recently I've been pondering parafeed stages, both for output and gain in amps. I've read lots of posts over at AA on the subject, and the great mathmatical derivation over at Valve, by Bill Ramsey, but that give me no idea about sonics. It's something I'd just try, but I have to pay for some big anode chokes for the output, and if it stinks (to my ears) then I'm out bucks that could have gone to something else.
Has anyone built a parafeed poweramp? Comments on measurements, sonics (esp if compared to a similar SET), topologies etc would be welcomed, whether it's R, L or CCS loaded.
My ideas for my amp are mentioned in this post from earlier today, and are inspired by the Outerlimit.
I see the topology as having two big advantages; extra isolation from the power supply,and the ability to optimise the OPT to run without a gap and no DC. The negatives are big anode chokes, ideally custom OPTs optimised for parafeed and a series cap with the OPT (yes I understand the PSU ac path).
I shouldn't have any trouble getting 20W with my proposed amp, but less would be fine.
FTR, I love IT PP amps, and have never gotten along wih SETs, and I don't find them to be 'magical'.
Hi Brett ,
Never built a choke loaded parafeed but I have built resistor loaded , CCS and 'SRPP' based amps . Hideously inefficient but who cares ! Currently using a 6C33 SRPP based output stage , very inefficient but it does sound good , plus it was very cheap to build as I used toroids as output transformers and the 6C33 were $10 each . What I can say about parafeed overall is to experiment with the output cap value and output transformer return (cathode or 0V) . Better off to parallel up output caps until you get the results you want , with some speakers the bass can be very 'farty' sounding with some values . The type and size of cap used has a lot of influence on the sound , oil caps in this position sounded awful , very sludgy . I've found that by tuning the cap to under the audio band gives the best results for me but may not be the case in your system
...sticks head up ... and ducks...
I thought the whole point of parafeed was to introduce enough variables into the SE design that it was finally possible to get it working properly :clown:
In other words - what are the advantages of parafeed?
Is it commercial - use a pp opt....lost by the choke load and capacitors (one for the opt connection, one for the psu...) cost ?
Is it sound - finally get proper bass out of an SE design ...presumably by the combined stored ac current flow in the OPT coils and parafeed capacitor:bigeyes: leading to psuedo proper push pull l.f. drive...
Or is it exploit the synergy of as many different phase angles of load impedance as posssible being present at once:rolleyes:
I think we should be told:devily:
Anyway, I'm also working on some ideas for parafeed - which I think ends up being fairly pointless as a topology, when the benefits are compared relative to straight transformer loads - EXCEPT when it comes to high current output stages that would saturate any core out there - like a 6AS7G/6080 stage, for example.
Maybe I should start a new thread on a 6080 parafeed??? Hmmm.... Can't seem to find any schematics.
...well have fun driving your 6080 and finding a tube with closely matched sections ! Better off using a 5998a instead of one of these grid-hungry things ! You also mention this
'which I think ends up being fairly pointless as a topology'
...well not that pointless if you can use 'unusable' transformers . I'm getting very good results from mains toroids , others have used 100v line transformers with success and small ungapped pushpull output trannies can also be used . <img src='http://www.diyaudio.com/forums/images/smilies/nod.gif'>
If you want to use that tube I can dig up a few ccts for you.
Actually the triode mismatch isn't half as bad as you seem to make out.
Even allowing for some bad luck when buying them, all you need to do is getting a bit of stock and do the matching yourself.
The 6080 has never been cheaper anyway.
Re: RE:6080 ?
A cross section of 40 NOS 6080 I tested a few months ago showed only two within 5% sectional matching for transconductance . About a dozen could just about manage 10% , the rest , appalling . Some testing with 30% mismatch (GE) . It seems that some testers can not accomodate the grid requirements of the 6080 , which when tested at lower anode voltages display better sectional matching . Further up is another matter , they literally fall apart . I dread to think what the curves would look like laid one on top of the other at the operating points used for audio designs . 6C19 is preferred in my view , these are a single triode and can easily be matched , 5998a again a double triode usually matched better and is less demanding to drive .
Do you want them per e-mail or are you starting a new thread?
I usually test them at an anode voltage of 150 V.
Naturally these valves were never designed as twin triodes and you're absolutely right, a lot of testers do have trouble with them.
Don't even think about putting a 6336A on those.
As far as the curves go: in a PP OTL they're extremely linear, almost all perfectly spaced straight lines.
If you need to match the 6C19 is a very good candidate indeed, I hear they sound very good too.
P.S.Brett, sorry for the OT.
Is srpp parafeed?
I guess so.
Scroll down on the above link and you all can see what I use and am still really happy with. BTW this amp has almost no tweek parts. No DC on the core and SRPP needs no big choke and no high parts count CCS. The 6111 can dump enough current to drive the 6080/6as7. I have a set of 5998 but have not tried them yet.
PS; this amp has super tight and big bass
|All times are GMT. The time now is 12:03 PM.|
vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2014 DragonByte Technologies Ltd.
Copyright ©1999-2014 diyAudio