Did I just kill my SX-Key & can I fix it
Turbo3
Posts: 4
I have been happily programming with the SX-Key and the SX-Tech board for several weeks now.
I am at the stage where I wanted to build my own prototype board. So I wired up my prototype board per the wiring diagram on page 147 of the SX-Key/Blitz Development System Manual 2.0 with one exception. I noticed the SX-Key USB 2.1 board label only listed that three of the four pins were wired or that is how I read the label. The label lists from top to bottom Vss, ---, OSC2, OSC1. So I did not wire 5 volts to the four pin connector.
When I tried loading my test progam into my new prototype board I got the error message "Vpp generation failed". At that point I started to think that 5 volts was really needed and that the SX-Key did not run off only the 5 volts from the USB cable.
I then connected the SX-Key back to the SX-Tech board and got the same error message "Vpp geneartion failed". Which kind of ruined my day. So it seems I have killed my SX-Key.
I am hoping Gunther is still around and might be able to suggest what I kill by not supplying the Vdd connection. I can do surface mount repairs and even have some surface mount SX28s if that is what I killed. Perhaps it is just a transistor in the 12 volt generation that is bad.
Any suggestions on how I can get my SX-Key USB 2.1 working again would be greatly appreciated. I see that Parallax is now out of stock on the SX-Key USB 2.1 so I can not even get another. I was so close to testing my CAN bus monitor when this happened. A very sad day for me. I will keep my fingers crossed that there is a fix to undo what I have done.
Regards, Jim
I am at the stage where I wanted to build my own prototype board. So I wired up my prototype board per the wiring diagram on page 147 of the SX-Key/Blitz Development System Manual 2.0 with one exception. I noticed the SX-Key USB 2.1 board label only listed that three of the four pins were wired or that is how I read the label. The label lists from top to bottom Vss, ---, OSC2, OSC1. So I did not wire 5 volts to the four pin connector.
When I tried loading my test progam into my new prototype board I got the error message "Vpp generation failed". At that point I started to think that 5 volts was really needed and that the SX-Key did not run off only the 5 volts from the USB cable.
I then connected the SX-Key back to the SX-Tech board and got the same error message "Vpp geneartion failed". Which kind of ruined my day. So it seems I have killed my SX-Key.
I am hoping Gunther is still around and might be able to suggest what I kill by not supplying the Vdd connection. I can do surface mount repairs and even have some surface mount SX28s if that is what I killed. Perhaps it is just a transistor in the 12 volt generation that is bad.
Any suggestions on how I can get my SX-Key USB 2.1 working again would be greatly appreciated. I see that Parallax is now out of stock on the SX-Key USB 2.1 so I can not even get another. I was so close to testing my CAN bus monitor when this happened. A very sad day for me. I will keep my fingers crossed that there is a fix to undo what I have done.
Regards, Jim
Comments
On your second test, what were you using to power the SX-Tech board?
I just switched to an even larger 2.5 Amp 8.4 volt supply and get the same error message.
I just noticed that the second large IC from the USB connector does seem a little warm but I don't know if that is a change from before or it has always been warm.
Anyone have an extra or unneeded SX-Key USB for sale?
Did Gunther ever release the schematics for the SX-Key board?
I was however able to use the Debug Again option even after the Debug failed. I monitored the Osc1 & 2 lines on the SX-Tech board with a Scope and do see the 50 MHz clock running.
So it seems the USB interface is working (I hear a ding when I pull/unplug the USB cable which meant the computer could see the SX-Key). Since I can see the clock running on the SX-Tech board I know the SX-Key USB logic can still control frequency generation logic and OSC1/2 pins. So it seems likely I only damaged the program voltage generation circuit.
Limited progress but at least the SX-Key is not totally dead. A schematic would be very helpful otherwise I will need to reverse enginner the board to see if I can pinpoint the damaged component(s).
I decided to slide the label out of the SX-Key board so I could see where the 5 volt signal from the four pin connector was connected. What I found was that as far as I could tell the label was correct and 5 volts go nowhere on the SX-Key. The SX-Key seems to generate the 12 volt program voltage from the USB 5 volt line. The old serial SX-Key would have used the 5 volts from the target board but the newer USB SX-Key has no need for external power since the USB cable supplies it. (Ok now I see another possiblity, I am using an unpowered USB hub so current is limited that is probably where I am having the problem. Will verify if it comes back.)
That of course left me with an interesting problem. Why was I getting that "Vpp generation failed" message.
I decided to try some different SX28 chips to see if somehow both chips I was using went bad at the same time in two different boards and caused a false Vpp error message. Much to my surprise the next chip programmed OK. No error message. I tried another chip and it worked too. Ok, what is going on, so I tried the original chips that had failed and now they work too. I even tried programming in my custom prototype board where I first had the problem and now it works too. (I did wire up the +5 volts just to be safe but I am not willing to tempt fate and try it without the +5 wired).
So I can not trace the problem to any hardware which leaves me with only the notebook computer I am using to run the SX-Key software. It is heavily loaded so I am guessing that the error message is some how software timing related and the problem is falsely detecting that Vpp is bad when it is not. That would explain why after getting it on my board the problem showed up on my previously working SX-Tech setup.
Anyway, I am happy now. I have programmed five working chips some several times without a problem. I did have one of the batch of SX28 chips give a fail to program message but it did do the erase step.
As you can see while re-reading this post I realized that the unpowered USB hub is the most likely problem.
And yes, the USB versions of both the Key and Blitz use the hub/usb port's 5v for power. The thing to remember is that the "Vpp" or high-voltage programming pulses use a "step up" regulator to generate the > 12v needed from a 5v supply, so that 5v supply needs to be pretty good. USB unable to supply > 100 ma may or may not work successfully.
Note also that when running or programming, the Key will *definitely* run quite warm. This is normal.