Vintage Synth Repair

Hi group,

still spending some time on a non-booting DX7, but close to give up. To summarize: 
- the DX7 doesn't boot half of the time, LED displaying a blinking 8, FA or other exotic stuff.
- When it succeed to boot, RAM contains garbage, but other than that the unit works (except main output; I'll check later, but headphone is OK)
- I replaced the battery; RAM show +3V when powered off, so replacement OK.
- I have another working unit, so I swapped the main boards between units: the trouble is really located on the main board as all other boards are OK.
- I nailed down the problem to the initial condition (located on G7 on the schematic); on power up, /RES should stay low for 100ms to allow Sub-CPU (IC13=6805S) to boot. It's the case on my working unit (confirmed by simulating it with Oregano), but on my defective, /RES goes high in 10ms.
- When I manually ground IC13 - PIN 28, the unit resumes Ok.
- I already changed the 10/16 cap, TR4 and TR2, but still no success. I checked all resistors: good values, no short.
- to save some ironing, I tested IC17 by trying another 74LS14 on a testing board: if I put the output of TR2 on PIN 5, PIN 9 shows the same signal as IC17; I conclude IC17 is OK.
- once again, I tried the DC board from my working unit; same behaviour.

I'm surely missing something obvious, but I don't know what.

--- In vintagesynthrepair@yahoogroups.com, "alesisfusion8hd" <babouche369@...> wrote:
>
> I like the smell of a cooking CPU in the morning...
> 
> Just checked the resistor values: OK.
> The Zener : I took the entire power board from my working DX7. Same behaviour.
> I made some comparative measurement on power on between my 2 DX7. My scope is DSO 082, I use external triggering (I trigger on CN6, pin 1=+5V) and I measure IC17 pin 9 (/RES). On the working one, I clearly see a spike and then a /RES frankly going high way later (not sure how long; not able nailed it down precisely, but close to 1 sec).
> On my defective, no spike and /RES slowly going high in 10 ms, with a curious wave in the middle.
> 
> The initial clear is really achieved by a few components; TR2, TR4, 4 resistors, a cap and a Schmitt trigger (IC17). A few bucks. I'll probably change them all. If it doesn't work, I'll give a look to a reset chip as adviced by GB.
> 
> --- In vintagesynthrepair@yahoogroups.com, "narfman96" <narfman96@> wrote:
> >
> > At least it isn't trying to cook your breakfast...
> > 
> > Is ZD1 the 6.2 zener on the power board shorted? Also have you checked the 39k and 82k actual values? Somehow the turn on is way too fast. I worked on a Digitech IVL 7000 Pitchrider for 6 months off and on. (It was mine so no customer blues...) Sometimes it would boot most often not. I ultimately found the factory installed a 2.2k feedback resistor in place of a 22k in the slave CPU clock driver circuit. Red,red,orange not red,red,red. "Missed it by that much!" Agent 86.... Ohhhhhhhh Max!  
> > 
> > --- In vintagesynthrepair@yahoogroups.com, "alesisfusion8hd" <babouche369@> wrote:
> > >
> > > Hi Narfman,
> > > 
> > > no the 6805 doesn't get warn.
> > > I plugged another 10uF in parallel (giving 20uF), but no change.
> > > I tried with another power supply (and another keyboard); still the same.
> > > 
> > > What buzzes me is /RES seems to reach +5V in 10ms; the datasheet says it should stay low for 100ms. That's why my manual reset works.
> > > 
> > > Strange thing: I wanted to sinc my one-trace small scope to the +5V (to precisely measure the time between power on and RES). I made a small BNC cable (coming from an old network cable), but the cable had a short I was not aware of. I took my supply from CN6 (pin 1=+5V, pin 4=GND). When I plugged it to my working DX7, I had the exact same behaviour.
> > > 
> > > --- In vintagesynthrepair@yahoogroups.com, "narfman96" <narfman96@> wrote:
> > > >
> > > > Good troubleshooting! Does the 6805 get warm? Since all the resets come from the same timing cap you could try a 22uf and see if the extra settling time helps. I found the 6805 on Ebay. Not sure if it's the same package...  
> > > > 
> > > > --- In vintagesynthrepair@yahoogroups.com, "alesisfusion8hd" <babouche369@> wrote:
> > > > >
> > > > > Just a follow up. I nailed down my issue to the Sub-CPU 6805 (IC 13) which sometimes does not boot up the right way. When I reset it by grounding pin 28, everything runs smoothly.
> > > > > My guess is the condition on power up does not reset the sub-CPU long enough (the 63B03 boots every time; less sensitive? ). That's the best case. The worst case is a defective 6805.
> > > > > I changed the 10 uF cap of the Initial Clear, but no success. I'll take a chance and change the 4700 uF on the PSU, or simply switch it with my other DX7.
> > > > > 
> > > > > --- In vintagesynthrepair@yahoogroups.com, "alesisfusion8hd" <babouche369@> wrote:
> > > > > >
> > > > > > Thanks man, the tuning procedure is exactly what I was missing. 
> > > > > > 
> > > > > > I've cleaned up all the connectors (on main board and sub boards); it didn't solve my issue, but it's a good start. I've also tipped on all chips & components to track any dry joint (I'd this issue on a Kurzweil 1000).
> > > > > > The issue really resides in the digital part during the init phase. Happily, I've got another DX7, in working condition; I can make some comparisons.
> > > > > > 
> > > > > > 
> > > > > > --- In vintagesynthrepair@yahoogroups.com, "z3sound" <z3sound@> wrote:
> > > > > > >
> > > > > > > 
> > > > > > > hi 
> > > > > > > there was no complete Service manual for DX7 (first edition)
> > > > > > > on the internet.but now it is 
> > > > > > > i just upload the complete  DX-7 first service manual in Files
> > > > > > > hope it will help......
> > > > > > > 
> > > > > > > --- In vintagesynthrepair@yahoogroups.com, Kashayar Dashtdar <vdashtdar@> wrote:
> > > > > > > >
> > > > > > > > Hi my friend.
> > > > > > > > 
> > > > > > > > Unfortunately there is no complete Service manual for DX7 (first edition)
> > > > > > > > on the internet.
> > > > > > > > 
> > > > > > > > Try these :
> > > > > > > > 
> > > > > > > > 1- Replace all the electrolytic capacitors in the power supply board.
> > > > > > > > 2- Send a fresh sound bank from the PC to the keyboard using a midi
> > > > > > > > interface.
> > > > > > > > 
> > > > > > > > I hope it will make your dx7 work properly.
> > > > > > > > 
> > > > > > > > Good luck.
> > > > > > > > Ciro
> > > > > > > > 
> > > > > > > > On Tue, Mar 6, 2012 at 8:49 AM, alesisfusion8hd <babouche369@>wrote:
> > > > > > > > 
> > > > > > > > > **
> > > > > > > > >
> > > > > > > > >
> > > > > > > > > Hi group,
> > > > > > > > >
> > > > > > > > > I recently bought a first generation DX7 for $ 50. Problem on startup:
> > > > > > > > > sometimes starts up OK, sometimes keeps looping, sometimes seems OK but
> > > > > > > > > actually frozen.
> > > > > > > > > When it starts up OK, I found only garbage in the internal memory, but it
> > > > > > > > > works with the cartridge. It does not memorize my settings, and main output
> > > > > > > > > seems dead (headphone OK). I've suspected and replaced the battery cell.
> > > > > > > > > I was able to run the self tests up to test #5 (analog inputs), but I have
> > > > > > > > > no foot controller to complete this stage.
> > > > > > > > > I checked the PSU and the voltage source on the main board (+2.5V, +7.5V
> > > > > > > > > and +5AV) and everything is OK.
> > > > > > > > > I suspect a signal is missing on the sub-CPU (6805), or a defective RAM.
> > > > > > > > > I picked up the schematic and the service manual on Dave Benson's site,
> > > > > > > > > but it seems to be altered after page 20. All other services manual I found
> > > > > > > > > on the web has the same issue. Does anyone have a full copy of it?
> > > > > > > > >
> > > > > > > > > Thanks,
> > > > > > > > > François
> > > > > > > > >
> > > > > > > > >  
> > > > > > > > >
> > > > > > > >
> > > > > > >
> > > > > >
> > > > >
> > > >
> > >
> >
>