I recently built 2 each of the Clock VCO (RevD) and D Flip-Flops (RevC) that I bought PCBs for direct from LZX. I ran into a couple of behaviors that I thought I should document/ask about.
Building the clock VCOs I got a little confused with the switches. The BOM says that S1 is the on-on switch, which makes sense for the range switch and I can confirm with a multimeter that pin 6 of the 74HC4046 is connected to the center pin for S1. However the panel says that S1, the switch further from the knobs, is the sync switch. So the switch placement is correct on the PCB and both modules work fine, but I was thrown for a loop for a bit.
For the D Flip-Flops, I was getting no output from either module but eventually figured out with my oscilloscope that the chip’s active-low reset pin was being held low with nothing patched to RST. The jack seems to be normalling to 5V as indicated in the schematic, and when I patch 0V to the RST input the modules operate but with 5V patched (or I reckon just > 0.5V, haven’t measured precisely where it switches) there’s no output. Not sure if maybe the easiest thing to do is disconnect pin 2A on J5 to let R9 pull the comparator input to ground with nothing patched?
My other Castle/Cadet builds have gone pretty much without a hitch. I’m having a little trouble getting the Cadet I to genlock to my Vidiot, but if I use a passive splitter so the Cadet II output can go to the display and the Vidiot’s camera input, I’ve got everything synced and have been having too much fun patching to investigate that yet.
The Castle VCO and Flip Flops had some errors that have since been fixed, but might you may have gotten an older PCB. First, the Clock VCO. The original at launch BOM called for 74HC4046 but there were some troubles syncing with that IC so it was switched to CD4046. The Sync and Range switches were swapped on the PCB, I suggested using short wires to connect the switches so the places could be swapped back to the original location. Use the short wires to take the switch closest to the jacks to the hole closest to the pots, and vice versa.
The Flip flops, I suggested people cut the PCB trace going from the +5V line to the switching pin of the jack and connect the switching pin to GND. But cutting the switching pin is probably easier
hope this helps.
Ha! That sounds like it solves my issues with my Clock VCO! Without knowing the switches were reversed, it was quite baffling (though still pretty useful).
Regarding the switching around of the switches im a bit confused. Is this the case on all revisions of the VCO?
As it is now i get nothing from the on-off-on switch placed closest to the pots.
The panel have this switch labeled as Range: HI / LO.
But the on-on sync switch closest to the Jacks seems to actually switch between horisontal and vertical sync, and when inputting the right sync signal from my Cadet1 i get synched vertical or horisontal bars. The panel have this labeled as sync, and a H and V accordingly.
If i change the switches around and wire them to the other pads as described, my panel will not match the functions, or am i completely off here?
Referring to the saved BOM i ordered my components from, it says revD, and have the CD4046 in it, which is also the chip on my board.
As stated Im having a hard time getting the range switch to change anything notable on my VCO, regardless of what i do.
Will this be corrected if i make the change around of the switches?
With this said Im still not sure what to expect from this switch, so it might be me just not understanding what to expect and input into the VC input. Could you elaborate on what the use for this switch is?
It has been nagging me for some time that i didnt get anything from this, and planned to aske here…
Could you walk me through what i need to do with those switches in the most basic way?
And the intended function of the range switch positions?
And on a related note:
Could you perhaps make a collected list of revisions and corrections for all the Castle modules? I have noted four corrections from the Facebook group and here:
The Sync Switch (On-Off-On) won’t do anything if you’re not on the same power buss as your sync source since the Clock VCO gets it’s sync signal over the power header. If you plug a sync signal in to the Sync jack it will override the power buss sync.
The currently uploaded BOM and schematic have the correct listings for S1 and S2 that match these refdeses on the (RevD, currently shipping as of 2019/01/22) PCB, but have opposite positions relative to the labels on the panel. This means that if you place sync = S2 (on-off-on) on the spot for S2 so it comes out the hole in the panel labeled “range”, and place S1 in the spot for range = S1 (on-on) so it comes out the hole in the panel labeled “sync”, the circuit is functionally correct, but the labels are backwards.
If the on-on switch is placed in in the hole on the panel labeled “sync” and the on-off-on switch in the hole on the panel labeled “range”, I would expect the sync switch (labeled RANGE in this case) to let you switch between vertical and horizontal sync, but the intended behavior is that the oscillator doesn’t sync to either the horizontal or the vertical pulse when the on-off-on switch is in the center position and that the range switch just has two settings - HI or LO for horizontal-ish and vertical-ish frequency ranges. In this case I guess placing the on-off-on switch (labeled RANGE) winds up floating pin 6 so the PLL/VCO has no external capacitor, I don’t recall what behavior this caused on the display other than that it seemed weird. If the on-off-on switch has no effect in this case then yeah I would expect that the module doesn’t have sync over the CV/gate lines from the power bus (my understanding is that Cadet I and Visual Cortex both generate this, I’m using Cadet I) or something else is wrong.
The fix I think Phil is talking about would be to put the switches in the positions that match the panel labels and wire all 3 pins on each switch to the 3 PCB pads in the opposite position, so that it remains electrically correct while swapping the positions of the switches, like so:
Personally I think I’m just gonna remember that I have em swapped for now Definitely will try the suggested resistor change on the ADC (R10 to 47k) to get more range out of the knobs though.
If you are using the Visual Cortex, the Sync to Power Bus selection switches on the back must be set to “on” otherwise it doesn’t send sync information over the power bus.
I burned a couple of days trying to figure out why this hadn’t been working for me.
Hey everyone, first post here, totally new to video synthesis (though @reverselandfill has been helping me with some questions… thanks!). Starting my first builds (in this case Cadet video input) and have a question on power - it looks like the +5V from the schematics is pulled off the L7805 and not a +5V power connection - can I connect 10 pin headers and cables to the cadet modules or do I need the full 16 pins jobbies? I have spare 10 pin headers and 10-16 power cables handy but not much by way of 16-16. If the 10 pin header is sufficient are there any cadet or castle modules that required the 16 pin connection? Thanks!!
The Castle VCOs will require 16 pin headers since they receive sync via the CV/gate pins on the power bus. As such they may also need to be on the same busboard as your sync generator. I believe these are the only ones since the other modules requiring sync receive it from the dedicated 14 pin sync header, but I could be wrong. Good to double check the schematics to make sure the unconnected pins aren’t used for anything.
Thanks for the quick reply. For castle modules I think I will be using the sync busboard (again, form @reverselandfill). Schematic for Cadet Video Input indicates +5V coming from… well, not sure (and admittedly, it’s these portions of schematics I struggle with). I traced it out and it appears the ICs get their 5V from the regulator and not the Power In but at any rate, I’ll hold off for now!
the 5v comes from the regulator. so you can use a 10pin connector.
in the schematics you can see this on page 2
The Casle Clock VCO is the only DIY module (that I know of) that uses the 16pin power header for sync signals.
But you can also wire it separately to the sync busboard.
Something along the same track - if I have multiple castle vco’s on the same 16 pin connector cable at the back and if I sync the first one via the front input, will the other ones sync as well or do I need to sync each one using a mult for example?
Inserting a plug into the sync input jack on the Castle VCO disables the connection to the sync from the 16-pin header (I checked the schematic). So if your VCOs aren’t getting sync from the 16-pin header, you’ll need to sync each one via the front panel.
That said, it’d be fairly easy to make a little 2hp module that adds some jacks that connect to the pins on your power bus that the VCOs use for V and H sync, and then you’d just need to hook up front panel sync to that 2hp module and you can then sync each VCO to either H or V sync.
But would all 3 VCOs be the same & set to either H or V? Or could they be set independently of each other? Have you tried this @Yatsek?
It’s been a while since anyone posted but now I at least know why my Castle VCO was a bit wonky
The Castle VCO uses two pins (“gate” and “cv”) on the euro power connector, one for HSYNC and the other for VSYNC, and there’s a switch on the VCO to select which one it uses (unless you override it via the panel socket). I made a custom power connector for my VCOs which splits the two sync wires off to an IDC connector which then plugs into the Cadet Sync Gen 14-pin bus in order to pick up those signals.
That sounds like a nice custom work around
If you have previously taken photos of it, please post one up here @VisibleSignals, I’m curious & I’ve gotten the sync bus recently from @reverselandfill so this might be an upgrade I’d attempt.
Definitely will try the suggested resistor change on the ADC (R10 to 47k) to get more range out of the knobs though.
