I'll plead the Fifth.
I've been wanting to do a bit of work on the UL mode, seeing what I can really get away with as far as the grid current limiting. If you can, pretty please...
I've been wanting to do a bit of work on the UL mode, seeing what I can really get away with as far as the grid current limiting. If you can, pretty please...
Pretty please, SIR, try to see how big of a grid current limiting resistor we can get away with in UL, SIR. Thank you, sir, this maggot will now get down and give you 50, sir.
Hi poobah,
-Chris
At a heights exceeding 6', muscle bound mods just may be able to pull this off! 😉
-Chris
At ease Sgt. Yaniger. I don't want nobody in this man's Marines calling himself a maggot unless I call him a maggot. And you ain't no maggot. Now give me 25 more. Got that Sgt.?
Sure, I can do that. It may be few days from now as my audio workshop time has been severely limited lately. Is that OK?
Sure, I can do that. It may be few days from now as my audio workshop time has been severely limited lately. Is that OK?
poobah said:
I'm not as young as I used to be. 100 is a bit much these days.
Brian, I meant when you redid your SCA35. But if you want to do it sooner...
😀
Brian.
I'm giving this (Ultralinear Mode) a go too. My interest is in modifying my "Baby Huey" (which is Ultralinear) to take advantage of the overload characteristics imparted by Stuarts LED biasing scheme. Got side tracked this week by (blush in embarassment) completing a SS Amp. Also about dissappear for 3 weeks on buisiness. Will attempt to get some results before then.
Aside: One other thing I want to progress on the design which was highlighted during work on "Baby Huey". By the time you get enough feedback (local shunt feedback rather than global in BH circuit) to get the top end sounding stunning then you have about twice the feedback required to get the bottom end and damping under control. This is perfectly understandable given that there are 2 poles associated with the OT HF response (leakage inductance and shunt capacitance givin 12dB/octave roll off) and ONLY one zero associated with low frequency OT response (Primary Inductance giving 6dB/octave roll off). This suggests to me that I need to take some of VoltSeconds (and others) ideas to find a way to compensate at least one of the HF poles associated with the OT. Limited success on this so far - and conflicting results. On a 4 x KT88 PPP UL driving Hammond 1650T I had good results with zobels between anode and screen connections. On the 6V6 version of "Baby Huey" I found the Zobel had to be across the secondary.
Cheers,
Ian
I'm giving this (Ultralinear Mode) a go too. My interest is in modifying my "Baby Huey" (which is Ultralinear) to take advantage of the overload characteristics imparted by Stuarts LED biasing scheme. Got side tracked this week by (blush in embarassment) completing a SS Amp. Also about dissappear for 3 weeks on buisiness. Will attempt to get some results before then.
Aside: One other thing I want to progress on the design which was highlighted during work on "Baby Huey". By the time you get enough feedback (local shunt feedback rather than global in BH circuit) to get the top end sounding stunning then you have about twice the feedback required to get the bottom end and damping under control. This is perfectly understandable given that there are 2 poles associated with the OT HF response (leakage inductance and shunt capacitance givin 12dB/octave roll off) and ONLY one zero associated with low frequency OT response (Primary Inductance giving 6dB/octave roll off). This suggests to me that I need to take some of VoltSeconds (and others) ideas to find a way to compensate at least one of the HF poles associated with the OT. Limited success on this so far - and conflicting results. On a 4 x KT88 PPP UL driving Hammond 1650T I had good results with zobels between anode and screen connections. On the 6V6 version of "Baby Huey" I found the Zobel had to be across the secondary.
Cheers,
Ian
Ian,
OK, please have at it. You'll probably get to it before me since I have a monster project sitting right now on the workbench ahead in the queue, little time for anything, and Sgt. Yaniger has let me off the hook for immediate results.
OK, please have at it. You'll probably get to it before me since I have a monster project sitting right now on the workbench ahead in the queue, little time for anything, and Sgt. Yaniger has let me off the hook for immediate results.
Two steps forward, one step back
Brian and SY,
I did some quick and dirty measurements with various grid stop values on Ultralinear Mode Amp last night - these where badly done without forethought for analysis BUT hre are the results:
1) Baseline:
I used the EL84 "Baby Huey", 43% UL, Individual 40mS CCS each EL84 cathode wuth 2x 470uF/50V BG bypasses and 4K7 grid stops.
- Zout = 1.5 Ohms
- Fh -3dB = 96 kHz
- Shunt feedback set resistor = 39K
2) Tacked a 1 Ohm resistor across the 39K shunt feedback set resistor i.e. Now no feedback
- Zout now 4.15 Ohms (i.e.there was 8.8 dB of feedback)
- Fh -3dB now 52kHz
3) Still with no feedback
Grid Stop = 47K + 4K7 => Fh -3dB now 33.5kHz
Grid Stop = 27K + 4K7 => Fh -3dB now 40kHz
Grid Stop = 16K + 4K7 => Fh -3dB now 46kHz
4) With Grid Stop at 16K + 4K7 removed the 1R restoring the 8.8dB Shunt feedback
Fh -3dB from 46kHz to 70kHz
The problem is of course is that those gridstop values in 3) are in series with the source impedance from the front end which in the Baby huey is a diffamp and is significant.
However we have 3 measurements and only 2 unknowns (Zs and Cin) so were all set - right??
Use the above as a guide, I will tackle the simultaneous equations later.
Cheers,
Ian
Brian and SY,
I did some quick and dirty measurements with various grid stop values on Ultralinear Mode Amp last night - these where badly done without forethought for analysis BUT hre are the results:
1) Baseline:
I used the EL84 "Baby Huey", 43% UL, Individual 40mS CCS each EL84 cathode wuth 2x 470uF/50V BG bypasses and 4K7 grid stops.
- Zout = 1.5 Ohms
- Fh -3dB = 96 kHz
- Shunt feedback set resistor = 39K
2) Tacked a 1 Ohm resistor across the 39K shunt feedback set resistor i.e. Now no feedback
- Zout now 4.15 Ohms (i.e.there was 8.8 dB of feedback)
- Fh -3dB now 52kHz
3) Still with no feedback
Grid Stop = 47K + 4K7 => Fh -3dB now 33.5kHz
Grid Stop = 27K + 4K7 => Fh -3dB now 40kHz
Grid Stop = 16K + 4K7 => Fh -3dB now 46kHz
4) With Grid Stop at 16K + 4K7 removed the 1R restoring the 8.8dB Shunt feedback
Fh -3dB from 46kHz to 70kHz
The problem is of course is that those gridstop values in 3) are in series with the source impedance from the front end which in the Baby huey is a diffamp and is significant.
However we have 3 measurements and only 2 unknowns (Zs and Cin) so were all set - right??
Use the above as a guide, I will tackle the simultaneous equations later.
Cheers,
Ian
Actually, no need to do the math- the UL just has too much input capacitance to do current limiting this way.
Really appreciate your help!
Really appreciate your help!
SY,
You MAY be right BUT I think you need to keep in mind that the source impedance from the BH diff amp (no feedback) is around 45 KOhms and that is in series with the grid stop values in my measurements.
With your 12AT7 Concertina Splitter the Source Impedance is going to be around 270 Ohms.
Theres 40 'ish KOhm you can use for a start - or is my ignorance showing again?
Cheers
Ian
You MAY be right BUT I think you need to keep in mind that the source impedance from the BH diff amp (no feedback) is around 45 KOhms and that is in series with the grid stop values in my measurements.
With your 12AT7 Concertina Splitter the Source Impedance is going to be around 270 Ohms.
Theres 40 'ish KOhm you can use for a start - or is my ignorance showing again?
Cheers
Ian
Without going and solving the equations (and I suppose I will anyway!), I can see that even if the source impedance were zero, making the current limiter larger will pull the poles down toward the output transformers'. That will not be a good thing for stability!
so Sy, you can golf your age?
And,......nice circuit. The far east and anyone under the age of 15 will love it! But the LED's need to be blue.
Too simple,as well, needs more transistors, maybe a few more caps and resistors, buncha compensation and feedback circuits,and presto! a transistor amp in disguise.
Is this where you guys gather together and throw me off a cliff?
But serially..a very nice and simple circuit. Just the way I like 'em. Tiny parts. big sound. The circuit is pretty. Always a good sign.
And,......nice circuit. The far east and anyone under the age of 15 will love it! But the LED's need to be blue.
Too simple,as well, needs more transistors, maybe a few more caps and resistors, buncha compensation and feedback circuits,and presto! a transistor amp in disguise.
Is this where you guys gather together and throw me off a cliff?
But serially..a very nice and simple circuit. Just the way I like 'em. Tiny parts. big sound. The circuit is pretty. Always a good sign.