What went wrong? AA864 / AB165 on toggle switch

[pompeiisneaks]

Here's a screenshot of what I mean. the stage I've marked as '3' seems to be missing on your standard amp, and you're using half of a gain stage on the V2 thats what it 'should' be no?

3rdStgMissing.PNG

~Phil

[fevzay]

Now I feel like a bit of a blockhead. The Y connection (220k, Buss ---> 500pf) has the 220k as the standard AA864 V3 out to the PI, and the Buss coming from the AB165 gain stage. The AB165 has a 220k resistor leading into the circuit (from the switch).

I'm sort of following how signal could bleed from there. Although I thought the 470k resistor (traced back from the buss wire and wire connected to it) would prevent this as on the original AB165? Though I realize I've added components which changes things.

Indeed the extra gain stage is 'optional,' and it is not a standard part of the AA864 amp when I built it.

I don't quite follow the other comment regarding 1/2 of a 7025 being missing. The amp has V3 as the gain stage, then off to tone stack, then recovery stage, then I go out to the switch to choose between adding V2's additional gain stage or straight to the 220k resistor feeding the 500pf cap.

[pompeiisneaks]

In the original AA864 OR the AB165 there is a dual triode used for each channel, (one tube) and then a half a triode used in between as part of the 'mixing' network before the PI. not sure if that's to help balance out the two channels, or just to provide a bit more gain,

in your original layout, you're using only two triodes, one of them is the PI, the other is the input, and recovery from the tone stack, but no half triode is being used (the third stage I mentioned above), until you added in the half a stage for the new functionality. Look at the layout of the AA864 you provided. That amp uses 4 tubes, 1 12AT7 and 3 7025's. You've omitted one input, which is just the ONE 7025, the V2 on the original layout is a required part of that amp, even if you remove the normal channel, as it comes after the mixer resistors but before the PI. It may NOT be needed since you're normally using two channels, but yours is only one, but that half of a 7025 from the layout is not in your amp anywhere, but it exists in both AA864 and AB165. Here's another screenshot of the area I mean:

thirdGainStage.PNG

~Phil

[fevzay]

Well, damn. This has proven more frustrating than I thought it would be.

Phil, I GREATLY appreciate your time and energy, but I still can't seem to grasp what is different. I felt like I did something simple, but it has proven to require more than my limited understanding on basic electronics.

To some degree, this was an attempt to learn why something didn't work. In practice, I'm going to keep the amp wired in as an AB165 front end (no switching involved). It's easy to convert to AA864 if I want to later on. The amp sounds great as is, so I'm not sure I want to tinker with it anymore.

[pompeiisneaks]

I just realized I am making this more confusing. I was wrong. The extra 1/2 triode is ONLY used in the bass channel, so ignore that deviation. On the other hand, my original statement, once I understood your wiring, makes sense still.

If you look at the next screenshot, you'll see the triode I was confusing, and can be ignored, as the 'exit' of the bass channel, in red and the 220k mixing resistor circled in red, and then in blue I did the normal channel and a blue 220k mixing resistor.

MixingResistors.PNG

You've added a 'second' channel by adding that gain stage, and in doing so, still need that second mixing resistor or it lets the signal from the one side bleed back to the other. That's the point of mixing resistors. Either that, or you need a DPDT switch that would disengage the connection from the 220k to the same stage. (The Y I mentioned before). I think if you restore things how you had them with a swtich, but just remove that jumper I've marked with an X and replace it with a 220k resistor, seen here:

ReplaceWith220k.PNG

you should be in business.


~Phil

[fevzay]

Okay, this gives me a better understanding of how things probably went wrong. Thank you.

So, replace the buss wire with a 220k. Got it. I understand why I was losing volume....I left the door open to the other circuit without it.

Two things:

1. Does that mean that the 220k resistor I put at the front of the AB165 circuit/gain stage wasn't necessary (and should be removed)?

2. Will adding the 220k resistor in the Y-junction change the tone of the AB165/gain stage circuit? It will now be in the signal path when I am on that side of the switch.

[fevzay]

All of this is really interesting to me. I'm curious of how the DPDT switch would have alleviated this issue.


Lots of options...

1. Leave it alone (AB165 front end).
2. Change two connections to revert to AA864.
3. Add the SPST switch back with additional 220k (AA864 and AB165 front end)
4. Add a DPDT switch to get same effect (both)
5. Wire a completely separate channel on the Bass side of the amp to separate them out

[sluckey]

Here's a schematic based on the hand drawn layout you posted. You really don't need either of the 220K mixing resistors because you ain't mixing anything. The SPST switch eliminates the need for any mixing. I recommend putting a .047µF cap between V3-6 and the center lug of the switch.

[pompeiisneaks]

But your schematic shows two 220k resistors, one on each leg. Or am I missing something?

~Phil

[pompeiisneaks]

More to my point, sluckey, isn't the signal, in the switch configuration you show, going to bleed some signal back down through the 470k, past the .01 cap, down the 470k to earth? This was what I though the point of the 220k mixing resistor would do, keep an even current flow that's moving towards the next stage and not allow some signal to bleed back out. (he indicated when eh engaged the AB165 stage, it was nice, but when disengaged, it got quieter, if I recall no?) Where the other option for the switch leaves the 220k side at the top just hanging with no route 'out' and it doesn't impact the signal in any lossy way, no?

I also realize now how I got partly confused about the 7025, the gain stage on the AB165 is shared between channels Bass and Normal on the AB165 but only part of the Bass channel on the AA864, this is what I missed earlier.

Help me Sluckey-won-kenobi, you're my only hope!

~Phil

[fevzay]

Here's a schematic based on the hand drawn layout you posted. You really don't need either of the 220K mixing resistors because you ain't mixing anything. The SPST switch eliminates the need for any mixing. I recommend putting a .047µF cap between V3-6 and the center lug of the switch.

Wow, thank you for the schematic and the tip. May I ask why and how the .047uf cap "works?"

It dawned on me that I wasn't mixing anything once I double checked my schematic. I think my original single channel AA864 would have been fine without the 220k resistor leading into the 500pf cap at the PI entry point.

[fevzay]

But your schematic shows two 220k resistors, one on each leg. Or am I missing something?

~Phil

He was kind enough to post my schematic (including the errors). Ha!

[sluckey]

Ain't worth worrying about. The SPST is not a good way to switch that gain stage in and out. Use a DPDT switch. Half of the switch just as shown, The other half just prior to the 500pF cap going to the PI. This will give total isolation.

[pompeiisneaks]

Very cool, that ends up completely doing what I was implying with the DPDT switch earlier, nice!

~Phil

[Roe]

Ain't worth worrying about. The SPST is not a good way to switch that gain stage in and out. Use a DPDT switch. Half of the switch just as shown, The other half just prior to the 500pF cap going to the PI. This will give total isolation.

Great but add a cap before the second switch so that you dont switch dc plate voltage