BZone /Blank Problems

[guddler]

Ok, this is starting to drive me potty now! Don't know how much people know about BZone but there's also some general signal stuff in this post too.

As previously mentioned this is a board from one of the DoL Battle Zone cabs. I've got the game running - albeit with a mathbox error that I'll get to later when the AVG is fixed. The picture is fine apart from the Z channel.

Normally the picture is dark to non existent and I have no Z control (my rig has pots for X, Y & Z so I can adjust Z when it's working). I'm not using a normal vector screen so I have no fear of screen burn, so if I turn the picture up nice and bright it's all there and being drawn lovely, but obviously because we're up so bright I have retrace lines.

If I snip the /ZBLANK line then the picture becomes considerably brighter and I now have Z control. So I believe my issue to be with the /ZBLANK signal. Sure enough, on tracing things back through, there is a 74LS74 at J10 which has /STATCLK and DVY10 being fed to it via an LS32 as the clock, DVY8 as the data, both of which appear to be doing their thing nicely, yet the output at Q is permanently low, giving us a permanently active ZBLANK signal.

SO, easy enough, right? Signals in, nothing out, so either the 74LS74 is bad, or the next one down the chain is holding the output low? Well, wrong. On the 7474, Q and /Q are directly opposite, so actually the outputs are fine.

So what's up then? Everything looks good on the inputs. Or does it? My fluke's logic probe is showing things to be good, It's telling me that /STATCLK is about 2.2Mhz so seems ok, but when I look at it with a scope it looks permanently high. Odd. Very odd.

After mucho messing about with the scope I can finally see the low pulses of the /STATCLK signal. But they're barely there. Frankly, I'm surprised the Fluke is picking them up. So I don't know what the problem is here, whether they need to be a stronger signal, or a longer duty cycle?

Any ideas where I might go from here? I figured I'd try to stretch the length of the duty cycle of the pulse by adding a 0.1uF ceramic cap between /STATCLK and GND and while it has an effect, all I get is a few retrace lines then so that's a bit odd. It does this regardless of whether I cap it to GND or +5.

Any thoughts??

(I'll add a bit more to this in a bit)

 

[guddler]

Right, following on from the previous post, as well as /STATCLK exhibiting the odd behaviour on the scope, so does /SCALELD. In hindsight, this isn't surprising since these signals are both derived from the same signal, being /STROBE2. /STROBE2 is first OR'd with OP2 and then the output from that gate is OR'd with both DVY12 and /DVY12 so give /STATCLK and /SCALELD respectively (all gates of K10). Lo and behold, /STROBE2 looks similar on the scope to the offending signals.

So my first thought is to swap out the 74LS42 that is producing the strobe signals as outputs from the state machine, but I don't particularly hold out much hope of that making any difference since if the strobe and latch signals were not ok then things would lock up real quick!

I'll update this evening but my guess is that I'm going to be back to looking at that signal and thoughts on what may be causing it to look so weak / marginal.

 

[guddler]

So does no-one know anything about BZone PCBs then? Even with the major influx of new members recently?? Poor show! Ok, actually, P-Man does but he's in Italy.

As suspected, it wasn't the 74LS42 causing the problem so I got my logic analyser onto the case. The 7474 at J10 isn't showing correctly on the LA so I need to investigate further. I did notice that the schems call for a plain 74S74 which I don't have so I fitted a socket to experiment with a few different types (HC, F etc.). At the same time I also de-soldered the LS86 @ B7 (or was it B6?) so that I could lift the pin that is where the output goes after the 7474. Sadly I never got to look further with the LA as at this point the board has crapped out.

All I'm getting now is a quick flash of the cross hatch pattern and then a reset. So something in the state machine side of the schematics is now playing up. It looks like it's when it tries to draw the text at the bottom of the screen, it kind of looks like it can't draw up or left, only right and down and then fairly quickly crashes.

On asteroids I'd therefore be looking at the Alphanumeric section of the state machine / counter area but BZ doesn't appear to differentiate between text and lines.

Possibly something in the stack? Maybe text uses enough jumps to push deeper into the stack than lines and it's crashing? Need to have a think about that but it's pissing me off so for now I need a break from it!!

 

[Mitchell Gant]

So does no-one know anything about BZone PCBs then? Even with the major influx of new members recently?? Poor show! Ok, actually, P-Man does but he's in Italy.

You seem to know plenty about it already!

Only thing I can think to add is that I remember looking at a friends BZ many years ago that had Z blank problems. I seem to remember the 'Window' circuit with some LM319 comparators being the culprit after much staring at schematics and a 'scope.

It sounds like you're already back past the window part on yours though from what you described.

My friends BZ was starting to develop problems on a more regular basis though, went through several 2114's (always unreliable IMHO), ROM fault, and more even though 5V was good and stable. I think we gave up on it in the end as he lost interest.

If yours has also deteriorated further then it points to multiple faults which is not good.

Please let us know how you get on!

 

[guddler]

I know plenty about Asteroids, and to be honest, I was rushing into this one hoping it was going to be a quick fix a bit too much without really taking it easy applying what I know about Asteroids to BZone. BZ is after all just the next iteration of vector hardware. Sadly though it doesn't seem to be as well documented.

I've had to give it a break for the minute. Real life has just popped it's ugly head and got in the way and from this point in, I don't think I'm going to see any leisure time now for probably 3 to 4 weeks. Nothing bad though so I'm looking forward to it

Getting back to BZone. Man, is there anyone who has worked on one who hasn't had issue with that Window circuit!! It seems a bit odd but when you read Jed Morgalan's (SP?) explanation of it then it makes sense. The last BZone I looked at that I documented turned out to have a problem with that circuit too.

I think this one's problem is just prior to it. I'm really hoping so as someone's hacked about royally with the TL084 with caps soldered to lifted legs and all sorts! Knowing my luck, once I get the /STROBE and the /STATCLK sorted that will be my next issue

Oh, of course, that's after I've got the board running again!

 

[guddler]

Ok, so I'm a bit pissed off tonight!!

You see, all the way through the above troubleshooting I was assuming I had a video blanking issue that try as I might I just couldn't track down. Well, tonight I happened to plug some faulty Star Wars boardsets in to look at and noticed they had the same issue. Too bright with retrace lines and no control on my rigs Z input adjustment pot.

Feck me, if that doesn't mean it's been my rig all along and not the BZone boards. After a bit of wiggling and re-inserting of Ampex connectors my Z signal (well, colour sigals since this is now SW) are back. Grrr.

Sadly the last thing the BZ board did was die properly so I now have bigger fish to fry otherwise one of the Duke Of Lancaster machines would have been up and running by now.

Not a happy bunny!

 

[IDCHAPPY]

Ooops,

. Atleast you know the board did work though?

 

[P-Man]

my DoL battlezone mini is in my lounge right now. Right shed it is. Don't remember it being this bad.. i'll do a restoration write-up when i get to it

Yes i used to know a lot about battlezone pcbs, but in all honesty, its been 6 years or more since i worked on them. I now have a machine in my lounge, several pcbs on my shelf, and several more in for repair, so i'm going to attack them rather soon, but i can't offer any nuggets of wisdom to you Martin until i get my head back in them.

Right now, my bz mini has no high voltage, so i can't hear chatter if there is any whcih i doubt... no sound from the speaker tho, not even a hum...

 

[guddler]

I'm in the same boat (no pun intended). The cabs pretty rough, but I guess with work it'll scrub up ok. It'll never be great unless someone does new control panels. Not sure if I've got HV or not, can't remember. I thought I had a glow in the neck though? Just can't remember if that runs off the HV or not.

And yep, also got a load of boardsets to do. But zero interest in doing them!

The one thing that would be handy to know when you do get to it is whether there's an easy way to isolate the program side from the vector generator side like there is with Asteroids. That would be damn handy if there is. But the PCBs been put to one side again for some time now anyway