Joust with 3 invisible horizonal lines & flimsy joysticks

TrevEB

Member
Joined
Jun 4, 2009
Messages
277
Reaction score
7
Location
Oakland, California
Joust with 3 invisible horizonal lines & flimsy joysticks

I have a joust that has three invisible lines running horizontally across the screen.
When the ostrich passes through it, the lance doubles. The graphics also get screwed up.

What I know so far.

1. There are ram issues. Sometimes the ram all checks out fine, sometimes random chips fail. It's rarely the same number. I'm assuming the best route here is to just replace all of them if I can.

2. Voltages: (taken from the molex connectors)
Main board: -5.07, +11.89, +5.18, +11.89
Rom board: +5.17, +5.17
Sound Board: +11.97, +5.14, (pin 8 = -5.07)
Pin 8 should be -12vdc according to the schematics.

Thats way off, but it is on the sound board.
Seems unlikely that it would cause visual problems.

Are the other voltages close enough?

(updated)
The power supply was switched to a modern unit many years ago. The old transformer at the base of the cabinet is still connected to the fluorescent bulb, the security off switch and the power cord.

3. Its not the monitors fault. I have a new monitor installed. The old one actually has the lines burned into it.

4. all of the rams, roms and other chips have been pulled, cleaned and reseated.
Rams were relocated in a different order. This does not appear to have relocated the invisible horizontal lines.

Question 2
The joysticks seem weak, like they should have springs but the springs are gone.
Review the picture and let me know if I'm missing something.
 

Attachments

  • CIMG9238.jpg
    CIMG9238.jpg
    20.9 KB · Views: 42
  • CIMG9239.jpg
    CIMG9239.jpg
    28.6 KB · Views: 37
  • CIMG9220.jpg
    CIMG9220.jpg
    87.9 KB · Views: 41
  • CIMG9234.jpg
    CIMG9234.jpg
    28.6 KB · Views: 46
Last edited:
You need to test the voltages at the RAM chips to be sure about CPU board voltages. It is not uncommon for craks at the headers or bad filter caps to drop the voltages dramatically in the RAM chips.

Measure the voltage at the 4 corners of one of the RAM chips.
The voltages should be
Pin 1 = -5V,
Pin 8 = +12V
Pin 9 = +5V
Pin 16 = Ground
The pins are numbered starting in the upper left corner, going counter clockwise.
If any of them are low or are fluctuating more than 0.1 V you need to check your power supply.

You can check som of the logic chips in the video circuit, but I would test the voltages first and make sure they are stable.

If that is a picture of your current joystick, there is a problem. You have no leaf switches mounted, so how are you controlling your Jousters.

You should have 2 leaf switches mounted through the holes in the top bracket:

attachment.php


attachment.php


ken
 

Attachments

  • P6206742_sm.jpg
    P6206742_sm.jpg
    85.2 KB · Views: 111
  • P6206743_sm.jpg
    P6206743_sm.jpg
    89.6 KB · Views: 113
I'll check the voltages in the morning.

The voltages right at the point of connection is lower than .1 already.
Chances are its the same or worse at the ram.

I removed the leaf switches to clean and adjust. Then I took the photo showing the empty insides. The joystick works, but it feels really soft. It returns to center, but it's got little strength to it.
 
Last edited:
If you look at the second picture, there are two spring metal strips between the leaf switches and the post sticking out of the square block. Those are what force the post back into the centered position.

It was not unusual for the spring metal strips to break and have operators remove them. Some players also removed them because it was thought that this made the controls more responsive and faster to switch directions.

ken
 
Re: The joysticks. Yes, the two metal strips are what I am missing.
Where to find those?
I'll also need to find "oval" dust caps.

Also, I am wrong about the power supply. It's actually no longer in the cabinet and the big transformer thing screwed to the base is the original, um, I don't know. But it belongs there.

Yellowdog: thanks for your help. Seems like you know this cabinet.
What's your opinion re: the voltage at the sound card? Wrongly wired or are the schematics wrong? I'm assuming whomever rewired the cabinet got that wrong.

I'm also getting a lot of noise from the florescent light. (edit)
Answer: Replace the Ballast at the bulb.
 
Last edited:
The big transformer in the base is the stepdown transformer to power the original power supply board. It is also the isolation transformer to power the marquee light and the monitor. If it is working, just leave it. You need the isolation transformer or you risk damage to the monitor.

You can check the voltage output at the monitor connection or pull the connector attached to the back of the reflector in the marquee area. They should read the same as they are wired together. One reason it may be so noisy is that it is mounted on what in effect acts like a diaphragm in a speaker. The reflector will amplify the sound the fluorescent light is making. It is also possible the ballast is ready to go out on it.

It may also be in the sound card or the speaker. I have a Joust that has a buzz in the sound. I thought it was the fluorescent light (there was a different one mounted right over the speaker). But when I unplugged it, the sound didn't get any better. If I plug a spare speaker in to the sound card and leave it in on the coin box, there is no buzz. But when I plug the original speaker in. It buzzes.

I don't know of any source for the original spring metal strips. I do know that I have had several sets with an operator hack that seems to work. It involves cutting and drilling a couple of pieces of hacksaw blade to the right length. I am guessing that the hacksaw blades are tempered and springy enough to work. The first time I saw one like that I was really surprised because it was only done on one side and I couldn't tell the difference between that control and the player 2 control that was all original material.

ken
 
Voltages at the chips are:
Pin 1 = -5.15V,
Pin 8 = +11.79V
Pin 9 = +4.95V
Pin 16 = Ground

I fabulously fried at least one ram while testing. To much coffee. Now my screen is crappola. No worries. I know at least one ram was already messed up.
Thankfully ram is available, and cheap. $0.58

So, new power supply and new ram.
Unicorn has a $25 min, so I have to buy 48 chips. hmmm.
 
Last edited:
The voltages are in the right ballpark.

Unfortunately, it is very easy when testing the RAMs to have the probe slip off and short two legs together. Not that it has ever happened to me:rolleyes:.

ken
 
Bad Lines:

I wonder if we are barking up the wrong tree looking for a bad ram chip causing the bad lines. The ram check should catch it. With this problem, and the problem of the Defender with a split screen posted earliner I wonder if we should be looking at 5C,5D,5E,5F and decoder 3C. If the counters in the 5's are marginal or the decoder is marginal you can get these symptoms. I once wrote.......

"Section 2 - Memory Addressing...

There are two potential users of memory on the system. The CPU and the Video Output. The btw correlates to having 2 video decoder ROMS. Lets talk first about the CPU.
If the CPU wants to get at the RAM , it uses A8 to A15 and A0 to A7. The higher bits are first put through the video decoder ROM at 3G to produce a column address select byte (CAS) and the lower bits are used as row address select bits (RAS). Actually since the RAM only needs 7 lines, A0 is used at the bit 7 (MA6) of the CAS, and only D0 - D5 are used for CAS out of the decoder ROM.
The 74153's are toggled to select one of 4 output scenarios. CPU address lines RAS and CAS, or video address lines to the CAS and RAS. Now these address lines are presneted to every ram chip, and every ram chip then outputs its data on to its own little bus. If the CPU wants the data, it will enable the 74LS374's corresponding to the bank it wants and the data is then put onto the main data bus. Alternatively, the CPU can select the write enable lines (so can the blitter chips) to put data from the main data bus into the RAM chips for a particular bank. As you may or may not be able to see thare are many possible points of failure in this loop. History has shown that the '374's will fail from time to time causing the CPU to report bad memory chips.
The video output is a bit different however...There are 4 binary counters constanly couting up, then getting reset. The output of this count drive, among other things, address lines gor the memory (RAS) and the decoder at 3C (CAS). When the '153's are passing this data through, the memory output goes down to 8 74166's. One bank of 4 is for normal mode, the other is for cocktail flip screen mode. These chips are clocked at 6 MHZ and simply shift the data out in 4 bit blocks to the 74LS257 (which appears to only select which bank of 4 166's to select from for flip screen). This data is then fed to the 74LS257 at 2C. 2C is the chip responsible for feeding the 7489's either address lines od the RAM data. The 7489's are the colour look up RAMS, which take the 4 bit info from the '257 and produces 8 bits of colour (two banks 7489's) and that 8 bits of colour ar buffered to the transistors which control the intensity of the RGB output. Alternativly, the 2C chip can send address data to the 7489, in conjuction with the CPU setting up the data to write to the RAM. In this manner the CPU can dynamically alter the colour pallet, choosing from 256 colours to get a max of 16 colours out on the screen at a time. "

Mike
 
Let's not forget the weakest link in the chain - the sockets themselves. Your RAM chip legs can be polished to mirror shine, but if the sockets themselves are dirty/corroded/broken/poor joints, then that will reflect in the operation.
 
Well, while I wait for my new chips to arrive, I reorganized the ram, taking a guess at what ram is bad and attempting to put them at the far end. Joust booted up. Still has 3 invisible lines, but its no worse then it was before.
 
The new ram is in. The wiring is mostly back to original. Sadly the step down transformer is shot. Must have died a long time ago. Voltages are all good coming from the replacement power supply.
New decoder chips coming. After that I'll be out of ideas.
 
I am with Mike Ranger on this. It looks like a problem with the decoders or counters. Do you have a spare boardset to swap decoder PROMs?

After looking at the images for a while, I think the problem is that a line is getting copied. If you look at the bookkeeping results, the "bad" rows in the image are an exact copy of the row two pixels above it. The same is true for the player 1 lance when it is doubled. The "bad" row is a copy of the row two pixels above.

The most puzzling thing is that you are not seeing this problem on more rows. But logic can do all kinds of fun stuff when it fails. :)
 
Back
Top Bottom