Hi, everyone.
Preemptive apology for the Wall o' Text.
So, I have the transformers and heatsinks left over from a pair of AX100s I had built years and years ago sitting around. I had retired the AX100s and built a whole surround system of 40w Mini UP Amps and I had always planned to recycle the AX100 hardware into 100w UPs to run my main L and R speakers, but just never got around to it. Well. . . I'm finally getting around to it.
I had decided on the UP amps earlier because I had built a bunch of UGS preamps and understood the circuit really well. I've been out of the loop for several years, so I've been cruising the forums to see what's new and have come to the realization that the balanced F5 circuit is really just a more elegant version of a UGS Power amp. So, I have decided to change it up and build two Balanced F5 Turbos (V1) with 3 pairs of output Mosfets per side.
So here's where I need some help: The threads for the F5 and F5x have gotten HUUUUUUUUGE!!! I've been trying this past week to skim through them and search for information, but I've been getting frustrated. What's worse is that if someone had posted a link to another post in the thread (i.e. "There is a bunch of info in this post (link)") it just takes you to the first page. So, I'd like run through a few questions and I hope you all can steer me to the info I'm looking for.
My starting point is the schematic in post #4 of "the F5X -EUVL Builder's Thread". Is this still the "standard" version? Any clever updates I'm missing out on?
Next, If I believe correctly, the F5X circuit in question is NOT a true "X" circuit (in the SuSy sense). There are a few tidbits I found in that it may be possible to make a true balanced SuSy F5 amp. Most notably, in the F5 thread Nelson says in post #1255: ".. The most obvious is to simply take the 10 ohm resistors to ground and connect them to each other instead. The other is to make an X circuit out of it."
There are a few follow up posts with people getting excited about this prospect and the Nelson's wording seems like he's begging someone to follow up on it, but I haven't found any other discussions on the topic. It would be trivial to not use the so-called "current feedback" and instead run standard voltage feedback to the opposite inputs. Is this what he was talking about? The circuit would be "X", but it really wouldn't be "F5" anymore. I'd love to revisit this topic if there is still a puzzle to be solved. I'll probably try it both ways just to experiment. Has anyone else tried it with voltage feedback to the inputs?
Moving on, The EUVL schematic forgoes both the current limiting and thermal compensation. I'm not to worried at all about leaving out the current limiting. What I've found suggests that while the thermal compensation does speed up the warm up time and provide better absolute offset stability, its benefits don't out weigh the loss in sound quality. Trying to find some ballpark numbers: How much difference in time between warm up are we talking? 15min? 5hrs? If there is a significant time difference, I'm thinking about making it switchable, so after the amp warms up, I'll just switch the thermistors out ouf the circuit. What's the offset stability like with and without? Are we in the mV range or multiple Volts range (or somewhere in between)?
What are the details about the 5ohm Rs added at the input p-jfets sources? I understand they are for better current matching with the n-jfets. But, I haven't found a detailed discription about them. Why 5ohm? Why haven't I seen them used before?
In my UP amps, the front end rails are regulated at +/-30V with 22V unreg to the output devices, any reason not to run the first stage regulated on the F5X? I still have some regulator boards so it will be easy to do. I'm planning on cascoding the jfets and raising the value of the drain resistors to compensate for the higher rail (probably 35V ish). The unreg power rails should be close to 26V (19V secondaries). I figure, get as much voltage from the front as I can. I know the F5x is decently quiet as is, so the regulation is overkill for noise, but, hey, why not? Any pitfalls I'm not seeing here?
Again, sorry for the huge post. Hope I didn't bore everyone. If you know where the answers are, feel free to just post a link to the appropriate posts. Like I said earlier, it's practically impossible to go through entire threads; the F5 thread is 1600+ PAGES long!!
-Scott
Preemptive apology for the Wall o' Text.
So, I have the transformers and heatsinks left over from a pair of AX100s I had built years and years ago sitting around. I had retired the AX100s and built a whole surround system of 40w Mini UP Amps and I had always planned to recycle the AX100 hardware into 100w UPs to run my main L and R speakers, but just never got around to it. Well. . . I'm finally getting around to it.
I had decided on the UP amps earlier because I had built a bunch of UGS preamps and understood the circuit really well. I've been out of the loop for several years, so I've been cruising the forums to see what's new and have come to the realization that the balanced F5 circuit is really just a more elegant version of a UGS Power amp. So, I have decided to change it up and build two Balanced F5 Turbos (V1) with 3 pairs of output Mosfets per side.
So here's where I need some help: The threads for the F5 and F5x have gotten HUUUUUUUUGE!!! I've been trying this past week to skim through them and search for information, but I've been getting frustrated. What's worse is that if someone had posted a link to another post in the thread (i.e. "There is a bunch of info in this post (link)") it just takes you to the first page. So, I'd like run through a few questions and I hope you all can steer me to the info I'm looking for.
My starting point is the schematic in post #4 of "the F5X -EUVL Builder's Thread". Is this still the "standard" version? Any clever updates I'm missing out on?
Next, If I believe correctly, the F5X circuit in question is NOT a true "X" circuit (in the SuSy sense). There are a few tidbits I found in that it may be possible to make a true balanced SuSy F5 amp. Most notably, in the F5 thread Nelson says in post #1255: ".. The most obvious is to simply take the 10 ohm resistors to ground and connect them to each other instead. The other is to make an X circuit out of it."
There are a few follow up posts with people getting excited about this prospect and the Nelson's wording seems like he's begging someone to follow up on it, but I haven't found any other discussions on the topic. It would be trivial to not use the so-called "current feedback" and instead run standard voltage feedback to the opposite inputs. Is this what he was talking about? The circuit would be "X", but it really wouldn't be "F5" anymore. I'd love to revisit this topic if there is still a puzzle to be solved. I'll probably try it both ways just to experiment. Has anyone else tried it with voltage feedback to the inputs?
Moving on, The EUVL schematic forgoes both the current limiting and thermal compensation. I'm not to worried at all about leaving out the current limiting. What I've found suggests that while the thermal compensation does speed up the warm up time and provide better absolute offset stability, its benefits don't out weigh the loss in sound quality. Trying to find some ballpark numbers: How much difference in time between warm up are we talking? 15min? 5hrs? If there is a significant time difference, I'm thinking about making it switchable, so after the amp warms up, I'll just switch the thermistors out ouf the circuit. What's the offset stability like with and without? Are we in the mV range or multiple Volts range (or somewhere in between)?
What are the details about the 5ohm Rs added at the input p-jfets sources? I understand they are for better current matching with the n-jfets. But, I haven't found a detailed discription about them. Why 5ohm? Why haven't I seen them used before?
In my UP amps, the front end rails are regulated at +/-30V with 22V unreg to the output devices, any reason not to run the first stage regulated on the F5X? I still have some regulator boards so it will be easy to do. I'm planning on cascoding the jfets and raising the value of the drain resistors to compensate for the higher rail (probably 35V ish). The unreg power rails should be close to 26V (19V secondaries). I figure, get as much voltage from the front as I can. I know the F5x is decently quiet as is, so the regulation is overkill for noise, but, hey, why not? Any pitfalls I'm not seeing here?
Again, sorry for the huge post. Hope I didn't bore everyone. If you know where the answers are, feel free to just post a link to the appropriate posts. Like I said earlier, it's practically impossible to go through entire threads; the F5 thread is 1600+ PAGES long!!
-Scott