First I want to say thanks for answering all my questions. My original idea was to replace a tube gain stage with a jfet gain stage. Seeing as I don't exactly know what I am talking about, per se, I am open to hearing other related ideas.
As to high input capacitance, looking at the SJEP170R550 (there are other options), it looks like 170pf. Is that that considered high? http://www.semisouth.com/products/uploads/DS_SJEP170R550_rev1.3.pdf
As to high input capacitance, looking at the SJEP170R550 (there are other options), it looks like 170pf. Is that that considered high? http://www.semisouth.com/products/uploads/DS_SJEP170R550_rev1.3.pdf
It is very low for a device with transconductance 1,500 miliamperes/volt and 60+W of power dissipation. If I remember right 300B has transconductance about 300 times less, and it's max power dissipation is 4 times less. I would ask, why do you need 300B at all if you already have SJEP170R550, but I am afraid Sheldon will yell at me again for not answering straight and directly. 😉
When I asked Gary Pimm that very question he said no problem. After all it's just a 3 watt SS push pull amp needed to drive a 300 B. Gary said his all SS PP Tabor, probably two stages, should do the job easily and that a SS CCS could be used as the input splitter.
Then I asked Rene' Jager the same thing and he said "lets do it". So, I am making and accumulating components. Level three interstage's, using nickle core, for PP to SE 1+1 to 1 and large, level 3 OPT's for the 300 B's. I only need 7 watts to drive a Lowther and an Eminenece Delta 8 drivers.
Fortunately for me I don't have to know anything except how to design and build the transformers. Rene' says we will use lateral high voltage FET's, whatever that is. I want the PP SS driver section to dump the noise and distortion that I have found in all of the other 300 B SE amps I have auditioned, not that this includes any vast number of them. I want the 300 B for the musical beauty it infuses Lowther's with, especially EnABL'd Lowther's.
Bud
170 pF is very low for a power FET. However, other than VDD = 300 V, it does not specify the conditions under which this measurement was performed. There's also no curve that shows the input cap vs gate-drain or gate-source voltage. So it's hard to guess.
It doesn't look all that linear, though. Especially not compared with an average triode.
~Tom
It's not bad. As the input device, you can drive it fine with any typical source below around 10k output impedance (For comparison, an 12AX7 stage with a gain of about 70 would have an input capacitance of around 120pF). But you'd want to keep the Drain/Source voltage above about 100V. If you look at the curve, you'll see that as the drain to source voltage drops below about 100, the input capacitance rises sharply. So your HF roll off starts to drop, but probably more important, it varies with amplitude. This creates a particular type of distortion.
Sheldon
I WASN"T YELLING! But yeah, we agree that it makes more sense as an output device.
Sheldon
Last edited:
Hi Sheldon, you compare apples and oranges: input capacitance of 12ax7 stage VS input capacitance of Semisouth JFET. The stage that uses this JFET would have input capacitance equal to 17 pF multiplied by this stage's voltage gain. It may be very high if used with common source. But if to use with common drain (source follower) it may be no more than 40 pF.
Last edited:
I refer you to the "Schadeode" amp I tested which uses a MOSFET as the first stage voltage-to-current converter which works into the local feedback from the output tube plate.
The input capacitance is partially bootstrapped by the unbypassed resistor in the source circuit. This resistance is needed to set the slope of the V/I converter into the feedback current swing and thus set the input sensitivity of the amp.
In practice, the Cgs is reduced by a factor of 10 or more, to reasonable values given most FET devices with ~500 pF Cgs resulting in ~50 pF input capacitance.
To drive a 300B, local feedback is not needed but big voltage swing is. I would treat the FET with it's source resistance as a pentode and work into a defined load resistance of the drain resistor in parallel with the grid resistor of the 300B. The source resistance value would be chosen for a particular current swing into the load resistance, again setting input sensitivity.
For A2 operation another FET could be used as a follower as in Power Drive or the Meteor shown above.
Significant source resistance linearizes the V/I characteristic to nearly a straight line for reasonable Gfs values of the FET. Note however that Gfs of any FET at low current will be many times less than the data sheet value.
The particular type of FET used may need different biasing. The IXTP01N100 is a depletion mode device that will self-bias with a source resistor.
Attachments
Last edited:
Let's compare apples to apples then. If we want to use the JFET as a replacement for the 12AX7, then what is source resistor, drain resistor (or combined with the following grid leak) to achieve the same gain? And what then is the input capacitance? This is not a rhetorical question, I'm interested in how to compute the answer. As you know, you are well beyond my competence levels.
Sheldon
Unity gain SICJFET source follower drives a really big Triode,
or small Triode source follower drives SICJFET and provides plate feedback for real Mu?
Either way makes sense, whats to fear?
or small Triode source follower drives SICJFET and provides plate feedback for real Mu?
Either way makes sense, whats to fear?
I'm curious to see how this turns out for you.
Now I picked the semisouth jfet because I knew it would work. Is there a smaller, less expensive jfet that might work?
Anyone have the parts needed to test out a tube driven by a jfet stage? Want to let us know how it sounds?
I have thought about trying this design idea out for my first venture into a completely DIY amp after I have learned a bit more. The problem for me is that it would be very expensive to build an amp using a 300b, those output transformers are an arm and a leg and a torso. Is there a more affordable tube I should try this idea with first? Is there anyone who has the parts who would like to try this out?
I don't think anyone has mentioned using an interstage xfmr to go from a highish current Fet input to a high voltage drive signal for the output tube. Maybe with a follower on it for the 300B grid. For $13 you could try experimenting with say an Edcor XSM 600/15K. ( 5 to 1 voltage gain)
EDCOR Electronics Corporation. XSM600/15K
EDCOR Electronics Corporation. XSM600/15K
How much current is to much for a tube? I feel like I only hear about voltage swing when it comes to tubes.
I'm just reviving this thread to see if there are more recent devices that can drive a 300b. At the moment I use a 4P1L but open to solid state ideas. Are these Exicon devices any good for instance?
Exicon Lateral MOSFETS - Our Product Range
Exicon Lateral MOSFETS - Our Product Range
Poor SemiSouth. Expanded, expanded some more, was awarded, was highly published, lionized, and finally had a quixotic and sudden closure. No apparent reason other than over-extension.
Financial hubris, not met with concomitant sales.
I, having been an owner-partner of a similar sized firm, can tell you: once “the staff” starts pulling in high-powered and highly rewarded sales executives, managerial and operations execs, and an endless queue of highly highly motivated Big Equipment suppliers … it is easy to start bending the rules, to start counting chicken farms that haven't been built, yet alone hatched any patented chickens.
To count sales-sort-of-guarantees(-and-assurances) by the sales execs, who, lusting for Big Commissions, convince “the house” to front them part of those commissions on relatively modest “commitment intent” sales. You know, slow markets, unexpected client setbacks, nothing that'll go for a couple of months...
That kind of leveraging is exactly what caused the Housing Bubble, among other nation-wide financial crises.
It is, on a microeconomic field, what causes small companies such as SemiSouth to blossom and go kaput soon after.
No roots to support the huge blooms.
Sapsuckers infesting the firm.
Blowflies, well-dressed liars, cheats, con-men.
Don't leave out the con-women.
Fewer for sure, but somewhat harder to detect.
Had 'em all.
Sold the firm to a bunch of Florida con-men, on excellent faith and regard.
Was sued BY them for their own poor new-management decisions.
Lost millions, personally.
Millions, like $15+ mil
Now I comment in my thread-bare “retirement” here at DIYaudio.
Poor SemiSouth.
I really feel for them.
⋅-⋅-⋅ Just saying, ⋅-⋅-⋅
⋅-=≡ GoatGuy ✓ ≡=-⋅
Financial hubris, not met with concomitant sales.
I, having been an owner-partner of a similar sized firm, can tell you: once “the staff” starts pulling in high-powered and highly rewarded sales executives, managerial and operations execs, and an endless queue of highly highly motivated Big Equipment suppliers … it is easy to start bending the rules, to start counting chicken farms that haven't been built, yet alone hatched any patented chickens.
To count sales-sort-of-guarantees(-and-assurances) by the sales execs, who, lusting for Big Commissions, convince “the house” to front them part of those commissions on relatively modest “commitment intent” sales. You know, slow markets, unexpected client setbacks, nothing that'll go for a couple of months...
That kind of leveraging is exactly what caused the Housing Bubble, among other nation-wide financial crises.
It is, on a microeconomic field, what causes small companies such as SemiSouth to blossom and go kaput soon after.
No roots to support the huge blooms.
Sapsuckers infesting the firm.
Blowflies, well-dressed liars, cheats, con-men.
Don't leave out the con-women.
Fewer for sure, but somewhat harder to detect.
Had 'em all.
Sold the firm to a bunch of Florida con-men, on excellent faith and regard.
Was sued BY them for their own poor new-management decisions.
Lost millions, personally.
Millions, like $15+ mil
Now I comment in my thread-bare “retirement” here at DIYaudio.
Poor SemiSouth.
I really feel for them.
⋅-⋅-⋅ Just saying, ⋅-⋅-⋅
⋅-=≡ GoatGuy ✓ ≡=-⋅
Last edited:
Sad personal story. I was in the IT business and sold our small company a few years ago without nasty problems or legal issues. I like to read your contributions though, a mixture of opinions, experience and technical knowledge. I keep learning every day, thanks to guys like you.
Regards, Gerrit
Regards, Gerrit