[DSP-10] Need some help with initial setup

[Bob Larkin]

Hello John,

It is good to have you back with the project!  I hope all goes well.  You 
may have already gotten your problem solved, but as Jiro pointed out, it is 
likely the timing problem relating to the waveform changes from the drivers 
used with the KK7P DSP. Here is the email from last March. I just realized 
I never posted it to the web site and will try to get it there in the next 
few days!!   73, Bob Larkin  W7PUA

----------------------------Copy of earlier email--------------------
To: <dsp-10 at mailman.qth.net>
From: Bob Larkin <boblark at proaxis.com>
Subject: [DSP-10] DSPx/KDSP10 modification needed
Date: Thu, 18 Mar 2004 13:55:50 -0800

Going back a week or so, Jiro JQ2LMG posted a note about improper data 
output from the U107/U108 shift registers. These registers are programmed 
serially from the DSP.  Since then there has been experimentation and 
discussion with Jiro, Lyle KK7P, and Steve  N7HPR.  This has resulted in a 
recommendation for a modification. This is particularly important for the 
DSPx board, but could possibly be beneficial for the EZ-Kits and in no case 
should be detrimental!

The 74HC595 shift registers are designed to work in cascade, but this makes 
the assumption that the serial clocks for the two will switch at the same 
time, within a few nanoseconds.  If the rise time of the clock pulse is 
slow, this requirement may not be met. This slowness is apparently 
aggravated by the 3.3 to 5 Volt transition required by the 2185 DSP in the 
DSPx.  If by bad fortune, U108 switches before U107, the data can get 
shifted incorrectly. A hardware fix, that seems to cure this, is a 
capacitor on the U108 serial data output to hold the data for an extra 20 
nanoseconds or more.  This turns out to be easily added to the existing 
DSP-10 main board.

Connect a leaded 470 pF to 1000 pF between pins 14 and 8 of U107. There are 
via holes near each of these pins that can be used to connect the cap.  The 
via connection traces are visible under the solder mask. The cap itself can 
go right over the U107 IC. The leads are not critical in length, but short 
is always good.  The value and type of the cap is not critical.  Ceramics 
would be fine. Values might even be as much as 2200 pF (0.0022 uF).  This 
modification can be made without removing the board, but, be sure the leads 
do not poke excessively through the board.  Obviously, turn off the power 
before doing this!

I would recommend this modification for all DSPx/KDSP10 versions.

Thanks to Jiro, Lyle and Steve for the help.

73,
Bob  W7PUA
-----------end copy------------------

At 11:41 PM 10/16/2004, you wrote:
>Ok,
>
>After a long illness this past summer I'm finaly able to get back to work 
>on my DSP-10.  The assembly is complete but I'm still having some 
>problems.  I have gone through the checks as follows:
>
>
>DSP-10 Nominal Voltages (no KK7P DSP)
>
>SOURCE 
>Expected        Measured
>U101 pin 1 10v Regulator input                          11.00   10.99
>U101 pin 3 10v Main regulator output                    10.00   10.00
>U102 pin 3 5v Main regulator output                               5.00  5.04
>U103 pin 3 126 MHz synthesizer 5v regulator               5.00  4.98
>U13 pin 3 - 19 MHz synthesizer 5v regulator               5.00  5.08
>U14 pins 2 & 13 - Audio outputs                           5.50  5.43
>Q106 pin 1 - Receive line voltage                                 9.70  9.79
>U1 pin 3 - First RF amp                                           3.50  3.43
>U2 pin 3 - Second RF amp                                          4.70  4.68
>R110 - Source of Q101 VCO                                         0.23  1.83
>U105 pin 3 - VCO buffer                                           3.50  2.80
>U106 pin 3 - VCO buffer                                           4.60  4.18
>R59 - Source of Q4 
>VCO                                            0.91  1.42  In the range of 
>.91 to 1.8v
>U8 pin 3 - VCO buffer                                             2.80  2.89
>U9 pin 3 VCO 
>buffer                                                       3.60  3.47
>
>
>With KK7P DSP Module
>
>Q106 pin 1 - 10v 
>receive                                                  9.70  9.77
>U10A pin 3 IF ampilifer                                           6.00  4.36
>U109B pin 6 - Transmit IF driver                                  2.40  2.43
>
>
>RF gain to 100
>
>D2-A1 - Antenna TR diode                                          0.73  0.01
>D1-A1 - RF Filter TR diode                                        0.73  0.48
>
>
>Mode to CW - 147.00 MHz
>R104-R105 
>Junction                                                3.5v  4.9 
>Adjusting L102 has no effect on the voltage
>
>The R110 at the source of Q101 is bad as are the TR diode voltages.
>The voltage at the R104-R105 Junction is bad as well and L102 seems to 
>have no effect on this voltage which it should.
>
>Now the strange stuff.  I also ran UDIAG1, UDIAG2 UDIAG3 and everything 
>checks out ok...  Both PLLs lock and seem to sweep the range.  BUT, when I 
>load UHF1.EXE I get "E30 PLL Unlock"!   What the heck.  What am I missing?
>
>Thanks guys
>
>John
>KC0OIE
>
>_______________________________________________
>DSP-10 mailing list
>DSP-10 at mailman.qth.net
>http://mailman.qth.net/mailman/listinfo/dsp-10
>