Re: [Usbpicprog-technical] Troubleshooting a low Vpp

[David <dev...@fo...>]

I think there's a design issue in how Vpp is being switched.

I figured out my board is closer to rev 644 (minus R13).  To that end, I
removed Q5 altogether as a test, and I still get the same 10.78V for Vpp.  I
would have suspected body-diode conduction from Vpp to the 5V supply as the
culprit (as this appears to be the design oops Frans refers to), but alas no
change.

On the other hand, I think the real issue is Q4: It's an N-MOSFET being used
to switch the 12V supply and the only thing to generate Vgs is the drop
across the MOSFET itself.  The BS170 datasheet calls for a Vth anywhere from
0.8V minimum to 3V max, with 2.1V typical.  As such, depending on the device
tolerances, the ~1.25V drop between the pump and the Zener shunt isn't
enough to drive the MOSFET into good conduction.  Moreover, the problem is
exacerbated with R1 being ahead of the pull-up for Q4, so as the Vpp current
increases (via the shunt or the load), Vgs drops even more.  The result
significantly limits the current on Vpp.

I may have *completely* misread the circuit and I'm a bit rusty on my analog
electronics (had to pull out the old text books), but this appears to be
what's happening.


The fix would be to use a P-MOSFET in place of Q4 (source towards the pump,
drain towards the Zener), with the rest of the circuit remaining unchanged
(though D4 could also be removed).  VPP_CTL would change from being active
low to active high.  The P-MOSFET won't conduct so long as the gate is held
towards the source, but will conduct quite well when Q1 pulls the gate down
to ground (Vgs=-13-14V, which will be too much for the TM2301.  Fix with a
different MOSFET or an extra resistor to make a voltage divider on the
gate.)

Thoughts?

Of course, another, less elegant fix is to bypass Q4 altogether and use Q1
to short out the 12V when not in use.  This will cost about 6mA in draw, but
won't require any other circuit changes.

If I have time later, I might try it.  Nothing like doing some hotwired SMT
soldering!

David

On Sun, Jan 09, 2011 at 12:03:45PM -0500, David wrote:
> 
> Hello everyone!
> 
> I'm trying to troubleshoot a low Vpp on my usbpicprog v0.3 hardware.  I'm
> only getting about ~10.78V with no load attached.  USB supply voltage is
> 4.98V.
> 
> I'm getting a solid 13.25V on the pump side of R1, with only a 0.17V drop
> across it, confirming little load.  However, by the time I get to the Vpp
> header, I'm showing only 10.79V.  The anode of D4 is also low at 11.27V.
> 
> Unfortunately, the board I have (SMT version, pre-assembled) doesn't exactly
> match any of the schematics.  I have what appears to be a Q5, and R13 is not
> assembled (which is what the v0.3.1 hardware calls for), yet I still have
> the 47k and 100k R16/R15 voltage divider on the Vpp sense (v0.3 design).  I
> also have no D10.  So, thus far, until I trace Q5, I'm a bit of a loss to
> troubleshoot further.
> 
> FWIW, my Vpp may have always been low -- I recall having some issues reading
> a device a while back, but I moved the interface to a powered hub and all
> was well.  I didn't investigate further.  Now I'm on a powered hub and I'm
> unable to get a PIC10F206 to respond, which I'm assuming is because Vpp is
> well below spec for program mode entry.
> 
> Any thoughts?
> 
> Thanks!
> 
> David
> 
> 
> ------------------------------------------------------------------------------
> Gaining the trust of online customers is vital for the success of any company
> that requires sensitive data to be transmitted over the Web.   Learn how to 
> best implement a security strategy that keeps consumers' information secure 
> and instills the confidence they need to proceed with transactions.
> http://p.sf.net/sfu/oracle-sfdevnl 
> _______________________________________________
> Usbpicprog-technical mailing list
> Usb...@li...
> https://lists.sourceforge.net/lists/listinfo/usbpicprog-technical