John Curl's Blowtorch preamplifier part II

Status
Not open for further replies.
. . . with the power amp schematics is the VALUE of the output inductor added.
IF it is more than 2uH, look for trouble. YET, I see a number of schematics with 10uH. This is too much, way too much.
.

Thanks for the tips John. I was just about to try some experiments changing from 1.5uH to 3uH.

Are you referring to objective or subjective trouble ?

In other words, can I trust my ears is this respect or are there some underlying technical issues.

BTW - I'm really pleased with the Jfet buffers I just added in front of my amps. Sometimes I have to force myself not to be a minimalist.

p.s. was gonna try 3uH in o/p lead, then swap it to the earth lead, then split it as 2 x 1.5uH, one in each lead. All chokes will be toroids and will assess subjectively.
 
Last edited:
The output inductor is there to act like a crude diplexer and isolate the amp from the load at RF. It should be as big as necessary but as small as possible. 10uH is 1.2R at 20kHz so could cause a little HF droop. This is an 'underlying technical issue'.

Splitting the inductor is unlikely to achieve anything. It could marginally reduce its effectiveness. Putting it in the ground lead to the speaker could reduce its effect even more. By 'effect' I mean the wanted effect of RF isolation. The unwanted effect of HF droop will remain the same, therefore the best place is a single inductor in the signal wire to the speaker.
 
The roll off of treble is well know - was wondering about anything beyond that.

I understand the normal practice but I have been focusing on reducing RF ingress into the amp and particularly keeping the system earth as clean as possible. Not expecting ground breaking results but I'm curious to see if I can hear any difference.

People have told me that there is absolutely no difference between having 4 x 2mH chokes in the all four lines from the transformer and having 2 x 4mH chokes just in the Pos & Neg rails ( both in CLC configuration ) but my ears tell me a different story so wanted to test the similar principle here.

Have made 2 x 1/2 toroidal elements that combine into a full toroidal coil. The inductance just more than doubles if the coils are in close proximity 1.21mH, & 1.22mH combine to 2.49mH

mike
 
Last edited:
I would like to address something that might make a difference with aspiring audio designers and builders.
I have noted on other threads where many schematics, both old and new, are put up proudly, as favorites. Not to dismiss any particular schematic, I would like to point out some potential compromises that many of these schematics show.
The first, with the power amp schematics is the VALUE of the output inductor added.
IF it is more than 2uH, look for trouble. YET, I see a number of schematics with 10uH. This is too much, way too much.
A second clue is how many power output devices are there for a given Wattage. IF there is only a single output pair, you are compromising peak current drive, as well as INVITING serious safe area protection triggering with real music playing at some nominally reasonable level. More output pairs give more peak current, AND a potential reduction in the protection circuitry sensitivity. This can be very important.
Look back at many of the schematics recently put up on this website, and take note.

Good info John on the output pairs. Lynn Olson in his thread in the speaker section has been in favor of no more that two output devices, for reasons that he goes into in the thread. Your experience designing power amplifiers speaks volumes.
 
My smallest amp has 2 pairs, the amp that I am using for my hi fi is 8 pairs, and the JC-1 uses 9 pairs. IF the output devices were super strong, and high current, then a single pair would do OK, even for a fairly powerful amp. But, most devices current limit below15 A or so, and start breaking down at 50V (timed breakdown) so parallel devices help us achieve +/- 90V operation, without protection circuits continually activating.
 
My speakers are 8" & 8ohm and dip to around 6ohm, 89db /W/m, The amp o/p rarely peaks above 10V so one o/p pair seems fine for me in this set up.

Perhaps British listening rooms tend to me smaller than US ones.

John, you have to design amps that will work in any situation, any load. My amp only has to drive my speakers in my living room.
 
Last edited:
My speakers are 8" & 8ohm and dip to around 6ohm, 89db /W/m, The amp o/p rarely peaks above 10V so one o/p pair seems fine for me in this set up.

Perhaps British listening rooms tend to me smaller than US ones.

John, you have to make amps that will work in any situation, any load. My amp only has to drive my speakers in my living room.

Whats Zmin mag/phase? And its not only SOA , you have to listen to the damn thing too , sounds like you listen nearfield and to dynamic limited music ..
 
Last edited:
Let us estimate the capacitance loading of this Quad loudspeaker:
 

Attachments

  • quad_impedance_graph.jpg
    quad_impedance_graph.jpg
    15 KB · Views: 185
Rick,
Now that you said it you are right. I didn't agree and don't with some of his ideas but I do like things he says.

Me to, I think he is an excellent writer, and always look forward to reading his posts.

You have to have multi output devices be it SS or tubes if you want high current . Lynn does not need high power with his speakers, so one pair could work for him, IF he could find a SS amp to his liking. I wonder if he ever heard Johns ML-2? I did in 1980 and it was one sweet sounding amp. Pure class A power. :)
 
Status
Not open for further replies.