- BS2 OEM 4X hardware communication failure!

Mrdavidjenkins

Good Morning,
I am having trouble with a project I am working on.· I want to build a Stamp 2 OEM with 4 switches, a 1302 clock with a CR2025 backup, a Seetron LCD 4 line and a PVN012 photovoltaic relay.· I have built 5 boards and reach a point with the programming where I get a "hardware communication failure!· check serial cable and power".· The OEM board·no longer work. ·Each time, I figure that I have done something stupid and revise the board - improving the board design and my techniques.·

What is odd is that It faills when I am just about finished - after a lot of testing each step of the way and a lot of planning.

And I reach a point where I get the same error "hardware communication failure!· check serial cable and power" just about when I am finished.· In these cases, the VIN reads 14 volts (I am using a spectre 9v·voltage converter that I purchased from Parallax), &·the·Vdd shows 5·volts.

Anyway, I am trying to figure out what I am doing·wrong here.··I test·every·step of the· way.· I improve the design·each time (note that I am using the ·PVN012 to reduce the compexity of the board as an expensive alternative.).· Each part of the project I do makes "Sense" to me and I am using designs that I get off the Parallax boards or parallax books.

By the way, the bad OEM boards all show same: VIN reads 14 volts &·the·Vdd shows 5·volts.· Pictures can be sent if it would be helpful.

Thanks for your help.·

Dave

Philip Gamblin

My suggestion would be to start simply. I'm still not convinced that you don't have a power issue, but we'll get to that later. If possible write the code and test the hardware for each item individually. At that time measure the load current to that hardware module, if possoble. Later you can sum the currents and compare the result to the spec for the regulator. THere is no reason to believe that the ds1302 or the LCD are the problem as they reprsent smallish loads. A backlight on the LCD could be the culprit, the specs on my Parallax serial display show a 100 mA requirement to drive the backlight. I am not familiar with your relay but I googled PV relay. Since it is a high current switching device, I would check it's connections. Additionally, if everything works separately then your code is probably ok UNLESS you have a conflict in software that causes a hardware problem once everything is assembled. Lastly, check all of your wiring with an ohm meter. Loose or intermittent connections can be a nightmare, and breadboards do wear over time. Remember check from component lead to component lead.The worst that can happen is you find out that everything is solid and correctly wired and you will then know where the problem isn't. Good luck. Photos never hurt and ·code and schematics always help.

Mrdavidjenkins

Thanks for the quick response.· Sounds like a good technique for·me and everyone else.·

I built the board in stages, like you suggested.· I got the clock working first.· Then I got the 4 switches to work.· I wrote code to run the switches and did not have the clock code running.· THe switches worked.· THen I ran the clock code and it worked fine.· I installed a 9 volt·LED on the VIN and it worked fine.· I put a 3300 uf capacitor on the board and hooked it up to the vss/vdd.· I disconnected this from the board and did not reconnect because it worried me.· I tested the board and it worked fine. Did not reconnect the capacitor. · At each step, prior to hooking up the power and the board, I tested the circuits and they looked good - I could see no faults or anything.· Lastly, in each case of the 5, I hooked up the relay/transistor to the board and this is where, in each case, the system began to go south.

IN every case, I breadboarded the new·circuit on my little board of education prior to installation.

I was also able to use one of the burnt out OEMs to check voltage in the circuits·prior to using the new board.· I think I mentioned that·some of the old boards·still had·VIN, VSS and Vdd.

I have tested the old boards on a bread board with a "Hello World"·program and they·give me the same·hardware communication failure error.

I tested a second spectre power supply (no resistors) and it too had 14.7 volts.·

THe LED display has no backlight.

I deeply appreciate your help.· I have been building boards and losing OEMs at a·high rate.· Attached are a·couple photos.

·

kb2hap

It would be best if you could post a schematic.· Its hard to see exactly how you have everything hooked up on the board.

If it seems that everything is good until you hook up the PV, you may want to concentrate there.· How are you hooking it up? What is this relay connected too?

Maybe if your using the same supply for the stamp as well as whatever is hooked to the PV that could be a problem.

There could be a short or some sort of overcurrent or backlash on a stamp pin?

but if you have time draw up a quick and dirty schematic so we can see where your coming from.

it is a 20v 2.5-4.5 amp relay.

FYI here is the data sheet for the PV

http://www.irf.com/product-info/datasheets/data/pvn012.pdf

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
DTQ

Mrdavidjenkins

I put together a drawing for this.· Will you be wanting Resistor values?· I can provide if you want.· I have checked and will confirm continuity for you if requested.· Thanks!

Mrdavidjenkins

I may not have gotten the drawing to you.· Here it is in BMP format.

Chris Savage

David,

·· Your photo relays require a resistor on the input (LED), and you do not have one in your schematic.· This could well be your problem.· You could be drawing too much current from the Stamp I/O pins.




▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Chris Savage
Parallax Tech Support
csavage@parallax.com