Welcome! Need support, you got it. Or share you ideas and experiences.
Return to MagneQuest/Peerless Forum
64.12.116.16
In Reply to: RE: Hey Mike, how about this idea? posted by keto on April 27, 2009 at 16:50:05
Keto
Since no one else is answering I'll give it a go -
1) A 15K OPT on a tube with ~10K output impedance
is not a recipe for good sonics IMO.
Mike specs them at >200H and then parallel them
with the plate choke so ~150H - barely enough.
But I never like having that steep of a slope
of a loadline (ie. almost vertical).
I like the sound when it is almost horizontal.
It will give you a gain of ~5.5
from the preamp - is that enough?
Is it worth it?
2) A 600R pot on the output would seriously load
the output WITHOUT anything connected to it.
3) Small signal choke on the input,
well I am in favor of grid chokes
on ANY tube ANYWHERE.
4) The EXO-99s are Mikey's highest inductance plate chokes.
I have a pair on my 12B4s - talk about overkill 8^D
IMO #1 and #2 will seriously hamper the 76 tube.
That's my take on this anyway.
DanL![]()
Hey DanL,
thanks a lot for the ideas.
I've gone back and looked at the 76 plate curves and done a lot of rethinking, keeping in mind the "toughest load" situations of either 15K or 20K, even though from Mike's post, the Cobalt 15K B7's will be "lighter", especially if in parallel with just a C4S (my thinking -- and maybe the 15K or 20K models would apply for the Cobalt 15K B7 + the appropriate plate choke?). I'll do this on an existing platform, rather than change the amazing looking FP3 kit I've been staring at. Given a mu of about 9, there wont be a plate voltage swing of more than 20vrms, probably. I'm seeing that given the C4S plate loads + 300V regulated B+ that are on the existing platform, deepest class A would look something like this:
@5mA, 15K load, 170Vp-k, -8Vg 50vrms swing
@5mA, 20K load, 185Vp-k, -9Vg 70vrms swing
On the 76 data sheet, plate loads of 50K and 250K are mentioned, so it is a logical stretch to apply (even) a (nominal) 15K load to the plate, but the numbers look okay to me now to give it a try.
I've got to order some parts before I try this out. I'll keep the exo-010, 1:1 input transformer (thanks Mike) and not loading the secondary output in mind as I proceed.
Thanks again, --keto
Hi Dan:
You framed the questions very well...
I'll make just a few comments.
Under your first point--- the reference L of 200H is for M6. In Keto's intended application I would opt for a nickel based core which will increase the L severalfold.
Re: loadlines and "nominal" impedances and etc--- please see the following post;
http://www.audioasylum.com/forums/magnequest/messages/8798.html
I agree with you Dan--- I would not load the secondary and, instead, would do the volume control on the "front side" of the voltage amplifier.
Will also make finding a suitably good quality attentuator\potentiometer much easier.
Another suitable plate choke--- physically smaller than the EXO-99's and less expensive--- is the EXO-010. The 010 is rated at 275 henries--- which at 20 hertz would give you a load impedance of a bit over 34K. I keep thinkin' that if it were my project I would first consider using the 010. Smaller core, smaller winding lengths equals less capacitance that must be driven.
grid choke on input side of voltage amp--- yep--- though I might also look at using a 1:1 with 10K volume control (think high quality Penny & Giles 10K stereo unit).
Thanks Dan for responding. I've been sidelined a bit--- seems I can work a few days and then must give my ribs a few days "off"--- just saw my family doc yesterday and will soon be seeing a muscle, bone, and joint specialist. This shit has to get better sometime soon.
MSL
![]()
Builder of MagneQuest™ & Peerless™ transformers since 1989
Mikey
Glad to help.
Yes the tube's load is the primary L
in parallel with the secondaries load.
that makes for an almost vertical LL.
I forgot about the EXO-010.
I have a pair loading my 6P5GT drivers
on my SE 6L6G amp.
Hope you do get better.
Use some wisdom on activities.
Most men have problems with that.
Me included !
DanL
![]()
Post a Followup: