What exactly am I looking for????

[digi2t]

Hey all!

I've run into a bit of a conundrum with the Synthi Hi Fli Phaser section. Over the past months of deciphering the schematics and building the clone, I've finally come to the point where I need to address the point of "adjust for null offset". Here is the schematic;



Each six stage section of the phaser shows a trimmer, with the note "Null offset at Ux/x". I did some reading here and there, and I think I understand that this is supposed to adjust the DC offset at the end of the chain. However, the Mk.2 unit that Phil and I have been working off of DID NOT have these trimmers installed. Not that they had been removed, but rather, not installed from the factory.

As a precaution, I installed sockets on the clone, and last night I noodled with throwing a pair of trimmers in, hooking up the o-scope to the points indicated on the schematic, and tried to see what was going on here. To be honest, I'm not exactly sure what I should be looking for. I even tried with and without a tone generator, at varying outputs, and.... I couldn't see anything happening when I would twisted the trimmers. Even to my ears, nada. Zip. Phaser sounds exactly the same with or without the trimmers.

So, a) I'm beginning to think that the trimmers were ultimately left off the Mk.2 units for a reason, and b) I still don't know what the heck I'm looking for!

Can someone show me the null offset light?

[anotherjim]

It does look like what it suggests -  DC offset null adjust, so zero volts on the output of U1A with no signal. It might have been thought to be a problem with regard to the diode VDR arrangement since the centre of the diode pairs live at an average 0v. At any rate, as the lower chain is DC coupled from the upper, the upper null (PR1) would have to be adjusted first.

[Ben N]

Could it be that the trimmers were temporary, and replaced as the last assembly item with fixed resistors after DC offset null adjust? In a typical mass produced unit something like that would be cost prohibitive, but these things were always a little--atypical.

[digi2t]

It does look like what it suggests -  DC offset null adjust, so zero volts on the output of U1A with no signal. It might have been thought to be a problem with regard to the diode VDR arrangement since the centre of the diode pairs live at an average 0v. At any rate, as the lower chain is DC coupled from the upper, the upper null (PR1) would have to be adjusted first.

Now that makes sense. I'll have to probe it some more tonight. Thanks Jim.

Could it be that the trimmers were temporary, and replaced as the last assembly item with fixed resistors after DC offset null adjust? In a typical mass produced unit something like that would be cost prohibitive, but these things were always a little--atypical.

Don't believe so. I've looked at Mk.1 unit pics with trimmers, and the Mk.2 without, and there are no changes in the surrounding topology. From what I can tell, it seems that they just ignored the trimmers altogether later on. Like I mentioned earlier, to the ears, they don't make a snuff of difference.

[italianguy63]

Don't believe so. I've looked at Mk.1 unit pics with trimmers, and the Mk.2 without, and there are no changes in the surrounding topology. From what I can tell, it seems that they just ignored the trimmers altogether later on. Like I mentioned earlier, to the ears, they don't make a snuff of difference.

Maybe that's it!



MC

[Mark Hammer]

As I've probably repeated too many times, when one has a lot of phase-shift stages - which are supposed to be unity gain - there is always a risk of a wee bit of gain being created.  I can see a trimmer being useful for counteracting any gain applied to DC.  On the other hand, if you simply pick your component values so that you never get cumulative gain, there's nothing to offset.  Does that make sense?

[digi2t]

As I've probably repeated too many times, when one has a lot of phase-shift stages - which are supposed to be unity gain - there is always a risk of a wee bit of gain being created.  I can see a trimmer being useful for counteracting any gain applied to DC.  On the other hand, if you simply pick your component values so that you never get cumulative gain, there's nothing to offset.  Does that make sense?

Absolutely Mark. In this case, I'm guessing, they were perhaps ready to compensate for any potential problems that component tolerance or drift may have presented. I suppose that it wasn't that important in the end, and they did away with them. I can't hear a lick of difference with or without them, so why bother trying to tune something you can't hear, right?

[samhay]

I'm with Jim, a trimmer strapped between the positive and negative power supply is most likely doing to be there to tweak the DC offset.
I assume U1/1 means pin 1 of U1, which is the output of the final phase shift stage. Again, this makes sense, being a good place to measure DC offset.

Question: What type of op-amps are you using? The original may have used op-amps with significantly more DC offset than more modern ICs often used today.

[digi2t]

I'm using 4558's. The original I worked off of, as well as another unit I have pics of, used CA3458's. "N5558V" is what's silkscreened onto the board.

I finally put the trimmers in after all. I set it up to a slow sweep, and set the trimmers to give me an equal + / - voltage sweep, centered around 0mV. Each section now hovers around +/-1.5 to +/-1.9mV. Without the trimmers, I would see between -30 to -40mV weighted voltage on either section.

I figure that's as "null" an offset as I'm going to get. I still can't hear any difference, but it's a good exercise for me to understand what's going on.

[samhay]

That sounds like a usual story - best practice as far as engineering goes, but no audible effect.

More generally, that's a lot of stages. How does it sound?

[pinkjimiphoton]

look on dino's youtube at the pompeii