5200 UAV & audio board install without case drilling (2-port)

[+x=usr(1536)]

A while back, I had the idea that it might be possible to install a UAV in a 5200 without needing to also drill the case (or expansion port cover) to install A/V jacks.  The goals were as follows:

• No case drilling of any kind
• RF shields to remain in place
• Composite, S-Video, and audio outputs to be fully-functional
• RF to remain intact and functional as backup video output
• Achievable with off-the-shelf parts

How to install a UAV won't be directly covered here; there are a ton of other resources out there that explain the process better than I could.

 

This is currently a work in progress (the audio board and 4050 need to be installed), but so far it's going very well.  Some photos:

 

A rough idea of how the cable routing will work; more on that below.

 

 

Wiring at the UAV.  More on that below as well.

 

 

And, finally, the A/V cabling coiled in the factory RF cable location:

 

 

Regarding some of the points mentioned above:

• Cable routing: if keeping the RF shields (recommended), the upper shield will need to have a hole made in it somewhere to route the cabling through.  I haven't decided yet if I'll run it through the side or the top, but that's a relatively minor consideration.
• UAV wiring: had I been thinking when the UAV was ordered, I would have gone for the kit and not the plug-in version.  The green terminal block would have been substituted for right-angle header pins, and Dupont connectors used for A/V connections instead of the standard screw-downs.  The jumper block that's on there now would also have been deleted, each location would have been manually jumpered, and the 4050 would have been installed into a socket running above the jumpers.
• A/V cabling: fits in the clip / channel in the case intended to carry the RF cable and loops without issue around the stock retainers.  It's only a 6' cable, which is good from the standpoint of being able to do this, but extension cables may be required depending on how far back from the display you want to be.

Other than that, it's going very well so far.  Parts used are:

• UAV.  Got that one from The Brewing Academy.
• Ditto the Internal Audio Board.
• 8-Bit Classics' Atari XL/XE 5-Pin DIN S-Video & Composite AV Cable.  These can also be found elsewhere without the DIN-5 monitor connector used on the A8 machines.
• Jumper wires with Dupont connectors.  I used these ones from Adafruit because they were what I had available.

 

[+-^CrossBow^-]

Why did you get the internal audio board? It isn't needed as you only need to attach from the top of R50 to tap audio. It is a bit too hot from there initially so I tend to add a 2k resistor inline from it but yeah...

 

As for the top RF shield...just punch out the smaller side portion on the right of the cartridge port area. That is actually one stamped piece that is part of the top shield and with some work it will come off pretty cleanly in most cases.

 

Oh..you should be able to get by with removing the jumpers on the header. Then clip the leads on the headers close to the base as you can. Use some cut leads to solder the jumpers on the bottom side of the UAV assuming you can still pull it off to do so? And I think there will be enough clearance to solder a socket to the top of the interconnect pins on the UAV and over the old jumper block section that remains to plug your 4050 into?

 

[+x=usr(1536)]

6 minutes ago, -^CrossBow^- said:

Why did you get the internal audio board? It isn't needed as you only need to attach from the top of R50 to tap audio.

Experimentation.  It's the first time I've got my hands on one, and figured that this particular install would be a good place to fool around with it given how much space the 5200 has to work with.

6 minutes ago, -^CrossBow^- said:

It is a bit too hot from there initially so I tend to add a 2k resistor inline from it but yeah...

Yep, and the usual 2 resistor / 1 capacitor mod will be my fallback if the audio board isn't quite what I want.

6 minutes ago, -^CrossBow^- said:

As for the top RF shield...just punch out the smaller side portion on the right of the cartridge port area. That is actually one stamped piece that is part of the top shield and with some work it will come off pretty cleanly in most cases.

Good to know, and thanks

[+-^CrossBow^-]

I actually don't mind the green terminal block. But I do still crimp on male dupont connector pins to the wires, put those into a 5pin housing and then plug them into the green terminal block and tighten down the screws. In fact when Bryan was selling the UAVs originally, I would spent like an extra $1 for him to add just the terminal block and the basic UAV board. Now I just order the DIY kit version of the UAV for the 5200 and 8-bit computers and order the basic board only for 2600 and 7800 installs from TBA.

 

 

[+x=usr(1536)]

8 hours ago, -^CrossBow^- said:

Oh..you should be able to get by with removing the jumpers on the header. Then clip the leads on the headers close to the base as you can. Use some cut leads to solder the jumpers on the bottom side of the UAV assuming you can still pull it off to do so? And I think there will be enough clearance to solder a socket to the top of the interconnect pins on the UAV and over the old jumper block section that remains to plug your 4050 into?

That should work, but having already rendered one UAV possibly-dead by trying to do exactly that, I'm recalcitrant to try it again.  That said, now that this machine has a UAV installed, I may try the other one in it and see how things go - if it works, I'm not adverse to using the ugly one in there

[+x=usr(1536)]

And the Internal Audio Board is in   Also did a bit of cleanup to the wiring.  Notes on the installation follow:

 

This is it in its final location, though a lack of heavy-duty double-sided tape means that it's not yet anchored down.  Decided to separate out the audio cabling (white & brown) from the video cabling (yellow, green, blue, and black) after it enters the machine in order to make things a little easier to follow:

 

 

Bird's-eye view of both the UAV and IAB.  The brown wire is audio out ground, which is tied to an unused ground location on the UAV's terminal block:

 

 

And, finally, here's a close-up of the IAB's wiring.  Notes on what goes where, etc. below:

 

 

Notes on the installation thus far:

• It's labour-intensive.  IMHO, it's a lot simpler to skip the IAB entirely on the 5200 and use the method that @-^CrossBow^- suggested above.  However, this is an experimental installation and there's a reason why I wanted to install it beyond just my choice of A/V cable; I'll come back to that shortly.
• Actually installing the IAB isn't difficult, but there's no documentation (that I could find) showing how to do so in a 5200.  Made it up as I went along, but it worked out.

One thing about the IAB that took some minor working out: the pinout for its right-angle header.  It's described in the manual in terms of connections to other components and what those components provide, but not in terms of describing what each pin requires.  To remedy that, the following list gives those equivalents.  All wire colours are based on the pictures above:

1. Audio in (far right; orange wire)
2. SIO audio (no connection)
3. Ground (middle; black wire)
4. +5V (red wire)
5. Audio out (far left; labelled '5' on PCB; white wire)

As for the connections from the IAB to the 5200 and audio output, we have:

 

Pin 1: R41 (lower leg)

Pin 2: Unused

Pin 3: Ground rail

Pin 4: +5V rail

Pin 5: Audio output on A/V cable

 

I didn't use R50 for audio because I'm experimenting   Attaching to R41 is the older way of doing it, and I'll move audio input over to R50 in a day or two to see how it compares to R41.  Everything seems to be working fine, however.  The audio may be a bit overdriven, but I haven't really cranked the TV volume yet so am not totally certain about that.

 

As for why I wanted to experiment with the IAB: it has two audio inputs.  The one on pin 2 would normally be used for SIO audio if the board was being installed in an A8 - but because it's in a 5200, it goes unused.  This may give a spot to tie in the CX-55 2600 VCS cartridge adapter's audio, and I have some ideas on how to handle the video output from it as well.  Those are things to look into down the road, though.

 

Only remaining items: tape down the IAB, drill a hole in the RF shield for the A/V cable, and reinstall the 4050.  Easy

[+-^CrossBow^-]

Very Cool! But again, you don't have to drill a hole into the RF shielding. I don't have one on hand to show you what I was talking about in regards to removing the right corner section, but let me see if I can find a pic because again, it is a separate piece that is covering that spot to the left of the FR heatsinks and the right side of the cart port section. 

 

Okay...not the best pic but you can see what it looks like with the section of he RF shield removed. To do this I just use some pliers and keep working at it until the tack welds break free. I've removed that section three times on 5200s that required it without much issue and it doesn't leave any burrs or sharp edges that could cut into wire later. Also makes it possible to simply remove the RF shields without your wire being in the way.

 

[+x=usr(1536)]

2 hours ago, -^CrossBow^- said:

Very Cool!

Thanks!

2 hours ago, -^CrossBow^- said:

But again, you don't have to drill a hole into the RF shielding. I don't have one on hand to show you what I was talking about in regards to removing the right corner section, but let me see if I can find a pic because again, it is a separate piece that is covering that spot to the left of the FR heatsinks and the right side of the cart port section. 

I took a look at removing it, and, ultimately, I agree with you that this would be the way to go.  More:

2 hours ago, -^CrossBow^- said:

Okay...not the best pic but you can see what it looks like with the section of he RF shield removed. To do this I just use some pliers and keep working at it until the tack welds break free. I've removed that section three times on 5200s that required it without much issue and it doesn't leave any burrs or sharp edges that could cut into wire later.

Yep.  The plan is to add some heat shrink to the outside of the A/V cable for safety, make the hole slightly larger than that, de-burr it, and run everything through.  With the A/V cable being held in the clip and channel in the lower case half, there shouldn't be any appreciable movement.  Having said that:

2 hours ago, -^CrossBow^- said:

Also makes it possible to simply remove the RF shields without your wire being in the way.

I totally agree with this, and had I thought about it from the outset this is how I would have chosen to handle it.  However, given the relatively short length of the A/V cable, not losing any more of it than necessary to routing is desirable.  A way around this would be to redo all of the connecting wires with longer ones, which would make up for the difference in distance between a hole in the shield and removing the top right panel - but that's just not something I want to get into right now given how much work has gone in at this point.

 

Right now, there's about five or six hours' worth of work in the UAV and IAB installs combined, though a chunk of that has been figuring out routing, component locations, cable lengths, etc.  While not a ton of effort in the grand scheme of things, it's certainly the most I've put into any UAV install up to now, and there's still a couple of hours to go.  Putting the final screw in the case is something that is being looked forward to greatly

[+-^CrossBow^-]

Well that section being removed isn't what I normally do. You will see that there are two more RCAs jacks installed near the power DC jack. Those are separated chroma and luma RCAs that were requested to be installed. So because of that, I felt it was easier to just remove that section of the RF shield as a nice clean part. I usually remove the overhang on the upper shield that goes around the expansion port because that is also a separate tack welded piece that will come up pretty easily. That makes is far easier to remove the RF shielding in the future and again, makes it easy for me to route my wiring to the expansion plate as needed.

 

Another thing instead of drilling, would be to use some tin snips or better...a nibbler and just notch out a spot on the RF shield bottom edge along the side where you need to route the cabling that way. I do this often on colecovision and 2600 installs when needed.

 

Only 5 or 6 hours? Pffft... all of my installs take at least that long even with everything ingrained in my head on where it needs to go etc. But I'm pretty meticulous in how all of my projects end up looking. Even though I know it is rare that anyone is going to see anything inside the system on what I did. It is just that I know... and I want it to look a certain way LOL!

 

Also where you are tapping the audio from isn't that different from the top of R50 because the only additional component from the top of R50 to where you have it attached is the .01µf ceramic cap that R50 attaches to. So yeah the signal is still basically just as strong there as it would be at R50. I think when I was measure the p2p of the audio signal on my scope this weekend, it comes out to right at 2v without the speaker load. Attach the audio up to speakers and it drops to about 1.3 -1.5v which is prosumer levels at that point.

 

[+MacRorie]

On 11/16/2022 at 8:16 PM, -^CrossBow^- said:

I actually don't mind the green terminal block. But I do still crimp on male dupont connector pins to the wires, put those into a 5pin housing and then plug them into the green terminal block and tighten down the screws. In fact when Bryan was selling the UAVs originally, I would spent like an extra $1 for him to add just the terminal block and the basic UAV board. Now I just order the DIY kit version of the UAV for the 5200 and 8-bit computers and order the basic board only for 2600 and 7800 installs from TBA.

 

 

If y'all don't want the terminal block, I am *MORE* than happy to just put some connectors in there that will allow dupont wires to be used.  Just let me know!

 

-M

 

 

[+-^CrossBow^-]

On most installs I just solder in a set of 90 degree angle bent headers on the output side most of the time. But I do use the green blocks on 5200 installs since it saves me from having to use my own materials and female dupont connectors. 

 

That said, the green blocks do add some height to the overall setup, but there are ways around that. But they aren't very useful for 7800 installs that might use the mount PCB setup solution unless they are soldered in facing backwards so that the output wiring goes over the UAV vs away from it.

 

[+x=usr(1536)]

All done   Everything has been sewn up and is working fine; I'll do the 4050 later as there are some socket experiments I'd like to try and just don't have what I need for them on hand.  A few addenda:

 

Went with @-^CrossBow^-'s notch-the-base idea for the RF shield, but decided to do it closer to the point of entry for the A/V cable.  The opening has since been made less jagged.

 

 

Also took a few video captures in OBS: ran one loop of Ms. Pac-Man's attract mode under both composite and SVideo with no audio, then played the first level of Beef Drop under SVideo with audio.

 

There is a noticeable difference between SVideo and Composite in captures that just doesn't exist on the CRT.  The issue seems to be with the HDMI capture device I'm using - it's different to the one I had previously, and I suspect is the cause of colours being slightly off as well as some missed / shifted pixels.  Output from the SVideo / Composite HDMI adapter straight to an LCD TV looks fine.

 

 

 

To compare the above two videos side-by-side, visit this link.

 

 

[+-^CrossBow^-]

Yeah it really is no contest between the composite and s-video. The s-video on the 5200 is some of the best I've seen next to much higher options that include RGB in the mix. You can really see the difference with the dot crawl effect that is still present along the edges of the maze sections on Ms. Pac vs the S-video's perfect lines. 

 

You should check out Fractalus between composite and s-video. The main screen that has the title and copyright along the top and starting level 04 in the middle of the screen will really show you a difference between composite and s-video.

 

 

[BIGHMW]

Onw question before I get my Big Sexy modded, since this a/v mod maintains the RF use as an option can I still, if I wanted to, use my VCS 2600 adapter on my unit? If so, sign me up, at the beginning of the calendar year I will want to get a proce wuote on shipping and the parts/labor needed to get it done, I do already have a composite-to-HDMI converter and if I can would be more than willing to get an S-video-to-HDMI converter, to go along with my current RF-to-HDMI converter for use whenever I use the VCS adapter, in which will be pretty much moot point once I get my 7800 I acquired earlier this month repaired 

[+x=usr(1536)]

After living with this for a few days, I've come to the conclusion that the audio is definitely overdriven more than I'd like.  In normal use it's not much of an issue - just crank down the volume until it's where you want it - but for doing video captures it's significantly less than great.

 

2KΩ resistor and 10µF capacitor incoming.  A better capture device also needs to be investigated, but that can happen after the New Year.

[+-^CrossBow^-]

14 hours ago, BIGHMW said:

Onw question before I get my Big Sexy modded, since this a/v mod maintains the RF use as an option can I still, if I wanted to, use my VCS 2600 adapter on my unit? If so, sign me up, at the beginning of the calendar year I will want to get a proce wuote on shipping and the parts/labor needed to get it done, I do already have a composite-to-HDMI converter and if I can would be more than willing to get an S-video-to-HDMI converter, to go along with my current RF-to-HDMI converter for use whenever I use the VCS adapter, in which will be pretty much moot point once I get my 7800 I acquired earlier this month repaired 

Ray,

 

In theory yes. Since the UAV doesn't require the removal of any parts, the VCS adapter should still work through the RF output side. I've not tested this myself because I don't own a VCS adapter and I've never had anyone send one with their 5200s when getting them serviced and upgraded.

 

53 minutes ago, x=usr(1536) said:

2KΩ resistor and 10µF capacitor incoming.  A better capture device also needs to be investigated, but that can happen after the New Year.

Interesting you still find the audio too hot through the IAB. I figured it would already have some additional resistance and capacitance in place to handle this?

 

[+x=usr(1536)]

6 hours ago, -^CrossBow^- said:

In theory yes. Since the UAV doesn't require the removal of any parts, the VCS adapter should still work through the RF output side. I've not tested this myself because I don't own a VCS adapter and I've never had anyone send one with their 5200s when getting them serviced and upgraded.

Late edit: I just realised that you were referring to the RF side, not the composite / SVideo side.  My mistake.  Leaving the original comment in place for posterity.

 

The issue with getting the adapter to output via the UAV is twofold: one, the adapter sends its video as composite; two, the 5200 bypasses 99.9% of its video section when routing that signal.  Essentially, the composite signal is injected into the RF modulator at more or less the last possible place it can be before the modulator turns it into an RF signal.  What this means in practical terms is that a) its video never goes anywhere that the UAV can pick it up, and b) even if it was picked up by the UAV, it couldn't do anything with it because the UAV expects luma and chroma (plus sync) as its video input, not composite.  Note that I'm basing this on my reading of the 2-port 5200 schematics; anyone else is welcome to verify this as I am far from expert in these matters.

 

Audio from the CX-55 is another matter and may actually be easier to deal with (hence the IAB), but without video it's a moot point.

 

Realistically, about the only way to make this work would be to implement a comb filter capable of separating chroma and luma out from the adapter's composite signal and somehow feed that to the UAV.  However, the end result would likely look crap even over SVideo, so it's more effort to go to than it's worth for what's basically a giant corner case.

 

One thing I have been considering doing (now that I have a semi-doorstop CX-55 adapter) is picking up a parts 5200, cannibalising its cartridge slot, and using that to supply power to and take audio and video output from the adapter.  In other words, making it a really bizarre standalone 2600 with composite-only video   The advantage to going this route would be that it could be powered from USB power supplies, and as it only outputs composite there's no need for add-ons to extract non-RF video.  Getting SVideo out of it would be different matter, but in all honesty it's just not worth the effort (again) when other solutions exist.

6 hours ago, -^CrossBow^- said:

Interesting you still find the audio too hot through the IAB. I figured it would already have some additional resistance and capacitance in place to handle this?

It doesn't seem to.  Granted, I have only been using the audio input pin so far, not the SIO audio input pin.  It's possible that that one does have some damping in place, but I'll need to fiddle with it to figure that one out.

[+x=usr(1536)]

On 11/16/2022 at 10:05 PM, -^CrossBow^- said:

Why did you get the internal audio board? It isn't needed as you only need to attach from the top of R50 to tap audio. It is a bit too hot from there initially so I tend to add a 2k resistor inline from it but yeah...

Added the 2KΩ (actually 2.2KΩ) resistor onto the audio tap at R50.  No noticeable difference, though it did take the edge off of things a bit.  The best way I can describe it is that the audio seems less overmodulated, but it wasn't terribly so to begin with.  It's a subtle difference, but it is there.

 

The IAB is still installed, so I'm wondering if it isn't acting as an amplifier / normaliser for the audio level.  It does require 5VDC, so this isn't a far-fetched idea.  There is another one kicking around here; I'll see if I can't figure out if that hypothesis makes sense or not later today when I can actually take a look at its PCB.

 

My suspicion is that there's an LM358 op-amp on there.  The pictures on TBA's website don't make the part numbers visible on the IC so I can't tell for certain, but it certainly fits the pattern and would be a logical choice for this application.

[+x=usr(1536)]

The IAB was removed today.  This wasn't due to it failing or causing other problems - rather, I was just tired of having to adjust the TV volume way up or down every time I moved between the 5200 and another system   While it was working perfectly fine, the audio was a bit hotter than I preferred.  Having only the 2.2KΩ resistor on the audio line has the level about right.

 

No photos of this; all I did was unplug the wiring and remove the IAB from the 5200's PCB, desolder the +5VDC and GND leads that were powering it, and connect the orange and white (audio) wires together with a repurposed header pin.  Everything went back together OK and is working fine.