakes points out.

That one (inverted serial pins) foxed me with an entire workshop full of people when I'd done the diagram they were all working to the wrong way round [facepalm].

Hi,

Done the loopback test, tried swapping the TX/RX still have the same problem!

My problems are detailed in another post on here... http://arduino.cc/forum/index.php/topic,121571.0.html along with a photo of the board...

Regards

Matt

[Reply #36 on:]

Just had a thought! Could it be the bootloaded chip? I mean I was under the assumption that the supporting hardware on the Uno was nothing more than a USB/Serial adaptor anyways so there shouldn't be any difference should there?  And what is the routine to put the 328P into programming mode? is it just a matter of timing on the reset? or is there something more complex?

Matt

[Reply #37 on:]

If you mean doing a manual reset, yes you have press that button at just the right time, which is a pain..

Otherwise, you need to look at the AtMega side now, check your wiring, is that crystal really oscillating?

The only way to be sure is to Nuke it from Space... I mean use a known working chip...

[Reply #38 on:]

I tried using a uplink and piloting the drop ship remotely....actually.....

Crystal is ok - I had the blink test program running on the board (i programmed the chip in my Uno just to make sure).
Chip is ok - I programmed it in my Uno just to make sure. (is there an echo in here???).


Matt

[Reply #39 on:]

@bromatt we should probably continue discussion of your layout/wiring/config issue on the other thread http://arduino.cc/forum/index.php/topic,121571.0.html but if you're still stuck after all the helpful advice, after I've finished my deadlines from 11th September, I can post you a full tested set of parts, then you can post your duff layout back on the breadboard for me to figure out what's wrong.

That way I gain a breadboard, (really useful for workshopping here in Morecambe), and maybe a 328P-PU (I'm working with the cheaper 328-PUs)!  At worst, I'll have swapped a good CP2102 for a fried one, but that's fine by me. If you're heading for Preston Raspberry Jam or OSHCamp we could cross paths there, or you could join us at a future http://shrimping.it workshop in Morecambe.

Let me know if that's interesting to you.

[Reply #40 on:]

Thanks for the offer, i may just do exactly that if I continue to have these problems, discussion moved to my other thread!

Matt