Strange Aikido problem when home power fluctuates

[wicked1]

I mentioned in another post about an issue I'm having w/ my aikido. When the power in my house droops in the slightest, the audio was cutting out. I've done more testing, and found that it only does that on my tube amps. My tubelab SE 300b being the worst, a st-35 clone being slightly better (cuts out for less time when the power fluctuates) and then on a solid state amp, the audio does not cut out.
BUT, while hooked up to my solid state amp, I happened to be looking at my speaker when the power fluctuated, and the cone went through it's full extension, as if there were a HUGE low frequency sine wave. (the cone went all the way back, and all the way forward, relatively slowly).
I guess the tube amps are being pushed past their limits and are cutting out. The SS one is taking the big fluctuation and passing it to the speakers.

I'm completely stumped by this one. Can anyone give me any ideas as to what could be causing this? Is it something inherent to the Aikido power supply noise canceling scheme? Even the slightest power fluctuation causes it. Turning on a bright light, a water pump in an appliance, things that always go unnoticed are causing this strange cut out, or speaker movement.

thanks a lot!

[SY]

What does your power supply look like?

[wicked1]

6ax5-10uf-L(10h 100ma)-200uf-[here it splits to right left, so 2x]-R(1k)-100uf

I'm using .1uf caps for the noise injection (I think thats what it is) which sits on top of the voltage divider in the aikido circuit.

I can find a schematic and post specific component numbers if that would help. But, other than my power supply, it's just a standard octal aikido. All of my component values are w/in the ranges specified.

[SY]

Nothing pathological there (I've seen some attempts to choke the hell out of a supply which end up ruining the transient response). What about the heater supply?

If there's nothing evident there, the issue may be the time constant of the RC cancellation network. A schematic will help.

[wicked1]

Ok, schematics.

Now that I've gone through and looked at everything some of my heater components are off. I used a 100uf cap for c14 and 1000 is suggested. I used a 4700uf for c13, and 10k is suggested. Could the heater cause an issue like this?

I forgot to label it on the schematic, but the first resistor in the power supply is as I said above, a 10h, 100ma choke.
C6 in the aikido sch, I have labeled as 1uf, and it is actually .1uf

[SY]

C6 in the aikido sch, I have labeled as 1uf, and it is actually .1uf

That may be the issue right there. If the supply moves with fluctuations in mains voltage, the RC network can't fully cancel that- think about the RC time constant of C6 and R15.

[richwalters]

Originally Posted by wicked1

I happened to be looking at my speaker when the power fluctuated, and the cone went through it's full extension, as if there were a HUGE low frequency sine wave. (the cone went all the way back, and all the way forward, relatively slowly).
I guess the tube amps are being pushed past their limits and are cutting out. The SS one is taking the big fluctuation and passing it to the speakers.

I'm completely stumped by this one.

Sounds like the preamp is the culprit. Somewhere your circuit has got a response right down to near DC and whatever voltage fluctuations come along it forces the circuit out of the operating conditions and simply blocks. It must be a large shift. My guess is it worse if youv'e got the bass control at max. (if youv'e got one)
It won't be the heater (the time constant is too long), it's in a sensitive input part of the audio path perhaps too high Cap values.
All tube amps I've come across are completely insensitive to line fluctuations within the permitted tolerance +6%;-10%. My line varies out of tolerance from 180-236V. 180V sees the heaters drop, but that just limits emission/available power. This is one of the merits of tube amps, they simply work fussless.(should do)



richy

[SY]

One more thought- what's the input voltage to the LD1085 heater regulator chip? The nominal heater transformer (or winding) voltage?

(I'd still start by getting C6 up to 1uF or higher)

[wicked1]

I'll dig around a little more, but I don't think I've got any appropriately sized caps around. I've got some on the way, but they're on the slow boat from eastern Europe. (couple weeks since I ordered them, so should be fairly soon)

The heaters are using a radio shack 12.6v tx which is outputting about 13v. I'm running my heaters just a little light, at 12.4v. (it's the best I could come up w/ with the resistors I have on hand, to set the regulator voltage)

Thanks for your help. I'll let you know what I figure out.

[SY]

You know you can get excellent polyprop caps of the right value from Digikey or Mouse overnight...