The positive feedback provided by the 220K off the cathode leg of the inverter helps achieve some OLG within the GNFB network. It's inclusion does throw the balance of the inverter off, such that the inverter's bottom leg 68K should really be about 52K for balance. As but one example, Dynaco used an identical approach in the ST-35, and used imbalanced phase splitter resistors to compensate for it. Again, for the most accurate balance, the bottom output tube coupling cap should be fed from the junction of the 1k cathode resistor, 220K feedback resistor, 470K grid resistor, and cathode load phase splitter resistor.
Using direct coupling into the splitter from the previous stage will surely eliminate some of the components in the splitter itself, and is a good idea if only from that point alone. From a theory standpoint, that will also allow for the application of more stable negative feedback. In reality though, MI amplifiers use such little feedback that converting to direct coupling for the purpose of being able to apply more is of no real value in this case. Still, it would lower the parts count -- and eliminate that pesky cathode cap, too. It also was there to try and maximize OLG before application of FB.
Since the original unit was a PA amplifier, and the intended application is for a bass guitar amplifier, I hope adequate tests were done to determine if the output transformer is up to this kind of service. The transformer will need to deliver full power output down to at least 40 Hz for this application, which many PA amplifiers can do -- but not all. Otherwise, this might end up being a great guitar amp, but not for bass.
Dave
Using direct coupling into the splitter from the previous stage will surely eliminate some of the components in the splitter itself, and is a good idea if only from that point alone. From a theory standpoint, that will also allow for the application of more stable negative feedback. In reality though, MI amplifiers use such little feedback that converting to direct coupling for the purpose of being able to apply more is of no real value in this case. Still, it would lower the parts count -- and eliminate that pesky cathode cap, too. It also was there to try and maximize OLG before application of FB.
Since the original unit was a PA amplifier, and the intended application is for a bass guitar amplifier, I hope adequate tests were done to determine if the output transformer is up to this kind of service. The transformer will need to deliver full power output down to at least 40 Hz for this application, which many PA amplifiers can do -- but not all. Otherwise, this might end up being a great guitar amp, but not for bass.
Dave
Rev 3
Bias points are adjustable, Plate and Screen to output tubes are switched together and regulated separately.
I realize the phase splitter is unbalanced, so I'm curious to see what kind of distortion I get out of it. If it's ugly, I'll throw in another resistor or two to try to even it out. PS regulator much simpler too.
Bias points are adjustable, Plate and Screen to output tubes are switched together and regulated separately.
I realize the phase splitter is unbalanced, so I'm curious to see what kind of distortion I get out of it. If it's ugly, I'll throw in another resistor or two to try to even it out. PS regulator much simpler too.
So, I found my 7189s. All the markings are gone, so the best I could do would be to look at the internal structure of the tube and see if it matches any known ones. Can anybody point me to a photo gallery of 7189s so I can see what I have? The only reason I know what they are is that is what the circuit called for and what the markings on the amp chassis say.
BTW, It's a Knight KN3220A. I'll post the original schematic shortly, when I have time to clean it up.
BTW, It's a Knight KN3220A. I'll post the original schematic shortly, when I have time to clean it up.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.