Jump to content


Recommended Posts


If you have some experience in bread boarding or wire wrapping; if you have the expertise in creating commercial grade PC boards and have the desire to upgrade your Astrocade video output from RF modulation to composite video, here are some upgrade options for you.

The options described below have been tested and are utilized by MCM Design. Their performance has definitely spoiled me. The only time I will use RF modulation is when I am troubleshooting someone's Astrocade motherboard.

The documentation below is in 2 parts. There's an elaborate upgrade and a simplified, piggyback upgrade. Both upgrades perform the same. The elaborate upgrade is more pleasing to look at plus offers a "go-between" audio upgrade to help protect the custom I/O chip's audio output line.

I recommend beginning the upgrade by bread boarding the simplified, piggyback option. It's the easiest, fastest startup.

Read both parts 1 and 2. If you follow all the recommendations, you should expect to see a major improvement on your TV screen compared to RF modulation. Attached are schematics and clear photos to help guide you into the world of composite video.

PART 1 - The Elaborate Upgrade

The Astrocade's Aztec RF modulator utilizes the LM1889 TV video modulator chip. Fortunately, all of the electronic RF components within the LM1889 scheme can be removed and still generate a composite video colored output.

Seventeen resistor and capacitor (RC) components in the LM1889 scheme help process 5 video related signal lines from the Astrocade motherboard, fed to the Aztec modulator via a quick-connect 8 pin connector. There are an additional 7 RC processing components near the motherboard's custom data chip, which are also connected to the 5 video related lines.

The 8 pin connector also provides +10v, -5v power lines plus an audio output line. The underside of the Aztec metal case is grounded to the Astrocade motherboard via a copper flex connector attached to the Astrocade's bottom RF shielding, which is fastened using tiny screws to the motherboard gnd contacts.

MCM Design has bread boarded, tested and wire wrapped a quick-connect board which includes:

17 RC component LM1889 video processing scheme
composite video driver
audio driver
3.5 mm audio/video output jack.

Rather than describe this board, attached to this posting are 4 photos and 2 electronic schematics.

The A/V board has room to accommodate one of 2 options for a video driver:

Option 1     The surface mounted THS7314 driver,
Option 2     Datamax UV-1R two-transistor configuration driver.

The THS7314 option is already posted on the Bally Alley website in the Miscellaneous Hardware Documents section.

Which video driver is the better choice?

The THS7314 driver might be a little clearer. The comparison is too close to determine which option is visually better. One obvious distinction was observed. The THS7314 driver displayed the somewhat transparent bottom text in the Astrocade Menu, differently.


1. The LM1889 scheme has some TV display graphic issues. For example, in Checkmate, there typically is a narrow white line along the top and sides of the score fields and the play area. The 2 video drivers documented in this report will NOT correct the already existing graphic issues generated by the LM1889 scheme.

2. An audio driver was included to help protect the custom I/O chip, should the audio cable accidentally be plugged into an OUTPUT jack on a TV (or remote audio amplifier) or if the TV audio input has some kind of a failure issue.

3. The audio video output jack faces the left side of the Astrocade instead of its back, which means the left side of the Astrocade console must have an opening to accommodate the A/V cable. This choice was made because MCM Design's hi-res Astrocade console (now referred to as the Ultimate Astrocade) will have a keyboard connector mounted on the left side and near the back to accommodate an upgraded 24-key keyboard that will be located in front of the Astrocade.

4. Video driver transistors can NOT be plugged directly into a breadboard. Apparently, the underlying breadboard contacts affect the transistors video performance (incorrect colors). Instead, the transistors must be hard wired onto a tiny Vector Board, then placed on top of the breadboard.

Tested substitutions for 1st choice MPSH69 PNP transistor
2N3906  ok
2N4125  ok
NTE159  no, slightly blurry text/graphics

Tested substitutions for 1st choice 2N4401 NPN transistor
2N3904  ok
2N4123  ok
TP3565  ok
MPS5179  ok

6. To lower the risk of line noise:
     The LM1889 chip was located close to the Astrocade 8 pin connector.
     The A/V output jack was located close to the video driver output line.

7. For the LM1889 scheme, because one .02uf cap was missing in MCM Design's inventory, two .01uf caps were wired in parallel.

8. The LM386N chip has a very low output impedance, capable of driving an 8 ohm speaker. The A/V board's audio output line was surprisingly very quiet (in most cases, no line noise when hooked up to a 20" Toshiba CRT TV) any time the Astrocade did not output any sound.

9. The A/V board gnd can be connected to the Astrocade motherboard gnd using a wire with a small alligator clip on one end with the other end of the wire soldered to the motherboard gnd.

10. The THS7314 chip, at this time, is in stock at Mouser Electronics, part# 595–THS7314D.

End Of Posting
MCM Design
June 2024


AV Board 1.jpg

AV Board 2.jpg

AV Board 3.jpg

AV BreadBoard.jpg

AV Scheme1.jpg

AV Scheme2.jpg

Link to comment
Share on other sites

PART 2 - Simplified, Piggyback Upgrade
Piggybacking 2 Composite Video Driver Schemes onto an Aztec RF Modulator

Part 2 details the 2 video driver schemes described in Part 1 for use with the existing Aztec RF modulator normally used in the Astrocade home computer.

Photos are attached showing the bread boarded piggyback setups.


A long time ago I discovered that replacing the 3.3K R33 resistor on the Astrocade motherboard with a lower resistance will increase the clarity on the TV display.

The R33 resistor, wired in series with the custom Data chip's video output line at the chip pin 24, is located near the RF modulator 8 pin connector. This resistor is shown on the motherboard schematic and component layout drawings in the Bally Service Manual PA-1, which is archived on the BallyAlley.com website.

Some time ago, I piggybacked a 3.3K resistor on top of R33 with only 1 resistor lead soldered, so I could push down the loose resistor lead onto the other side of R33. When the menu was displayed, there was an obvious increase in clarity when the piggybacked resistor was pushed downward to make contact with R33. Two 3.3K resistors in parallel will yield an effective resistance slightly above 1.5K ohms.

I have asked the question if lowering the series R33 resistor to 1K ohms would increase the custom Data chip's video output line current enough to create a situation which might raise the chip's operating temperature hotter than it was designed to operate at.

One could probably use a kitchen digital thermometer having a metal probe to monitor the custom Data chip's plastic package temperature. Perhaps the thermometer could sense a temperature increase comparing a 1K resistor installation versus the normal 3.3K factory installation. I haven't actually run any data chip operating temperature test.

For years, I have been running my hi-res Astrocades with a 1K R33 to get that higher TV display clarity. Because I run my Astrocades with forced air cooling, there has been no issues. The custom Data chip never runs hot to the touch.

I also suspect that both the custom Address and the custom Data chip will both operate a little hotter in the hi-res mode compared to low-res operation. I won't give my reasons for this claim here in this posting.

You only need a little air passing over the Data chip. A 120vac mini fan purchased at Jameco Electronics, set up to blow a little air over the motherboard data chip, would be sufficient.

Users who run their Astrocade with the console top in place should probably not attempt any R33 modification.


In Part 1, I indicated this scheme seemed to be just a little clearer than the DataMax UV-1R video driver scheme.

I piggybacked the THS7314 schematic, shown in Part 1, on top of a very old (45 years?) Aztec modulator. The attached photos show my set up. The motherboard still had the factory installed 3.3K R33 resistor.

You can use the small LM78L05 regulator (Jameco part #51182?) to lower the modulator input +10v to +5v or tap +5v from the motherboard C67 0.1uf bypass capacitor solder pad. One end of the cap (toward custom Data chip) is +5v. The other side is gnded.

The set up was powered on and to my surprise, the video driver wasn't working right. I saw 2 issues. The Seawolf game playing field was terribly filled with some kind of interference. What?

This driver scheme runs great on my Ultimate Astrocade (upper assembly). Then I recalled my hi-res Astrocades use 1K R33 resistors.

The THS7314 scheme utilizes a 1200 ohm resistor wired from the video input line to gnd. This resistor is probably being used as a voltage step down for the THS7314 video input. I thought the 3.3k R33 resistor on this low-res Astrocade motherboard might be creating too low of a voltage swing at the THS7314 video input.

So I replaced the 3.3K R33 with a 1K resistor. When this revision was powered on, I ran some low-res games. My reaction was wow! I ran Checkmate, the Store Demo, Seawolf and the Incredible Wizard. Now this video driver was running as good as the same scheme on my Ultimate Astrocade. What a major improvement compared to the RF modulated display.

At the time of this writing (Feb 2024), the THS7314 chip was readily available at Mouser Electronics for under $1.
595-THS7314D, $0.94, will ship separately in a mini static proof box?
595-THS7314DR, $0.70, will ship sandwiched between a "sticky tape" and placed in a static proof pouch?

Don't let the tiny size of this chip intimidate you into not choosing to build this high performance, low component count, composite video driver.

Yes, this is a tiny 8 pin SMD, however, you only have to solder pins 1, 4, 5 and 8, which are the SMD package end pins. Pins 2, 3, 6 and 7 are not connected.

For a pro-grade PC board, you would only need to provide solder pads for pins 1, 4, 5 and 8. Otherwise, you can buy a high quality SMD to DIP breakout board at www.schmalztech.com. Choose 8, 14 or 16 pin breakout boards ($1.39 each, must buy qty of 5).

I used a cheap 25W Weller soldering iron with a Weller MT1 tip. Ideally, a long pencil type tip would be a better choice.

For the first pin soldering, I taped down the chip in the desired location using masking tape. I also placed tape over the neighboring pin to help prevent solder spill over. I repeated similarly, to solder the other 3 pins.

I also used an ohmmeter to check neighboring pins for a short due to a miniscule solder spill over.


Note that this scheme may also require a 1K R33 resistor for optimum performance.

I found this scheme's performance really close to the THS7314 scheme.

You will have to deal with a few more components and choose transistors that are compatible with a video driver application. I listed some possible transistor choices in Part 1.

Note also that you can NOT plug the transistors directly into a breadboard. Apparently, the underlying breadboard connections will affect the driver's performance, producing false colors. Instead, wire wrap or solder the components on a small hobby board. Then place the hobby board on top of the breadboard. Insert the +5v, -5v, gnd, video in and video out lines into your breadboard. See the attached photo.


Add an electrolytic coupling capacitor to the audio line with its positive polarity lead wired toward the custom I/O audio output line pin 24. I used a 47uf 16v electrolytic.


When performance testing your piggyback set up, you might have to (want to) tweak the TV display parameter sharpness, brightness or color intensity to your preference.


Tired of seeing a modulated TV display with waves, swirls and snow. Consider bread boarding one of these 2 composite video driver schemes. 

Don't forget to replace the 3.3K R33 resistor with a 1.5K or 1K resistor and consider using forced air cooling.

The 8 pin modulator connector has no gnd line. If you are using a motherboard bottom RF shield, there should be a copper flex gnd that contacts the modulator case bottom. Make sure that flex contact is clean, not tarnished if you gnded the driver board via the mounting screws. Otherwise, use a jumper cable with alligator clips to gnd the driver board to the negative end of the large 1500uf electrolytic cap on the other end of the motherboard. 

The components for either video driver scheme are cheap.

Cut down a quick-connect breadboard or just mount the longer version to the removable modulator top cover using  number 2-56 screws/nuts.

Use a 3.5mm video out jack instead of RCA phono jacks.

If you are just going to use a 3.5mm male to RCA male stereo audio adapter cable, use an ohmmeter to determine which RCA plug is the video plug and tag it as "Video".

Welcome to the world of composite video. Say goodbye to modulated video.

End of Posting
MCM Design
June 2024

Piggy-Back CV1.jpg

Piggy-Back CV2.jpg

Piggy-Back CV3.jpg

Piggy-Back CV4.jpg

Piggy-Back CV5.jpg

Link to comment
Share on other sites

20 hours ago, AstroHRman said:

If you have some experience in bread boarding or wire wrapping [...] and have the desire to upgrade your Astrocade video output from RF modulation to composite video, here are some upgrade options for you.


Thanks, Michael.  I still use RF-out on my Astrocade.  It's good enough for me under most circumstances.  I have used an s-video upgrade, but the performance of it doesn't work for me (it has ghosting).  Are the colors accurate on these composite upgrade options that you describe in these two posts?


I would like to see some people upgrade their Astrocade's RF video with Michael's detailed composite video instructions.  Afterward, I would like to see before and after pictures posted here.



Link to comment
Share on other sites

I haven't spent any time running any intense visual comparison of color correctness.

However, it is obvious that the composite video (CV) colors look just like the RF modulated colors, plus the CV graphics are clearer with no RF interference present.

I can't prove my claim using my Canon camera because the camera can't capture accurately what I see visually on my CRT TV.

I could ship you the piggybacked THS7314 composite video driver on a breadboard shown in the above photo, so you can check out the driver yourself, but you would have to be willing to swap out the motherboard 3.3K R33 video line resistor with a 1K ohm resistor. You would also have to ship the piggybacked modulator back to me when you're finished testing it. If you're interested in running some tests, let you know via email.

MCM Design
June 2024

Link to comment
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

  • Recently Browsing   0 members

    • No registered users viewing this page.
  • Create New...