Galaxian static/hiss on audio

cwilbar

Well-known member
Joined
Sep 8, 2015
Messages
1,209
Reaction score
31
Location
Massachusetts
Working on a board. There is a general static/hiss to the audio background when no sounds are occuring.

I find Galaxian is like that a little, but this one seems to have a bit more than the usual to me.

I have verified it is coming from the Fire audio circuit. The static seems to be the frequency of the free running oscillator (7S). I determined this by grounding the control of the analog switch. There is a small amount of voltage on it (which is roughly in line with another working board) I've compared it to.

So far I have tried swapping the following:

7R 4066 analog switch
7T LM324 quad diff op amp
C26 0.01MF 50V which is in series with the output of the LM324 for the fire sound
C25 1MF 16V Tantalum which sits between control of the analog switch and ground and provides for the duration of sound production of the fire sound
I have blindly tried C29 (0.01MF) between the control of 555 timer 7S and ground.
The C28 cap had already been done along with the other electrolytics.

I'm unsure why there is a difference. A good board (low hiss) and this one both have similar voltages on the analog switch control line for that sound. But as there is a slight + voltage here, I think the analog switch is allowing some to get through. I did not have a new 4066, so I did have to test this with a used 4066 pull of unknown condition.

Looking for any recommendations of those familiar with the Galaxian sound circuits.

If there is a marginal part in there I'd eventually hit it (there are only 7 resistors, 5 capacitors, 1 diode, a 555, a 4066, and a LM324 in that circuit. Though, maybe some of the caps in the audio output final stages of amplification might serve as a noise filter.... and it might be normal, but is filtered out on a 'happy' board.

Hoping someone can shed some light on this.
 
Not one response ?

When the 'control' pin of the analog switch is grounded, the hiss stops. However, I'm not sure how this line is ever brought to ground. The diode which blocks the capacitor holding the charge for the envelope of the fire sound would block 'dc', so when discharged, the diode prevents the low state of the output of the driving chip to reach the 'control' pin of the analog switch.

Therefore it is left floating ?

I measure some mV of DC on it, and that is enough to let through the fire oscillator's 'hiss' on anything by quiet volume settings.

I have a few Galaxian boards, and they all hiss, but this one does a bit more.... probably a variance in some parts, but I'd like to quiet it down.

I was wondering if a large value resistor across the capacitor that allows the analog switch to create the envelope for this 'hiss' would bring the 'control' pin of the analog switch closer to ground w/o effecting the drain time of the envelope forming capacitor (which would change the sound of the game).

Surely someone must have some better analog electronics experience that can shed some light on this....
 
Last edited:
I'll do some more chatting with myself ;-)

So, I've tracked down that the 74LS259 that triggers the analog switches, does not have 0 volts at 0 value. Instead it is about .2V.

So, I have to next figure out how that voltage changes through the resistor and the diode. On my board that hisses much less the volts at the LS259 are just under .15. So I shouldn't have to drop it much. I'll also meter the diode and see what it's voltage drop is in comparison to my working board. Maybe the diode is marginal ?

I did try 200-300K resistors across C25, and it improves it, but the lower the resistance, the faster C25 discharges, which starts to affect the length and decay of the fire sound. It may be an acceptable solution, though I'm going to look into the resistor and the diode there and see how they compare to my reference board.

Anyone else following this ? Any comments, suggestions, observations ?

Does anyone else with Galaxian have hiss at volumes above home quiet (say at arcade volume levels) ?
 
Last edited:
Nothing I have looked into a tried has found any damaged part. All Galaxian boards I have have this his (at least on my bench rig with the speaker I am using).

This one board simpy would 'hiss' sooner as the volume was turned up.

In the end I believe this is due to component variances/etc.

I have however, found a solution.

R920 (1K) is fed form the 74LS259. After R920 is a diode (from memory D1 ?). This diode is a blocking diode to prevent the low logic state from discharging the capacitor that provides the decay sound envelope for the fire oscillator. Rather than use a 1K resistor here I decided to try another diode, thereby making the circuit two diodes in series vs the resistor and diode.

This worked great. Unlike other solutions (such as a 200K to 300K resistor accross the capacitor) which would change the rate of the decay to the fire sound, this solution does not audibly change the fire sound, but at the volume levels I'm testing at all but eliminates the hiss.

I thought I would pass this along for anyone else ever running into this.

It irks me that this is the way I solved it, as it means the board then differs from the original design, but no amount of hunting could find a failing, marginal, or out of spec component to explain it. But by trying different 4066 and 74LS259, I was able to subtly change the hiss, but not come close to eliminate it. So, I suspect it is just the amount of variance of a number of components in the circuit helping to create it.

I like the results so much, I'll probably do this to my other Galaxian board as well.

If you use this, enjoy as 'hiss' free Galaxian sound experience as you are likely to get w/o seriously upgrading some of the design.
 
Back
Top Bottom