diyAudio

diyAudio (http://www.diyaudio.com/forums/)
-   Solid State (http://www.diyaudio.com/forums/solid-state/)
-   -   Watt Sucking Fireball Series (http://www.diyaudio.com/forums/solid-state/154070-watt-sucking-fireball-series.html)

Roadbagger 28th October 2009 05:58 PM

Watt Sucking Fireball Series
 
2 Attachment(s)
Thank you for the very motivating event this year. For those of you who saw and heard the Watt Sucking Fireball No.4, and asked for a discussion of the circuits, here is the schematic for the power supply and one channel of the Amp. It is a simple application of J-Fets in a minimized gain chain of linear gain stages. An input stage that I call a differential-cascode-transimpedance amp is used to generate the initial amplification and convert the single ended input to a full differential drive that continues through to the P-P output stage. Being new to this forum, let me know if the Schematics didn't attach and help me out. More on this will follow if wanted.

Kirkwood

Roadbagger 29th October 2009 02:50 AM

Parts, Parts and more Parts
 
1 Attachment(s)
There are those things that a DIY will always need as I did for this project. Since this project was a High speed path from a clean piece of paper to the demonstrable show unit, Parts and Parts quick were needed. If any of you are in the Bay area, one of the great surplus parts sellers is in Milpitas. The EXCESS SOLUTIONS company www.excesssolutions.com Excess Solutions was essential to the successful completion of this amplifier, from toroidal transformer, speaker terminations, RCA jacks, heatsinks, resistors, caps, and wire all were from the Excess guys. They really have it all and my thanks to Mike, Todd and AJ for running the ultimate house of parts and for just being there or this project would have taken much longer.

Roadbagger 29th October 2009 06:16 AM

process pieces
 
3 Attachment(s)
some mess before the event

carpenter 30th October 2009 02:11 AM

I liked the sound of your amp--smooth! I'm very interested in your topology and, if you have the time, would love to read a detailed explanation regarding its operation.

Thanks for being at the 09 BAF.

John Inlow

slam 30th October 2009 08:59 AM

Quote:

Originally Posted by Roadbagger (Post 1964070)
some mess before the event

Thanks for sharing the design! Looks like you have some really nice test gears. That Tektronix 570 curve tracer certainly looks great. And the ther box is some kind of semiconductor noise analyzer?

jackinnj 30th October 2009 01:53 PM

I have never seen a Quan-Tech in real life.

jacco vermeulen 30th October 2009 03:38 PM

Mr Rough,

do you mind telling what type of JFEts are used for the driver section ?

A version with LU1014D for the Sony power devices would be fun.

jan.didden 30th October 2009 03:45 PM

Quote:

Originally Posted by Roadbagger (Post 1963496)
Thank you for the very motivating event this year. For those of you who saw and heard the Watt Sucking Fireball No.4, and asked for a discussion of the circuits, here is the schematic for the power supply and one channel of the Amp. It is a simple application of J-Fets in a minimized gain chain of linear gain stages. An input stage that I call a differential-cascode-transimpedance amp is used to generate the initial amplification and convert the single ended input to a full differential drive that continues through to the P-P output stage. Being new to this forum, let me know if the Schematics didn't attach and help me out. More on this will follow if wanted.

Kirkwood

Nice trick to get that -45VDC. Not many here would remember how to do that, I suspect. It does date you ;) . And me.
Anyway, enjoyed our chat at BA.

Jan Didden
Linear Audio

Roadbagger 31st October 2009 08:05 AM

Well..........
 
The Quan-Tech was used extensively to match the gain, IDss/Vgs and noise characteristics of Linear Sys parts used in the BAF demo amp. Some years ago I worked with Mr. Stansbury to refine a vacuum tube adaptor for the 5173 to also measure equivalent input noise of small signal triodes. The WattSuckingFireball4 can be operated without feedback and have only the gain shift over temperature. Distortion and IM are about the same over the audio passband of 20-20K. Linearity was a combination of carefully matched parts, including resistors and caps, but primarily established by the balanced signal path Gm and successive reverse impedance isolation from stage to stage by strict differential device matching, optimized bias centering, and massive differences in source to load impedance.
The Tek 570 is my best tool for matching up transfer functions of various triodes used in tube preamps designed here. Another tool that has provided a great amount of data is the General Radio 561-D Vacuum Tube bridge designed by Tuttle which produces remarkably accurate Gm, Rp, and mu numbers for all the operating bias points of a thermionic triode or J-Fet at various audio frequencies.

K-wood

Roadbagger 2nd November 2009 06:36 AM

WSF continues
 
Life on the Bench

There are 3 fundamental principles in this amp design. Linearity, elimination of DC gain terms, and stage gain consistency. A review of all gain elements from the first Thermionic Triode to a FET (conceived before bipolars by Shockley) to bipolar transistors, mosfets, depletion mode or enhancement, IGBT's etc., show a progression of greater gain and greater nonlinearity over time . The nonlinearities are many and varied from square law to exponential with complex biasing variations from current crowding to beta shifting over collector current and reverse feedback differences with gain/impedance changes over the operating range of use. A careful review of any gain element's transfer function will reveal operating regions to use and to avoid. Personally I prefer low mu triodes as the Avatar shows. The WSF input stage is at first, a simple differential pair of J-Fets, but only at AC voltages above the passband of a coupling R-C network between the sources. This eliminates DC offset terms and drift of a normal differential amp while allowing independent manipulation of biasing conditions. This comes in handy later in the output circuit. Differential J-Fet pair sources fed from a resistance to some minus supply will not perform well in either common mode rejection or single ended to differential conversion because there will always be a difference in source current produced from the driven and undriven input gate voltage. A dual tracking current source was used here employing a low noise LSK170 as the (I) source and a pair of low noise dual NPN’s configured as a full Wilson Mirror. Keeping the design simple, there are no cross coupled things going on but the need to keep miller effects negligible dictates a cascode loading of the Diff input pair. By stacking a pair of fets as a GM amp and feeding back a 10K ohm resistor to the GM amp gate, we can produce an amp stage with about 46 db of open loop gain and feed the Diff amp drains into a virtual ground node at the input connection of a 10K GM feedback resistor. The 1 ma drain current biases up 10 volts across the 10K and swings it's output side to the change in diff current times the 10K resistor while holding the Diff stage drains to about 1/200th of the output voltage swing. Now we have a single ended to differential non inverting voltage gain and a bandwidth of about 1 MHz. Driver output Z is about 50 ohms…..More through the looking glass later.

Kirkwood


All times are GMT. The time now is 01:03 AM.


vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2014 DragonByte Technologies Ltd.
Copyright 1999-2014 diyAudio


Content Relevant URLs by vBSEO 3.3.2