Replacing CRT pots with digipots

[yo1dog]

A few questions for analog tri/multi-sync monitor owners:

• When adjusting the picture for each horizontal frequency, can you always turn the pots to the same place each time? Meaning, do the pot positions for the best picture at 15khz change each time you switch back and forth?
• Can you harm the monitor by setting something too high or low (turning a pot all the way up/down)?

Preface: I have no personal experience with multi-sync CRTs. I have some experience with programming microcontrolers like the Arduino and Raspberry Pi, but little EE experience.

As I understand it, the problem with analog multi-sync monitors is that when they switch between 15/24/31 horizontal frequencies, you have to manually adjust the picture settings using knobs/mechanical potentiometers. This is because the monitor is analog and has no way of "remembering" your picture settings for each horizontal frequency.

My idea is to use a microcontroller to control a relay to change the horizontal frequency and servos to turn those knobs replace the mechanical potentiometers with digital potentiometers. This way the microcontroller would "remember" what the settings should be for each horizontal frequency. It would be told by the PC what horizontal frequency to switch to and it would automatically adjust the settings. This turns an analog multi-sync into a crude "digital" one.

I want to test this idea but first need to know if it is even remotely possible before shelling out $600 for a multi-sync CRT.

[yo1dog]

Rephrased the post after learning a bit more.

[nexusmtz]

Allowing for a reasonable drift over time, yes the adjustments will be the same. My way of handing my similar situation was going to be geting a usb driven variable resistance board and just put the base range that doesn't change before the board. That board was supposed to take the place of a remote adjustment board.

It never got important enough to puruse.

[yo1dog]

Thanks for the info.

It never got important enough to puruse.

Why is that? Are the adjustments when switching horizontal frequencies so subtle that it wasn't worth it? Or did you end up only rarely switching?

I was searching around for "variable resistance board" which led me to "Digital Potentiometer" which led me to products like this one: https://www.sparkfun.com/products/10613. This would be a much better solution than having servos physically turn a knob. However, I am pretty novice when it comes to EE so I would definitely need some help from the community. I would also be worried about harming the monitor.

My biggest question is this: How at risk is my monitor and/or myself while I play around with this? I will only be messing with the detached control board. Does the control board or monitor have safe guards in place to prevent damage? For example, what if you were to turn the existing pots all the way up or down? What if the existing pots were removed (infinite resistance) or shorted (0 resistance)? What if the control board was completely disconnected?

Doing some reading on digital potentiometers (digipots) vs normal/mechanical potentiometers (pots), I have come up with this list of considerations:

• Resistance range: Would need to find the resistance range the existing pots provide to ensure the digipot provides the same range. I assume this can be done by simply turning the pot all the way up and down and reading the resistance.
• Current limitations: "Maximum current through A, W or B: 2.5mA" Not sure what the current running across the existing pots are. If it is too high, what is the solution?
• Discreet steps: Pots have continuous adjustment while digipots have steps. For example, if a 16bit control is used would the 65535 steps be fine enough for proper adjustment of the picture? I would think so as 1 step on the digipot would be roughly equal to 360/65535 = 0.005deg turn of a pot which is a ridiculously fine adjustment. Or is that not how it works? What about a 8 bit controller with 257 steps?
• Resistance tolerance: Both pots and digipots have poor tolerances (+-20%). Does this only affect the pot/digipot's resistance range (ex: 10-90Kohm instead of 0-100Kohm)? Do I need to worry about this since the existing pots suffer the same problem?
• Accuracy: I saw this mentioned somewhere but I am not sure what it is referring to. What is it and do I need to worry about it?
• Temperature: Related to the previous two, temperature effect the actual resistance provided by the digipot. Do I need to worry about this?
• Linearity: I assume this means that the resistance difference between steps varies. Is this variance enough to prevent proper picture adjustment? I assume if the number of steps is high enough this variance becomes negligible.
• Noise: Do digipots produce enough noise that it would effect the picture? I would assume not.
• Zero crossing: This was frequently mentioned regarding preventing popping noises in audio when using digipots as a gain/volume control. Do I need to worry about this? You can get digipots with zero-crossing detectors built in.
• Maximum operating frequency: Does this refer to the frequency of the signal the digipot is providing resistance on? Do I need to worry about this?
• Power-up/default resistance value and off-state resistance value: If the monitor receives power before the digipot receives power or before the digipot is set, the resistance provided by the digipot is undefined and could be at the ends of the resistance range (turned all the way up/down). Could this be harmful to the monitor? You can get nonvolatile digipots that retain their resistance value without power and on power up.

Any other considerations I should know about? If someone can provide some insight to the above considerations and questions, I would be very thankful.

Thanks,
 - Mike

Edit: Here are the schematics for the monitor. I think the control board is on the last page? Maybe?

[nexusmtz]

I recall there being pcbs that had 4/8 channels (all the same) along with controlling circuitry, and they were expensive for the problem they'd be solving. They didn't seem to be for trimming, and I knew I needed something with more flexibility. For example, here are the trim pot values from the k7200 remote board (copy/paste from realbobroberts)

10K Ohm Trim Pot  -  Contrast
2K Ohm Trim Pot  -  Black Level
30K Ohm Trim Pot  -  Vertical Position
200 Ohm Trim Pot  -  Vertical Size
500K Ohm Trim Pot  -  Vertical Hold
500 Ohm Trim Pot  -  Horizontal Position

I can't use that same example to tell you the current limitations because he doesn't show the actual part numbers.

Steps, tolerance, and accuracy for the pots aren't super critical. They just need to have a range and not melt themselves or the monitor. You're not trying to calibrate a picture to a fine laser-etched overlay grid; you just want pac-man to look more like a pizza than an egg.

I've had my WG on without the board attached. It wasn't usable, but it didn't die. Then again, I didn't leave it like that to find out if it was going to.

To answer your 'Why' question, I didn't bother with it because I added a vertical cabinet (std res monitor) to go along with my horizontal, and I put a D9800 multisync in my horizontal. Both use GroovyMame, so I don't have to manually adjust very often. On the vertical, I just slide the control panel forward to get to the knobs if I need them.

It still seems like a cool solution, and I do have a couple Arduinos sitting here, so I might have order some parts and take a closer look... when everything else is done around here.

[yo1dog]

So it seems like you are saying I don't have to worry about:

• Discreet steps
• Resistance tolerance
• Accuracy
• Temperature
• Linearity

Which leaves

• Resistance range
• Current limitations
• Noise
• Zero crossing
• Maximum operating frequency
• Power-up/default resistance value and off-state resistance value

Again, my biggest unanswered question is about how vulnerable the monitor is. I would be happy to mess around with it but I don't want to fry my $600 tri-sync. I'll repeat:

How at risk is my monitor and/or myself while I play around with this? I will only be messing with the detached control board. Does the control board or monitor have safe guards in place to prevent damage? For example, what if you were to turn the existing pots all the way up or down? What if the existing pots were removed (infinite resistance) or shorted (0 resistance)? What if the control board was completely disconnected?

I made a list of all variable resistors in the schematics for the monitor. Not sure which ones are on the control board, but you can see some of them have obvious labels like Brightness, contrast, Horizontal/vertical position, etc.

[yo1dog]

We lost some posts after the last sever crash so I will try to recap:

We discussed using pots behind relays as an alternative to digipots. This would allow the CRT to act like a traditional digital multisync with settings for each horizontal frequency. The disadvantage this has over using digipots is that you can not programmatically adjust the settings on the fly (ex. on a game-by-game basis).

The biggest remaining question is: How at risk is the monitor if unspecified resistance levels are supplied? And if it is or we don't know, is there a way to design a circuit so the resistance level is always between a given range?

[obcd]

If you use the relais to add a variable resistor in parallel, the resistance will never become higher than the original one.
If you use the relais to add a variable resistor in serie, the resistance will never become lower than the original one.
You will first need to adjust the frequency with only one variable resistance, as the setting of that will influence the setting of the other frequency as well.

If the variable resistor is set as a voltage divider, you could add a capacitor so that the existing voltage remains while the relais switches between the 2 variable resistors. It will adjust to the other voltage slowly after some seconds.