You may have tried some/all of this already, but in case not:

I would test it first without connecting it to any other modules, no sync connection and no front connections. Just composite output to the tv, and then in a separate test just YPbPr to the tv. Make sure you have the dip switches set correctly for the output and format you want. In either of these states (only composite output or only component output), you should get a black output image if you have all the knobs and switches (normal switches, not DIP switches) set to their middle positions. If that works, test further by flipping the normal switches one by one up – when you do that, the module should output the respective color. If this works, that means the module is generating sync correctly (internally at least), and at least partially working.

If that doesn’t work, first plug something else that outputs the same format video into the same input on the TV that you were using for the ESG3, to check if there’s an issue with that format going into the specific jack/jacks on the tv. If the tv works fine with something else generating the same format video into the same input jack(s), then there may indeed be a problem with the ESG3.

If that did work when not connected to any other modules, the next step is to try it connected to another module. I’d first try to put something into one of the R, G, or B input jacks on the ESG3, when it’s hooked up in the way that was working before, just to see if that part of it’s working ok. Then I’d move on to trying to sync it to you Cortex, and trying to sync the Cortex with it, which you likely already tried, but you might have better luck now if you successfully got it working on it’s own before (maybe you were missing something before for some reason).

LZX Support will probably be able to step you through some more pointed testing too.

You can tell that it’s not genlocking correctly because the LED below the format selection DIP switches is yellow.

It will immediately turn green when a video signal running at the same timing format as ESG3 is plugged into the rear Sync In.

Building on that, I should add that the light under STANDARD is yellow for me when I’m using my ESG3 as the sync generator without any sync input. So during the tests I described where it’s not hooked up to anything, I think it’s normal to be yellow in that case. (Edit: Just checked the docs and that light is green if sync is detected, and yellow if no sync is detected.)

Yes, I tested without inputs. Sadly, output is similar to with inputs (just without an image).

ESG_06461920×1440 667 KB

ESG_06481920×1440 721 KB

ESG_06491920×1440 1.01 MB

Color controls do seem to work however (this is with RED switch up; other switches and contrast/brightness pots all affect signal as expected as well).

ESG_06511920×1440 568 KB

I assumed that since it was labeled as Sync Input LED in the doc, but wasn’t 100% sure TBH.

Will test HD formats next, to see if I have any more luck with those, but need to rearrange some stuff first since my current rig is set up for an SD workflow.

One last thing to try would be powering the module using the 12V barrel connection, to see if that improves things. If you see no improvement with this or any of the above suggestions, could you send an email to support@lzxindustries.net?

I am also having the same issue. It seems as though my esg can not create a stable output. I also pointed this out in the discord.

Things I have tried in HD (multiple formats on the esg front panel):
esg → bmi mini analog to sd → md-hx → atomos ninja
*When this is tried the md-hx will recognize that there is a hd signal coming into it but nothing will show up in the atomos until you go into the md hx change the “hdmi output” to “scaler” and change the “duc ref” to “free run”. Tried this both from barrel and eurorack power with things plugged into the esg and nothing plugged into the esg. everytime you do get an image in atomos it is accompanied by glitchy horizontal lines.

In SD:
esg → v4 → composite ntsc monitor
big glitchy blocks appear on the screen.

Thank you @joem for your suggestions, sadly none of them fixed the problem.

Thank you to all who are reaching out to support@lzxindustries.net and submitting support tickets. When sharing potential issues here on the forum or via email, could you please include photos and video, both of the patch and its output? It always helps to see what you are seeing.

Figured I’d follow up on this.

Received repaired ESG from LZX a while ago and finally had a chance to test it out again today. And… sync now seems to be working as expected. This includes input into a Blackmagic Analog to SDI box — at both SD and HD resolutions (at least the ones I anticipate using; didn’t test every single sync rate) — so the fix they have cooked up for that issue seems successful as well.

Thanks, LZX!

This is excellent news! Thanks.

Great to hear, I need to get mine in for its RMA soon. I’ll do a similar test to this one with VU007B + BMD Analog to SDI + Video Assist but with ESG as the encoder once it’s back in the studio.

My assumption is that the Syntonie decoder is not at fault, it should just be passing sync through with no timing modification.

Seeing horizontal jitter from other sources as well, such as these cheap Chinese TV tuners I’ve got. But I chalked that up to … cheapness. Figured the Roland switcher would be more robust.

Need some sort of image onscreen to see the jitter, a flat matte from ESG3 wouldn’t be a good test. But I am seeing the jitter when patching DSG3 patterns into ESG3.

I’ll continue to run tests, e.g. bypassing the VU003B. Maybe it’s a power issue. In this rack, the Gen3 modules are powered by a 5A DC Distro, everything else powered by Doepfer PSU3.

Also had the rig running for several hours yesterday for the first time, and saw sync completely drop out for a few seconds, a couple of times.

Just got my capture system working, don’t have any examples to share yet.

@Z0NK0UT @creatorlars

Here is the issue I am seeing with ESG3. When ESG3 is connected to external sync, the picture starts to jitter. Look in the center of the frame and you will see the vertical lines are wiggling left to right.

I’ve bypassed all other modules and am connecting a 1080i59 Y signal directly to the ESG3 sync input. ESG3 DIP switches are configured for 1080i59.

In this clip, DSG3 is generating ramps and going directly into ESG3. I have not configured the DIP switches on the rear of DSG3, I left it in automatic mode. But the jitter also happens with external video from a Syntonie VU0003B decoder, so that would seem to indicate that the issue is not with the DSG3.

I have tried several different sync sources, and so far the problem persists. The jitter is intermittent, coming and going according to no pattern I have determined.

As soon as ESG3 is running on internal sync, the issue immediately goes away and the picture is rock solid.

To make matters even more confusing, if I send genlocked ramps from Syntonie VU009 into the ESG3, then I do not see the jitter issue.

The ordinary sync signal path in this rack is this:

External source → VU003B → ESG3 → DSG3 → FKG3 → VU009 → VU009 → VU009

The sync signal is unterminated at the end of the chain. But I tried it with a 75 ohm terminator plug and it didn’t make any difference.

In the clip posted above, the sync path is this:

External source → ESG3 → DSG3

I am also seeing the whole system lose sync randomly for a few seconds every now and then, with no idea why. The picture just goes away and then comes back. ESG3 is going through an AJA HD10AVA to convert it to SDI before hitting the monitor. I can try swapping that out, but honestly the more I fiddle with this the more I think the issue must lie with ESG3. The cornerstone of the whole studio, and apparently not reliable.

In both of these cases, your external source is generating the reference timing for the entire system. Here’s what seems to be be going wrong:

ESG3 is having difficulty attaining a perfect lock to your external source. But that glitch may potentially be coming from the external source rather than ESG3 natively (since you are getting stable timing when ESG3 is in control of the output timing.) Or it may be related to the transmission line (cables, connectors, adapters) between the two devices. Testing with a variety of external sources and in different video formats may illuminate the issue. To my eyes, it looks like noise in the transmission line somewhere. If you have shorter cables, swap some out – make sure there aren’t any passive adapters/ interconnects or long cables ( > 3-5 ft) in the sync path.

One fix is of course to use a TBC to acquire your input video source, and that way all timing will be generated from inside the modular system. With HD timings, this may be even more critical (especially if the source itself might be noisy.)

So the more data we can get on the external sources tested, the better. If we can recreate the issue here, then we are able to make measurements and arrive at theories. So far, this isn’t an issue we’ve recorded: ESG3 locks without glitch to what we have available here (various Sync Gens, SDI to Analog converters and cameras.)

Thanks, I’ll keep testing with other devices. Once TBC2 arrives, this problem will probably go away.

The interference theory is solid. Keeping cable lengths short is not easily done, I’m forced to use 6’ and 12’ component cables. But I am using good shielded ones.

Still mystified by the fact that DSG3 would jitter and the Syntonie oscillators don’t. Cracked open the case to double check the sync connections and it looks as expected, not terminated after the VU009s. But Bastien assured me that would not interfere with their function. And of course we know that the Gen3 modules have buffered sync.

The other thing I want to try is making ESG3 the master clock. But I am confused about the Roland V1600-HD switcher Reference input. The docs don’t say whether that’s an analog or SDI input. So far, I haven’t been able to get it to lock to an external Y 1080i signal, but I will keep trying.

What are the external sources you are using to test ESG3? Have you tried a signal that uses a format other than 1080i5994? Do you have a way to monitor without the AJA HD10AVA? The more you can pare down the setup, the closer you can get to the issue.

Thanks Chad. I’m utterly married to 1080i59. Of course I can test other formats, but I really need this one to work.

I can monitor the ESG3 YUV output directly, that’s next on the list. After the firmware upgrade I haven’t seen the problem with the Lilliput monitors not syncing to ESG3. But they are only small field monitors, and I can’t see the jitter issue very well. But I do have other component monitors I can test with.

So far I’ve seen the jitter issue when syncing ESG3 to the Roland V-1600HD and the cheap Chinese TV tuners. I have an analog HD camera I can try next.

@Z0NK0UT @creatorlars

After another afternoon of testing, the problem with ESG3 syncing to external sources has only worsened. I’m afraid I’m going to have to send it back in to you because I have not been able to get it to work as expected.

• Bypassed the AJA converter, sent ESG3 directly to a component monitor - problem persists
• Shortened cable runs, eliminated all adapters - problem persists
• Sent Y 1080i59 sync from a like-new video camera - problem persists
• Changed camera and ESG3 format to NTSC - problem worsened
• Changed ESG3 power source from 5V DC Distro to Doepfer PSU3 - problem still worse than before

In the most stripped down testing scenario, I had the Y output of the camera DIRECTLY connected to the ESG3. Literally one shielded cable, no adapters. And the output of the ESG3 DIRECTLY connected to an HD monitor.

So I have eliminated all possible variables I can think of. At this point the problem is actually much worse than it was yesterday. The jitter is not subtle, it’s really really really bad. In the realm of completely unacceptable. Unless you can think of something else I can do, I am going to have to request another RMA.

Aaaand now the system has been running for a few hours, ESG3 is locking to external sync perfectly. But I didn’t do anything.

Keep it under observation – it’s definitely possible that there could be some glitches while everything is still warming up (a couple minutes after startup). If you feel like there’s an issue with your unit, contact support and we can test it here.

Well, VU009 is fairly simple in the design, sync is extracted using a sync extractor IC, and the resulting pulses are resetting the oscillator, as most analog oscillators designs, whereas DSG3 is a digital core, so it needs to genlock to the external sync signal, in a similar fashion an encoder does to an external video source, which means it may be more picky depending on what’s in the sync chain.

What I meant to say was that VU009 should operate properly even if the sync isn’t terminated (as the sync extractor IC, LMH1980, can handle 0.5Vpp (sync terminated 2 times) to 2Vpp (sync unterminated)), however it may cause issues with the modules upstream in the sync chain, so it is good practice to terminate the last VU009 in your chain. As even if the sync is buffered on DSG3/FKG3, having the output unterminated may yield unexpected behavior. I’ll try to reproduce the sync chain your experiencing issue with and report, in the meantime, make sure the last VU009 is terminated.

Planning on buffered sync for the next modules I’ll make that requires sync, as it makes a lot more sense than having to manually set the termination.

@creatorlars Thanks Lars, I am also going to try changing up the power. Maybe the StarTech power distribution unit is at fault? I didn’t have the budget for the Lite-On version, had to opt for a less expensive Chinese PDU.

@syntonie Thanks Bastien, I appreciate you weighing in with an explanation as to why Gen3 may be more picky with sync. I will put that terminator plug back on, even though it didn’t seem to make any difference in my tests.

I’m running into a similar situation. Was the cause and/or fix for the OP’s issue every identified?

I’m running 1080p29.97, the patch takes output from DSG3 and DWO3 into the ESG3.

Here is what I’m seeing.

1. Slight wiggles in the display.
2. When using Sync In on the back of the ESG, the display will occasionally go blank for a second, then re-sync. I’ve tested this with Sync Input from a VU003B and a BMD Sync Generator. The problem only occurs when using the ESG3 Sync In.

Edit: Here is a link to a video that shows what is happening. https://youtu.be/lDYw5TuefQs
FYI this ESG3 has the firmware update applied for the BMD Analog to SDI issue.

Thanks!

Pete

FYI re my previous post, the ESG3 was returned and repaired. All is good now!

I’m trying to use the ESG3 in 1080p 25fps and on my screen the picture is divided in two and it’s wiggling. The problem goes away when I use the 1080p 60fps which would be fine otherwise but the file sizes are getting out of hand.

Any tips for fixing this issue? I’m using the BMD Ultrastudio HD Mini with a M1 MacMini.
On the Desktop video software the settings change accordingly when I move the dip switches around. Only the one mode (1080p 25) I’d use out of all of them is not working.

Could you please make a new post specific to your issue and share relevant images and video? I’m sure we can help you determine what is going wrong.

Hey Lau!! Vidiot does not output.

To answer your technical question, black burst basically means “black video signal with color burst and sync”. It’s a standard normal video signal that’s just black. The sync output from Vidiot is more like “black + sync”. It is black video and sync but without any colorburst. You can use any video source that is synchronized to your main system as the video input to Vidiot, and it will sync. Vidiot already has the same genlockable sync converter circuit as found in Cadet I.

Does this answer your question in full? Love having technical discussions, so make sure all your questions are answered.

Thanks for the explanation @creatorlars, but its not exactly what im after. I will try to rephrase my question and give a little background.

I know that the sync generators, both the cadet I and the vidiot outputs a black sync signal from the out RCAs when no signal is plugget to the input, but what im after is some output (in the 14 pin sync bus) of a proper sync signal which is black, nomatter if a video signal (with an image) is plugget into the input of the sync generator.

Usually i sync my cadet I sync generator to a signal coming from either a camera, that does not take genlock, or from one of my blackmagic capture cards (that have no genlock,) and then genlock all the other signals i have before inputting them to the cadet video inputs and the vidiot. (i have 3 Cadet Vidio inputs and the vidiot so that means that i often use 4 different sources into the system at once) I then route all the synched signals around in my analog composite system (containing the synth, V8, mx50s etc…)using a matrix router. This is all pretty simple. (for me at least )

But what im after is synching my vidiot to the cadet system, without getting a visible video input into the vidiot. (because the video signal is then always present in the luma section of the vidiot, whichI sometimes want, sometimes dont want).

For now i generate such a black signal outside the lzx system and use this to sync the vidiot to the rest of the system, without inputting an actual visible video stream into the vidiot. I would like to ditch the rackmount sync generator, and use a signal already present in the sync bus to sync the vidiot, if possible.

Is there a black sync signal generated inside the cadet I sync generator, which is sent through the 14 pin sync bus outputs that could be used to sync the vidiot to the cadet I:

I would of course have to DIY some kind of adaptor from the sync header to RCA, that could be input into the RCA input on the vidiot, but this seems pretty doable

In other words will my vidiot slave itself to the sync generator, if i feed the vidiot with Burst Gate, C sync, or any of the other signals present here?

This all relies on the assumption that what the sync generator sends out through this header is only sync signals without any image information. This might be a wrong understanding?

Hope this makes sense? And thanks for the explanations!

you might be able to use c sync out to the vidiot. not sure that it would need color burst or not, since it only gets luma on the input, i wonder if the sync generation is similar to the cadet in that it will generate a colorburst (edit: nevermind that, sync gen doesnt generate colorburst, the output encoder generates the colorburst. everywhere in between is just rgb signals). give it a shot anyway.

For what you want, just build another C2 RGB Encoder for your Cadet system and don’t patch anything to the inputs. This will give you black video output and serve as a black burst generator.

The other option would be to make a DIY circuit that attenuates and buffers the CSYNC on your sync bus and drives it thru a video op amp buffer. @reverselandfill has incorporated this approach into his sync busboard design, with the full thread located here:

Alternately, you can drive the Vidiot from Visual Cortex’s sync out chain on the rear of the module.

The CSYNC on the 14-pin sync bus is TTL level (5V logic). When you attenuate it to 0.6V, buffer with video opamp, and drive thru 75R resistor, it is now “black video.” (Not quite black burst, which also includes subcarrier embedded along with CSYNC, but that’s a different topic.)

Thanks Lars. Seems a little more complex than i had hoped for. Had not grasped it in my first read through of the sync busboard thread.

Making a full C2 module just for a black video signal seems a little over the top for me, and I have no Visual Cortex (module that is! ) so i think i will go for some kind of diy solution as described.

You could also go with something like this:
https://www.markertek.com/product/bmd-convmsync/blackmagic-design-convmsync-sync-generator-mini-converter?cvsfa=3786&cvsfe=2&cvsfhu=424d442d434f4e564d53594e43&ne_ppc_id=1629435546&gclid=CjwKCAiA7vTiBRAqEiwA4NTO65g3w55KZjeLZovH8muY_071orkAzsIRVFx3oqEqenxbXKoDK_-S-BoCPxkQAvD_BwE

Oh, yeah, thanks. That is indeed smaller for the rig, not bad! (but still a bit pricy)

Think im going for Martijns Sync bus board that have an rca sync output though.

The RCA sync busboard is the obvious solution!

take a look here specifically at the sync section
there is a vidiot picture that I’ve put up that describes my setup
most of the time the issue with the vidiot has to do with a wrong input / output setup

so the most helpful thing you can do if you aren’t sure would be take a picture of the back of your vidiot with everything plugged in also take a picture of the back of the visual cortex

Thank you for the link! I will try that out. I was running the video out of the Vidiot into the video in of the VC. I thought that was how the LZX YouTube video directed set up.

You will also want to connect the sync output from the Cortex to the sync input jack on the back of the Vidiot.

Thanks for the tip! I should be getting my sync cables in soon!

You will only need a 2.5mm (1/8") cable for the connection.

I found some extra cables and got the Vidiot synced.
Now I noticed a visual issue. Its not the Vidiot, because I see it with it connected and not, it does it with the Visual Cortex. It is a line on the left side of the screen and the bottom of the screen that is alway there. Does anyone know what could cause it? I wanted to try to have a consistent symmetry with the imagery and this trows it off for me. It smudges the pattern a bit. You can see it in this screen shot.

That is odd. Is it there no matter if you are just using the Cortex on its own, patched internally?

Do you see the same issue on multiple displays/recording devices?

I broke it all down and patched it back up with just the VC and don’t see the lines. I will run it back again through the Vidiot and see. Might have bee my Blackmagic too. That thing is a bit quirky.

Just received my Memory Palace, very excited, but one quick syncing question. Do I run sync out from the Visual Cortex to the in/loop thru of the MP then sync out from MP to sync A or B of Vessel? Also do I set it to external on the MP? What do I set loop thru/ext sync/terminate to? I want to be able to also sync my Vidiot to the Vessel A or B sync out for Vidiot use.

Yes, these are the correct connections to use Visual Cortex as your sync source. You want the MP sync source set to EXTERNAL since it’s coming from VC, and you want the loop switch set to THRU since MP is not terminating the sync signal, you are passing that through to the Vessel bulkhead connector. Note that you want to use the two Loop Sync RCA connectors on the Memory Palace, not the Sync Out, since MP is not generating sync in your configuration. This is the same as my setup, except I’m also going THRU a Prismatic Ray after MP before leaving the Vessel. I also just received my MP today and set it up like this (VC → MP → Prismatic Ray → Vessel sync connector → Vidiot). Also make sure you center the MP sliders, with them all pulled to the bottom it’s likely that the image will be out of frame.

Awesome! Thank you for the help!

How do we only need an 1/8" cable for sync between VC and Vidiot? The Sync on the back of Vidiot is an RCA connection. What am I not understanding? Thanks

Apologies, I mis-posted. You’ll want to use the RCA video input. Vidiot cannot receive sync via the 1/8" luma input on the back.

Hello, I have a few questions I’ve been trying to understand about the syncing the Vidiot’s RCA input and I feel this might be the right thread to ask,

Is the sync signal that goes into the Vidiot’s RCA video usable? In other words, does it feed into the the Vidiot’s Luma processor, or would I need to feed a different signal into the Luma processor from the dedicated 3.5mm input?

Also, I noticed in another thread, the second composite output of the Visual Cortex was used to sync to the Vidiot. Could I do the same with a Memory Palace? Additionally, do the rear panel sync signals contain video, or is it only timing code? Finally, would it make sense to sync up a system where the Visual Cortex is synced to the Memory Palace via the rear panel RCA sync jack, and synced to the Vidiot by way of the Visual Cortex’s second composite out?

I’ve synced both MP and Vidiot to the Cortex composite output without issues, but I am not sure if this is a “supported” configuration. The downside to syncing Vidiot this way is that the input is a video input… so the output of Cortex is now going to be feeding into your luma processor. Which I think kind answers your first question? Any video going into either luma input (RCA or 3.5mm) gets mixed together into the luma processor. If it’s only a sync signal then you can patch into the 3.5mm jack to have video+sync together

Oh but it should be mentioned that everything has to share common sync, so if Cortex is synced separately from a video source, then you won’t be able to use a second video source going into Vidiot unless the two video sources are externally synced together.

You can work around this if you have a video mixer, perhaps. Some have the ability to display opposite buses via the “program” and “preview” outs, so what I do is connect Cortex and Vidiot to these different outs, which will be synced, allowing me to use two different video inputs between Cortex and Vidiot while staying synced together.

Awesome! Thank you very much for answering my question so quickly. I really appreciate it.

Since you are using a VHS input source, you probably need some kind of time base corrector, such as the upcoming LZX TBC2 or an outboard video mixer. See How do I avoid glitchy behavior when using VHS tapes as an input source?

I was reminded that the Vidiot has a time base corrector built in and there is a switch at the back that enables it above the video input, it has a loop logo , I flipped it up and the scrolling is fixed. Thanks!

I have a different problem in that it’s the vidiot that is scrolling when I plug into The 3trins. Yeah I know there is correction but not enough.

Does it lock okay if the 3trins output goes into the Vidiot input instead?

Yes but I’m creating feedback loops.

Realized I hadn’t actually searched the forum for a possible answer yet and well, crap, I guess I had completely forgotten that I had had problems with my ESG before and already sent it in for service once.

Ugh. Yeah.

Guess that goes to show how rarely I’ve been using my system lately. Things have been pretty torn apart since I started to build an audio rack as well (and cannibalized one of my cases for that for that last 6–9 months).

If you are in NTSC/PAL modes, Composite output should be active. If it’s not, it could be the CVBS input on your display or a bad cable. If you’re still having problems, contact support@lzxindustries.net for repair or replacement.

Ah, good call on making sure to test out multiple cables. Will include that on my list for further testing. Thanks.

DIP switches are in the right arrangement?

jep!

14pin sync from VC to the busboard sync input.
16 to 16pin power cable from your power busboard to the sync busboard
and then you have 10 x 14pin buffered sync connectors for connecting your modules.

You can cut off/remove the strain reliefs on the sync distro cable to get it to fit between Cadets.

Losing color is a strange behavior. What camera are you using? Do you have Vidiot’s sync loop through turned on? Are you patching the Vidiot rear RGB outputs to the RGB inputs on Memory Palace?
If you have Palace set to internal sync, is S-video still black & white?

After some more testing tonight, it seems the black and white MP analog video output is caused by using Vidiot as master sync source for MP. Everything seems fine when Vidiot gets sync from camera(or anything else).

For example, if my cheap mini security camera is plugged into RCA video in on back of Vidiot with loop thru switch on or off, and system is getting sync from the back panel RCA sync out/loop thru, luma or chroma, MP analog outs are in color.

If no camera or other sync source is plugged into Vidiot, using it as master sync for system causes MP to be in black and white when analog outs are going directly into TV (but color if patched into Structure)

20200324_0051543264×2448 3.35 MB

I can certainly work around this if its normal, just seems weird. Hopefully this explanation makes some sense. It’s gotten late

On a somewhat related note, is it possible to get rid of or adjust the borders around tiled image as seen in attached image?

Thank you!

Excellent. Glad you found the path!
Aside from the black & white issue, everything sounds normal–including the borders, which are an anomaly of the Vidiot/Memory Palace combo (and are a byproduct of Vidiot’s sync output).