Proper wiring for potentiometer on ES15?

[bmichaelb]

I've seen the thread on using a potentiometer on the Y line of the svideo cable to try and prevent luminance clipping with the DMR-ES10/15/16's, but I'm a bit confused with the wiring in the photo.

https://www.digitalfaq.com/forum/vid...iometer-2.html

Is the Y grounds of both svideo sockets tied to the potentiometer as well as the positive Y's to make a voltage divider? I originally tried using the potentiometer as a variable resistor, with 1 outer leg tied to the middle leg, then input Y on 1 leg, and output Y on the other leg, bypassing the ground altogether. Both work, but I'm not sure which is more stable as my ES16 seems to be trying to bump it back up in some scenes that I know don't change on their own.

Also... I originally connected between the ES16 and the Pinnacle capture card, but it did absolutely nothing. It only works between the vcr and the ES16. I tested using VirtualDub's histogram. It didn't move if placed between the ES16 and the Pinnacle. Why is that?

And lastly... is it possible to cut the Chroma the same way? I tried it, but it didn't change anything. That was in a variable resistor config though, I haven't tried with a voltage divider. Unless 100 Ohms aren't enough for Chroma? I know it can be reduced using the capture card's levels sliders, but I just wanted to try in analog first.

-- merged --

Well it seems 100 Ohm isn't enough for Chroma, at least according to a Reddit post about mod'ing an svideo cable for Chroma. The user found it best at 500 Ohms. And it was a variable resistor, not a voltage divider. I'm going to get a bigger pot and try it out.

https://www.reddit.com/r/c64/comment..._svideo_cable/

[aramkolt]

I have a theory why it may not do anything in certain situations - since you are compressing the entire waveform by a fixed amount, certain devices will re-amplify the signal to get the sync pulse back to -0.3V (or -40IRE) as any drops in the known sync pulse level would be assumed to be due to voltage drop over longer cable lengths. Same is true for chroma levels - the colorburst is supposed to be a fixed amplitude and some cards may re-boost the chroma if they see the color burst being low and then assumes the chroma is also low for the rest of the waveform (which is generally a reasonable assumption for the hardware to make). Generally, you do want the luma reduction before the ES16 anyway since the ES16 may clip if incoming levels are high and there'd be no benefit to lowering those levels at the output if those details are already lost. So I'd say it is a bit of a blessing that the ES16 doesn't re-amplify luma if it sees the sync pulse is not "as deep/as negative" as it should be there. With S-Video, chroma is bypassed, so putting your voltage divider on luma alone shouldn't affect color levels, so that at least shouldn't be re-boosted.

As to whether different capture cards re-boost low signals is probably an automatic gain control thing or some internal feature set that likely varies from capture card to capture card manufacturer/model.

If my theory is correct, if you hook your VCR to the divider then to your capture card via S-Video (No ES16), it should have no effect either when you turn the potentiometer.

I would say that the original circuit is indeed a voltage divider using a variable resistor to accomplish that, but I'm not an electrical engineer or video broadcast engineer by any means haha

[bmichaelb]

Cheers. The way I understand it, is it's a voltage divider using a potentiometer, where you draw between 2 different sources, depending on which way the knob is turned. It's only a variable resistor when that spare leg is tied to the middle leg, making only 1 input, 1 output.

As for the chroma though, yeah, I get that it's bypassed in this circuit, I was only trying to see if I could lower it while still in analog, to try and fix some bleeding. If it works, it works. If not, oh well. I'm just testing.

-- merged --

I guess a better explanation would be that a voltage divider affects the voltage, while a variable resistor only affects the current, while voltage stays the same. I think that might be the issue I saw where the ES16 tried to compensate on a scene change. I went back to the variable resistor like I originally tried, and it seems to be more stable, possibly because there's no drop in voltage.

I then tried a 500 Ohm trimpot on the Chroma... no difference. Blah.

-- merged --

Just a follow up... with the 100 Ohm potentiometer configured as a voltage divider, I found that VDub's histogram would want to bump itself back up after 3/4's of a turn, so 70-75 Ohms(?). But when configured as a variable resistor, it doesn't want to correct itself. I now have a 500 Ohm trimpot, in parallel with a 330 Ohm resistor, making a 203 Ohm variable resistor. Even at full turn, the DMR still won't correct itself. And... it looks even better than the 100 Ohm at full turn. Success.

It's because I'm lowering the current, and not the operating voltage, similar to using a dimmer on an LED. Keep the same operating voltage, but lower the wattage by adjusting the current. Especially if the DMR's expecting a certain voltage on the input... if it's lower than expected, it might think it's a long cable run, and adjust itself to rectify.

[aramkolt]

Picture of what you came up might be helpful, I'm pretty sure I can visualize it, but may eventually try myself, so would be good to have for reference.

[bmichaelb]

Quote:

Originally Posted by aramkolt

Picture of what you came up might be helpful, I'm pretty sure I can visualize it, but may eventually try myself, so would be good to have for reference.

I used a 500 Ohm Bournes 3386. I bought it locally, but I'm pretty sure it's the 3386T-1-501. I tied 1 outer leg with the middle leg, then connected a 330 Ohm 1/4 watt resistor to both legs to make a parallel resistor. I then used a Raspberry Pi case to mount 2x 4-pin PCB mount svideo jacks in the holes already there for the Pi's USB ports, using wire from an old USB cable. A cheap svideo extension cable would have been easier, and cut down on the total cable length, as it wouldn't need 2 cables.. 1 in, 1 out. They come in 6' lengths, but you could just cut out 4' worth in the middle. I was thinking of that, then use a DB9 housing to mount everything in, then cut and mount the trimpot on top. Nice and compact.

The pins on the svideo jacks are pretty small, and have 2 case ground tabs either side of the pins... I simply bent them 90 degrees outwards. I could have cut them off, but I wound up soldering a lead to both jacks, so they ground together. The Y and C pins weren't too difficult to solder to after that.

I can't find the exact trimpot I have on Amazon, but they do have others. Best bet is to see if you have a local electronics supply shop nearby... it would definitely be cheaper.

https://www.digikey.com/en/products/...T-1-501/412600
https://www.amazon.ca/gp/product/B00CQNU5TK
https://www.amazon.ca/Poyiccot-S-Vid.../dp/B0BWT3LYVM
https://www.digikey.ca/en/resources/...eries-resistor

[lordsmurf]

What's your total cost to make this?
I wish the pot was recessed on the case, that'd be a slick/clean job.

[bmichaelb]

Quote:

Originally Posted by lordsmurf

What's your total cost to make this?
I wish the pot was recessed on the case, that'd be a slick/clean job.

Trimpot was like $2 or so, (maybe $5? it's been a few years), resistors were like $0.70 for a pack of 10, the svideo sockets were $15 for a pack of 5, and the Raspberry Pi case was from an old project, but was $14. And a small piece of prototype board that the trimpot is soldered to... free leftovers from a previous project. Yeah, I'm actually planning on recessing it... I was just testing to see if it actually worked first. There's 4 spindles underneath the lid to mount a small fan, so I'll mount a piece of protoboard to that, then mount the pot to the board. The depth is perfect for the pot.

I'm actually curious to see what the full 500 Ohms would do, but I want to finish the batch of tapes before testing any more. I know that while the WJ-AVE5 mixer is dark as hell, even more than the 200 Ohms, it actually looks really good passed through the DMR-ES16 when I tested it... except for the oversaturation part of the mixer. But the whites look good.

I did have to adjust the Pinnacle's proc amp sliders... I usually had it at Brightness +10, Contrast -20... but with this setup, it's +70 Brightness, +50 Contrast.

-- merged --

ok... well it seems 200 Ohms is a little too much, causing vertical jitter every few seconds. A slight backturn... less than 1/4 of a complete turn, and it's stable. I dropped the Brightness down to +65, keeping contrast the same +50. So... I'm guessing ~150'ish Ohms(?). My multimeter needs a new battery, so it's not giving consistent readings.

edit: actually, I needed to drop the contrast to +35... it was hitting a bit red at +50.

[aramkolt]

Hmm. That doesn't seem to be the voltage divider configuration like others have discussed (if it were, the output side would be the wiper and one of the legs would go to ground). This appears to just be an inline variable resistor on the signal line. You could basically do the same thing then with just cutting the luma wire on an S-Video cable and connecting that to the center conductor. If shielding were to be an issue, you could mount it inside of like an altoids box or something and just have one of the grounds make contact with the box. I personally wouldn't have a box that has two S-Video jacks, I'd just have two male S-Video ends coming out of the box so there's fewer overall connections.

Me saying it's an inline variable resistor assumes that both white wires in you picture are the luma signal in/out. I have wondered if just putting a variable resistor to ground would more or less accomplish the same thing, but it won't be hard to test once I get out a breadboard and some random variable resistors I have floating around here somewhere haha.

I always felt that even basic proc amps were quite a bit more complicated, but may be if we're messing with luma only of an S-Video signal it could just be a lot less complicated. I do think that some devices will still not like the decreased depth of the sync pulse which could then cause the device to re-scale everything back up, but perhaps not.

[bmichaelb]

Quote:

Originally Posted by aramkolt

Hmm. That doesn't seem to be the voltage divider configuration like others have discussed (if it were, the output side would be the wiper and one of the legs would go to ground). This appears to just be an inline variable resistor on the signal line.

Yep... that's what I explained in my first reply to you. A voltage divider is basically a voltage regulator to reduce 1 voltage down to a lower voltage. But with electronics, you don't really want to control their outputs that way. Instead, reduce the current like when dimming LEDs. The devices require a certain operating voltage, and reducing it can cause stability issues. And if the Panasonic expects a certain voltage on the input, and it's lower than it should be, it's probably compensating for it, which is why you can only dim it so much with the voltage divider before it wants to bump it back up on it's own. Reduce the current, and it still operates normally, so you can dim it further.

Quote:

You could basically do the same thing then with just cutting the luma wire on an S-Video cable and connecting that to the center conductor. If shielding were to be an issue, you could mount it inside of like an altoids box or something and just have one of the grounds make contact with the box. I personally wouldn't have a box that has two S-Video jacks, I'd just have two male S-Video ends coming out of the box so there's fewer overall connections.

True, that's why I said it might be easier to just get a extension cable and cut it instead, then house it in something like a DB9 housing. You know, those old serial cable connector housings. The smallest the cables come in though is 6'... at least on Amazon. But after reading your input, it would make more sense to get a male to male, and just keep a single cable. But when in a box, you can mount it on top of the Panasonic so it's easier to access when you want to adjust it. But once it's adjusted, you should be fine anyway.

Quote:

Me saying it's an inline variable resistor assumes that both white wires in you picture are the luma signal in/out. I have wondered if just putting a variable resistor to ground would more or less accomplish the same thing, but it won't be hard to test once I get out a breadboard and some random variable resistors I have floating around here somewhere haha.

Yep... 1 wire in, 1 wire out. I posted a picture showing the 2 variations of the variable resistors, depending on which outer leg gets tied too the middle leg. It determines which way to turn to raise or lower. I'm not sure of reducing ground... I've seen it in LEDs, but not with something that uses some kind data flow. I think it would create an issue. It's actually how we used to 'glitch' into D!shnetwork cards. Rapidly sending the command to write a bit of code to a locked card will do nothing... but reduces the voltage as you do it, and there's a sweetspot where it sneaks through. Shhh... don't tell anybody. Actually, it's been dead for more than a decade after the updated cards.

Quote:

I always felt that even basic proc amps were quite a bit more complicated, but may be if we're messing with luma only of an S-Video signal it could just be a lot less complicated. I do think that some devices will still not like the decreased depth of the sync pulse which could then cause the device to re-scale everything back up, but perhaps not.

As I said above, I think it's due to the drop in operating voltage that causes the thing to scale back up. It's not receiving the expected voltage. Reduce the current, not the operating voltage. But yeah, I think dimming it too much was causing vertical jitter, so maybe messing with the horizontal sync.

But... give it a try. It definitely looks better.

Cheers.

[dpalomaki]

Correct video signals levels are designed around specific circuit parameters; voltage and impedance. The industry has standardized on 75 ohms for both input and output impedance for video signals. The circuits will look equivalent to a 2-volt (peak-to-peak) constant voltage source behind a 75 ohm output impedance feeding a 75 ohm input impedance. The 2-volt being the video signal including sync. This results in the 1 volt p-p at the output/input connection we normally think about.

If any of the impedances are different the levels will change accordingly. In general adding a trimmer pot will reduce the input signal seen by the circuit; both video, sync, and color burst. How much this effects the ultimately captured image will depend on the specific configuration and and effects of any AGC in the device being fed. Connecting video inputs in parallel can result in an effective input impedance of 37.5 ohms, and a the expected 1 volt at the input drops to .67 volts.

Simply adding a pot to the input, depending on how it is wired will effectively increase the output impedance and/or lower the input impedance of the connected gear. A level control called a T-pad (essentially 3 special pots on a common shaft) can preserve impedance while adjusting pass through signal levels

Bottom line: if a piece of gear is knows to output too hot a video image the better way to correct it is to feed the signal through a good proc amp. As noted above some digitizers may not handle video signals above 100 IRE or below 7.5 (for NTSC) well.

On the other hand, most consumer audio is much less sensitive to impedance variations. In general most modern line level audio outputs are relatively low impedance (2000 ohms or less is common) and line level inputs are usually 10,000 ohms or higher. (Broadcast professional may be different.)

[bmichaelb]

Quote:

Originally Posted by dpalomaki

Correct video signals levels are designed around specific circuit parameters; voltage and impedance. The industry has standardized on 75 ohms for both input and output impedance for video signals. The circuits will look equivalent to a 2-volt (peak-to-peak) constant voltage source behind a 75 ohm output impedance feeding a 75 ohm input impedance. The 2-volt being the video signal including sync. This results in the 1 volt p-p at the output/input connection we normally think about.

If any of the impedances are different the levels will change accordingly. In general adding a trimmer pot will reduce the input signal seen by the circuit; both video, sync, and color burst. How much this effects the ultimately captured image will depend on the specific configuration and and effects of any AGC in the device being fed. Connecting video inputs in parallel can result in an effective input impedance of 37.5 ohms, and a the expected 1 volt at the input drops to .67 volts.

Simply adding a pot to the input, depending on how it is wired will effectively increase the output impedance and/or lower the input impedance of the connected gear. A level control called a T-pad (essentially 3 special pots on a common shaft) can preserve impedance while adjusting pass through signal levels

Bottom line: if a piece of gear is knows to output too hot a video image the better way to correct it is to feed the signal through a good proc amp. As noted above some digitizers may not handle video signals above 100 IRE or below 7.5 (for NTSC) well.

On the other hand, most consumer audio is much less sensitive to impedance variations. In general most modern line level audio outputs are relatively low impedance (2000 ohms or less is common) and line level inputs are usually 10,000 ohms or higher. (Broadcast professional may be different.)

I just figured that by lowering the voltage, the receiving unit might see it being low due to a long cable run or something, thus boosting it back up to where it expected it to be. But by leaving voltage alone, and only reducing current, it continues to function normally.

As for a proc amp though, I almost snagged a SignVideo dual proc amp about 2 months ago in a ebay auction... started at $135 USD, I put an offer of $200 USD. 15 seconds to the end, someone offered $215 USD... I didn't have time to offer $225... lol. It was my first auction, so I'll know better next time.

But according to 12voltvids, I'd still need a scope to calibrate it. I think(?).

[dpalomaki]

ebay has an automatic bid feature. You enter a bid amount as an initial bid that is higher thatn the previous high bid, then the maximum about you are willing to bid. If anyone outbids your initial offer eBay will enter a new higher bid for you up to your max bid amount. Keep in mind that buying video gear from ebay can be risky because many sellers may not accurately describe the condition of the item either through lack of knowledge, or shady operation.

[bmichaelb]

Quote:

Originally Posted by dpalomaki

ebay has an automatic bid feature. You enter a bid amount as an initial bid that is higher thatn the previous high bid, then the maximum about you are willing to bid. If anyone outbids your initial offer eBay will enter a new higher bid for you up to your max bid amount. Keep in mind that buying video gear from ebay can be risky because many sellers may not accurately describe the condition of the item either through lack of knowledge, or shady operation.

Yep, I read that about the bids... $200 USD was originally as high as I wanted to go, which is $280 CAD, and the current bid stayed at $155 for 2 days, then in the last 15 seconds of the 3rd day, it jumped to $215. I was p*d.

As for the risk though... ebay seems to have a guarantee of their own, as the video mixer I bought recently wasn't working properly, so I opened a return ticket, and the message said they'd also refund the shipping. They even emailed me a pre-paid shipping label. I wound up keeping it thinking I'd just get it serviced, but just found out in the last few days it's a PAL unit, while I was expecting an NTSC unit... which is why it was so dark, and colours were all whacked out. I should have sent it back.

[aramkolt]

I've never been too clear as what makes SignVideo proc amps so desirable, but I decided to take a gamble on a parts unit recently, so I'll see if I can get that working to see what all the hype is about. They do have some unique features such as split screen and a video level LED meter that shows you if the signal is going above 100IRE. I suppose one of the main things is that they have S-Video in/out, but there had to be other proc amps that had that as well? I'm not exactly sure what to google since most probably didn't have "Proc Amp" in the title of the device haha.

Since we are mainly just concerned with luma clipping and crushed blacks, why not just pass the luma wire in through a regular composite proc amp and leave chroma untouched? Should accomplish the same thing? Would be easy enough to test, I just have to get around to doing it. Would definitely apply to the original poster's concern about changing S-Video luma levels if it works which I don't see why it wouldn't.

[dpalomaki]

Quote:

I wound up keeping it thinking I'd just get it serviced,...I've never been too clear as what makes SignVideo proc amps so desirable,

FWIW: Finding service on any of the old analog video gear is difficult at best and bench time from a competent technician is costly. Most mfgrs drop support within 7 years of the end of production. One is often better off seeking a known working model.

Bottom line is that they were reasonably priced and worked. Something that cannot be said about all things that were on the market.

Quote:

take a gamble on a parts unit

A gamble it is. If you are very lucky it is just the lack of a wall wart power supply. For parts is a honest way to market an item that you cannot test or do not know whether or not it works as it should.

Quote:

why not just pass the luma wire in through a regular composite proc amp and leave chroma untouched

This might work sort of, especially if it is a B&W signal to start out. But the following are a couple potential issues to watch for depending on the unit.

1. The Y/C separation circuit used for the composite input might be a simple one that limits the bandwidth of the Y signal.

2. Running only the Y signal through a proc amp may add excessive signal delay relative to the C signal that bypasses the proc amp resulting in misaligned color information when the Y and C signals reach their destination. Running Y and C through their individual inputs on the same proc amp can mitigate this possibility.

[bmichaelb]

Quote:

Originally Posted by dpalomaki

FWIW: Finding service on any of the old analog video gear is difficult at best and bench time from a competent technician is costly. Most mfgrs drop support within 7 years of the end of production. One is often better off seeking a known working model.

We have a few repair shops in town, including one that deals with broadcast equipment. I merely meant I wanted to get the board recapped, and possibly have the various trimpots tweaked. That's it. That's all. I'd do it myself, but after opening the thing up, it's more boards and wiring that I want to deal with. My mom's husband used to own a repair shop, but he passed away a few years ago. Otherwise I would have asked him to do it.

https://youtu.be/vmT5pZxHdLg?si=YqDqoNphUufcDSwz

[lordsmurf]

Quote:

Originally Posted by bmichaelb

As for a proc amp though, I almost snagged a SignVideo dual proc amp about 2 months ago in a ebay auction... started at $135 USD, I put an offer of $200 USD. 15 seconds to the end, someone offered $215 USD... I didn't have time to offer $225... lol. It was my first auction, so I'll know better next time.

It's probably for the best.

SignVideo (aka Studio1 aka Vidicraft) had design/engineering flaws with their knobs. Any stresses, even from normal careful use, caused them to break down. Everything from knob de-solder to mainboard cracking. The values should be buttery smooth as you turn the knob, but would begin to step, then glitch step, then just be glitches outside unity.

Compounding this issue is the fact that most eBay sellers are morons when it comes to shipping, and too often I've seen knobs arrive broken and non-responsive. I really wish eBay would show when items get returns, so people could truly understand what a shit-show buying video gear off eBay is.

Quote:

But according to 12voltvids, I'd still need a scope to calibrate it. I think(?).

Utterly ridiculous. These devices are for visual corrections. Eyeballs, not scopes.

To be blunt, 12voltvids has been proving himself to be an idiot lately, when it comes to higher-end gear, like S-VHS decks, proc amps, TBCs, etc. He needs to stick to consumer VHS VCRs, which he does a great job at. But his information on the higher-end has a mix of myth and outright BS, as discussed in other threads this year.

I respect him for his VHS VCR repairs, but I really wish he'd avoid high-end gear, because it's extremely evident that he has near-zero knowledge in the area, and it continues to mislead newbies (as they don't know what they don't know).

Quote:

Originally Posted by bmichaelb

As for the risk though... ebay seems to have a guarantee of their own, as the video mixer I bought recently wasn't working properly, so I opened a return ticket, and the message said they'd also refund the shipping. They even emailed me a pre-paid shipping label.

It still doesn't necessarily work how people think. There is a dispute process, and lots of sellers use it. The eBay "guarantee" isn't guaranteed. It even varies on your age/status as a member. New eBay users often get shafted.

Quote:

Originally Posted by aramkolt

I've never been too clear as what makes SignVideo proc amps so desirable,

Range, smoothness, noise-free, s-video, and stability of analog-domain values adjustments. That's is the only reason. Most other proc amp options step (not smooth), limited range, noisy, composite-only, not stable -- or combo of those.

Quote:

Since we are mainly just concerned with luma clipping and crushed blacks, why not just pass the luma wire in through a regular composite proc amp and leave chroma untouched? Should accomplish the same thing? Would be easy enough to test, I just have to get around to doing it. Would definitely apply to the original poster's concern about changing S-Video luma levels if it works which I don't see why it wouldn't

That's an idea.

Quote:

Originally Posted by dpalomaki

FWIW: Finding service on any of the old analog video gear is difficult at best and bench time from a competent technician is costly. Most mfgrs drop support within 7 years of the end of production. One is often better off seeking a known working model.
Bottom line is that they were reasonably priced and worked. Something that cannot be said about all things that were on the market.
A gamble it is. If you are very lucky it is just the lack of a wall wart power supply. For parts is a honest way to market an item that you cannot test or do not know whether or not it works as it should.

This.

Quote:

2. Running only the Y signal through a proc amp may add excessive signal delay relative to the C signal that bypasses the proc amp resulting in misaligned color information when the Y and C signals reach their destination. Running Y and C through their individual inputs on the same proc amp can mitigate this possibility.

Yep, delay is essentially guaranteed. Must also address that.

[bmichaelb]

I'll believe 12voltsvids over you dude, any day of the week. He's actually worked in a tv studio, servicing actual broadcast equipment. Just stop it with this bullshit already.

https://youtu.be/V0evc3azqZI?si=mOSK1qNHJZ6Atrln

[lordsmurf]

Quote:

Originally Posted by bmichaelb

I'll believe 12voltsvids over you dude, any day of the week. He's actually worked in a tv studio, servicing actual broadcast equipment. Just stop it with this bullshit already.

I've never heard him discuss a former TV career anywhere, just repair shops.
Regardless:
- TV stations are/were not where proc amps were used (small studios were),
- TV stations are/were not where S-VHS decks were really used (unless small), as it was a "rich consumer" (prosumer) format, starting in the late 80s. The chroma bandwidch was not acceptable for broadcasting recordings, unless you were a tiny operation (schools, local cable co, rural, etc).

You PM'd me the 12voltvids SignVideo repair link to "prove" that he knows about SignVideo proc amps (and proc amps in general). But that's not what the video showed me.

- It's 40 minutes of him mostly rambling about multiple topics.
- He clearly does NOT understand s-video (Y/C) vs. composite, which has surfaced multiple times in his videos. Specific to this topic, he doesn't seem to realize the benefit of Y/C when it comes to proc amps. Nor the fact that VHS is a Y/C format, and NOT composite as he insists.
- He didn't even comment on some aspects of the very proc amp he was demonstration, meaning he knows very little about it. Not weaknesses, not benefits.
- He's being extremely abusive to the unit.

Look at how he treats the buttons and knobs at this exact spot in the video:
https://youtu.be/V0evc3azqZI?t=1906

People like him are why so many of these units are now failed.
- You cannot turn the knobs back and forth so quickly.
- You cannot grab the knobs so hard that you move the entire unit.
- You cannot smack the buttons, like a kid turning the light switch on/off/on/off, etc

He treats this gear like a child in a video game arcade, beating the hell out of the cherry buttons and joysticks. He does not treat this equipment like professional equipment. Use it, don't abuse it. But none of this surprises me. He constantly throws things in his videos, and is rough on everything.

He needs to stick to VHS VCR repairs, which is the only time he isn't spewing garbage.

[aramkolt]

I'm not sure how much delay an analog proc amp would add. There's no analog to digital conversions going on, so really the only delay would be additional "wire" (well, traces) that it travels through while inside of the device I'd think and electricity travels pretty fast. This could be compensated by having the chroma wire be a bit longer length if it's actually an issue I think. Should be pretty easy to visualize on an oscilloscope though if there really is any delay - just tap both the input and output on different scope channels and see if the output signal is behind the input one and have it trigger on the first sync pulse.

As for potentiometers on old equipment, shouldn't be hard to find replacements that are electrically equivalent, or cleaning them also is probably a viable option. Where I've heard people go wrong is dousing them in deoxit spray (which can soak into the composite fiber board that has the resistive carbon track on it and cause them to swell/crumble) rather than disassembling first and applying only to the carbon track sparingly. That all depends on the construction of the potentiometer though, if it doesn't have the light brown/manilla fiber board, it's probably fine to douse in deoxit. Deoxit also makes fader lube/grease to get that slightly resistive feel when turning knobs which is nice, but not required for function.