JuneBug - BS250 orientation

[August Treubig]

Backards/Foards

Actually,

Backwards or Forwards BS250S, the programmer should see the 18F1320. I ran with my homegrown one in wrong for quite a while, before I discovered the ERRROR or my ways .

The only problem is that programmimng software will think that it sees voltage comming from the target board. Even if it is self-inflicted voltage. I finally got to the bottom of it and reversed it and things got better.


But the key is the programmer part will still work with it in Backwards.

 

[LabRat]

August,

I think we're getting two threads crossed over here.
On my board it was the 2n3906 that was incorrect, and thus could not see the 18F1320 (well.. I hope that is why)

There was also "theoretical discussions" about what if the BS250 was reversed.

 

[August Treubig]

That makes things go west

Yea, if the 3906 is backwards, then the Vpp will be messed up seriously and the 1320 probably won't answer

 

[LabRat]

(in my best "Montgomery Burns voice")

"Ehhhxcellent"

It's always nice when the problem you found does seem to be related to the problem you are trying to fix. Can't wait to give this a try when I get back home and/or the kids are asleep.

 

[LabRat]

Ahh nuts... still something amiss.

Error from MPLAB -
Found PICkit 2 - Operating System Version 2.20.0
Target power not detected - Powering from PICkit 2
PKWarn0003: Unexpected device ID: Please verify that a PIC18F1320 is correctly installed in the application. (Expected ID = 0x7C0, ID Read = 0x0)
PICkit 2 Ready


Suggestions on what/where to check next?

 

[blueroomelectronics]

Make sure dip switches 1,2,3 are ON.

Also try the PICkit2 2.40 software as it has trouble shooting tools.

 

[LabRat]

Will try that.

The switches are on, and oddly enough a couple of times it DID actually probe and get back the correct value. (yeah.. go figure).

Took it back to the bench and revisited the majority of the solder joints, but I'm back to no connects. Will search out that software now..

 

[August Treubig]

Go down that Road

As Bill says, use the Pickit2 V 2.4 tool and go to Tools -> Troubleshooting.

From there you can verify what is missing. Voltages / Signals.

 

[LabRat]

SHORT detected - short or heavy load on VPP - VPP shutoff.

Reviewing the schematic shows R10 as being 10K, but the kit has it as 1K.
Now this is only a "control" signal, but as I've replaced the 2N3906 with a
2N4403, would this make a difference? Perhaps I need a larger resistor here?
(or smaller for that matter)

Or .. could the presence of the 2N2222 in this spot have caused further damage?

BTW: I do appreciate all the suggestions and help.

 

[blueroomelectronics]

Wow, I can't belive noone's noticed that including myself. I go by the parts list when putting them together. The 2N2222 should have not caused any problems, but the short is a mystery.

 

[LabRat]

Ahh.. it's R10 vs R1D

My eyes.. plus a poor printout. (Would explain why no-one noticed )

 

[August Treubig]

Ok, what's right

Home brew one here follows schematic. R1D = 10k, R10 = 1k.



I am assuming the schematic is correct, since mine works.

 

[LabRat]

Yes confirming that it was my mistake on reading the schematic (I blame the laser printer with the fading toner). I read R1D as R10.

There is no error there. (Still have something wrong, but alas my time has run out until next week (or perhaps Saturday evening))

 

[petrv]

Have a look at the datasheets of BS250 - there is the version from Vishay (BS250KL) and also from Zetex (BS250P). You can see that the Vishay and Zetex versions have opposite pinout, this is I think the source of the confusion ...

BS250 is not really the greatest P-FET, you can get cheaper PFETs with better parameters (Rds-on etc...) but most of those newer parts are SMD (I was looking for a PFET myself and bought FDC5614P from Fairchild, there is a bigger one FDD5614P as well).

But if you must use through-hole components and you don't need the big power MOSFETs in TO-220 package the choices are limited.

Petr

 

[LabRat]

Q1?

Hey everyone. I downloaded the test app, and it responds with a failure on the VPP line. After poking at the source, and reviewing what it is doing (purely reading back from the VPP_FB line), I have the following question.

For a standard 18F family part, VPP should be up to 12V (according to the troubleshooting).

Anyone have a pointer to a document that details how the charge pump is supposed to work? I can see the pulses on the VPP_PUMP line, but I'm not sure if Q1 is actually doing anything. (Again.. software guy here). Is there any way to test Q1 in place? Or do I need to pull the part from the board?

 

[blueroomelectronics]

Jumper U5V & +5V with a wire (use the CON3 connector on the top)
Do you have a logic probe or scope? There should be a square wave on the base of Q1.
Q1 works like the switch does in the graphic below.
**broken link removed**
**broken link removed**

 

[LabRat]

Logic Probe, Multi-meter, and a GameBoy Oscilloscope (see GBDSO)

"scope", and the diagnostic software I saw the clock pulses going into the base of Q1. The Voltage on the Collector is fixed at about 5v (4.99). Assuming that Q1 was operating, shouldn't this be measuring as higher than the 5V? (Similar measurements at the "other" side of D1.

oh.. and don't you EVER sleep??

 

[blueroomelectronics]

It should be pulling the inductor to gnd and letting go for a nice HV spike, Sounds like Q1 is bad, must have killed itself when it saw the 2N2222. Now that 2N2222 would make an excellent Q1 replacement.
5V solid means Q1 is not doing its job.

Sleep, I sleep in on Sundays

 

[LabRat]

Ok. We're both on the same page, and I have 3 minutes before you go into your weekly hibernation cycle. I hope to try and replace this tomorrow evening. Luckily I had a stock of 2N3904's kicking around.

Will keep you posted. Did you make progress with your PK2 firmware tweak?

 

[blueroomelectronics]

Actually yes. It's seems to be quite easy.
The bootloader has a byte in high memory, you set it to 0x55 and your program will be in control.
Your program should watch the state of the bootloader pushbutton and clear the byte to release control.