What's new
Hi @Frank_fjs, the through-holes are much too small for the axial components. If you can please increase the through-hole size for the resistors by 2x, then make the through-holes for all the other axial and radial components the same diameter, that should solve the problem. You should also know the through-holes for the caps that are in my parts list are very precise. But if someone uses different caps that have larger diameter leads, they are going to run into issues.

Since I have the PCB's in hand already, I'm going to either surface mount the axial resistors or bore a larger hole. Does anyone know if either method presents a problem?
 
Sorry about that. The holes fit the resistors I use, 0.5 watt metal film.

You will be fine to drill the holes out larger. Within reason of course :)
 
Looking at the datasheet for the Vishay resistors, the leads specify a 0.6mm diameter which is what I designed the PCB for.

In any event, I'll revise the PCB and make the holes larger. Let me know if anything else needs to be revised.
 
Dry fit a resistor into the diode footprint. Tell me if that's too large a hole or just right.
 
Mounting holes added, board size increased to 50mm x 30mm.

SYNC_RESTORE_03.png
Why is your silk screen showing R1 as 680K ?

On Martin's page he says it should be: 6.8K ?

https://martin-jones.com/2014/09/16/that-syncing-feeling-classic-arcade-games-that-wont-stay-still/
 
Last edited:
Typo in the silkscreen.

Schematic incorrectly shows 680R, I must have then read the correction to 6.8K but mistyped it. Thanks for pointing out, will correct it now.
 
Increased through hole diameter for resistors and caps. Changed caps to suit poly caps. Fixed R1 silkscreen value. Beefed up copper pads.

Sync Pulse Generator.png
 

Attachments

  • SyncPulseGen.zip
    15.6 KB · Views: 206
Hand drilled the through-holes with my precision drill bit but just ended up lifting traces on a few (I went slow and easy). I am just going to order a new set from pcbway.com.
 
Try JLC. 5 x PCBs for $2 with free DHL express shipping on your first order. You'll spend $2 and have them in a week.
 
New boards from JLC come in on Tuesday. One week turn around time. Not bad. I'll snap some comparison pics with PCB's I got from Osh Park when they get here. Won't be able to work on this till next weekend though.
 
Wokee, here are some snaps and an update:

OSHvsJLC.jpg
Comparison between OSH Park and JLC's.

FullyAssembled.jpg
Here is a fully assembled PCB. The fitment of the holes with the components was perfect. The hole sizes are very precise. Snug but didn't have to jam anything in at all.

SyncPulseGenReference.jpg
Snapped next to a dollar bill for size reference.

SyncPulseGenProduction.jpg
And I started putting them all together in earnest. Then realized that I should probably test one before I use up all the parts. Will get to that today and report back.

If ths works for the Sanwa PFX monitor, makes the monitord on par in terms of image quality and sync compatibility with the Toshiba Pure Flat monitor. Exciting stuff!
 
Welp, the preliminary report is they don’t work. Assembled 5 of them. Made a harness for them so I can disconnect them if ever needed. Double checked my wiring, even reterminating. Tested voltages. Tested all 5 of them. I get a ‘No Signal’ message on my Sanwa PFX. Without the generator, I get garbled noise which is known and expected. Damn.
 
Damn. Maybe try adjusting some of the values to change the timing somewhat?!
 
Damn. Maybe try adjusting some of the values to change the timing somewhat?!
I have a version of the circuit that Hatsune Mike designed. 300wins loaned to me. It works to solve my issues with the F3 and Sanwa PFX monitor. So we know the stated parts work.

The other thing is if the resistors I put in are mixed up. That is a possibility. That would mean that Digikey sorted them incorrectly. Because as I removed them from the bags to install them, I put them right back in. Nothing was left out the bags so they could get mixed up. What I will do will head over to Fry's Electronics and grab some resistors off the shelf. Replace them all. I could look at the band colors too. But a little too frustrated right now to think that hard on a weekend. I trust that Frank designed PCB correctly. But it is possible this may be at fault too. There is a chance that my soldering is messed up, but I assembled 5 of them and they all test the same.

I'll check the resistors I put in. Can someone please check the PCB design?
 
Damn, PCB design is fine. Same problem I have, I don’t know why.

The only thing I can think of is the schematic is incorrect. Kinda like how 680R was supposed to be 6.8k in his notes.

I built it on a breadboard before the PCB was designed, parts/values/routing is all correct.
 
In that case i’d drop chris (at) martin-jones an email (from his About and Contact page) and ask...
I’ve emailed him before...
 
Back
Top