ST-70 mono makeover

[DF96]

If you simply inject feedback between the LEDs, with no series resistor, then you will need quite low feedback resistors which may load the output and you will get distortion from the LED non-linearity (small, but not negligible in this situation). You must add a resistor.

You appear to have positive feedback. Reverse the OPT secondary connections. This would explain the bad bass with feedback. The 'no feedback' lumpy response would be even worse with positive feedback.

[deicide67]

Yep,

No matter how I work the feedback, it is positive feedback.

I should reverse the primary connections on each power tube? Wouldn't it be easier to just move the coupling caps from the PI?

[20to20]

Originally Posted by deicide67

Yep,

No matter how I work the feedback, it is positive feedback.

I should reverse the primary connections on each power tube? Wouldn't it be easier to just move the coupling caps from the PI?

It should come from the (+) speaker connection. Whatever the signal is there "phase-wise" must be 180 degrees from he place you put it into.

Your original plan put it on the grid of the PI. What is the phase of the signal there? Your choice of phase inverter type may be why the FB seems to be positive to the driver cathode instead of it being negative there. The phase shifts through each device. It all has to add up to being 180 degrees out from where it goes back into the circuit.

If you started with a proven design, stick to it. If you just pieced it together from different topologies then you might have a wild child to tame.

[DF96]

The simplest method is usually reversing the secondary, as I said. Swapping PI coupling caps will work, but is usually a bigger change and may introduce HF problems as the stray capacitances will be different. Similarly, swapping OPT primaries can change strays. If your wiring is short and direct, as it should be, then you won't have enough slack to swap over. The secondary, being a low impedance point, is not fussy about stray capacitance.

[deicide67]

Thanks guys!

I swapped the coupling caps at the phase inverter because it was easiest, and now I get NEGATIVE feedback

I have a 1K cathode resistor in series with a 100 ohm to ground. 4.97V on my cathode of my CA stage. At 1K, I get .78v to full output, and I'm actually hitting my 70W before clipping.

At 15w, or 10VRMS output, the output voltage does not fluctuate more than .2V from 16Hz - 20KHz. Same for low voltage output, but I wanted to see what it does at some power.

4.7K seems to be the best value so far.

Sounds very nice as well. Now, I have to build the other one, or settle on mono.

Blair

[20to20]

Originally Posted by deicide67

Thanks guys!

I swapped the coupling caps at the phase inverter because it was easiest, and now I get NEGATIVE feedback

I have a 1K cathode resistor in series with a 100 ohm to ground. 4.97V on my cathode of my CA stage. At 1K, I get .78v to full output, and I'm actually hitting my 70W before clipping.

At 15w, or 10VRMS output, the output voltage does not fluctuate more than .2V from 16Hz - 20KHz. Same for low voltage output, but I wanted to see what it does at some power.

4.7K seems to be the best value so far.

Sounds very nice as well. Now, I have to build the other one, or settle on mono.

Blair

You is FAST!

[deicide67]

It helps that it is my layout

This has been a monster learning experience for me!

I've built a few amps in my day, but only a few off the paper schematics

Blair