• Please note that Audiokarma will be offline briefly for updates early on Monday morning, September 28th.

Improving the Fisher SA-100 with EFB II

I have one waiting in the wings, with a little different tact planned. This looks to have turned out fabulous.:thmbsp:
 
I know Dave has been super busy as I have been trying to keep up with some of his other threads.

But...what is the status of the article detailing the different EFB implementations?

And...I'd like to know where the SA-100 schematic is available so I can follow this thread along better. I looked online and here at the AK archives and cannot find a legible copy, thanks!
 
Hi Rust -- A great question, with the obvious answer that it's gotten sidetracked -- but still in the works. About the time I started on it, I also started getting barraged with questions and requests concerning the demise of Audio Regenesis, so I took some time to first deal with that concern. While doing so, it became clear that what was really needed was a universal EFB board, that could work in almost any application, with nothing but slight alterations to component values as required. That effort has yielded two basic designs, that can cover anything from cathode bias to fixed bias designs, employing triode, UL, or pentode output stages. The basic elements of these two designs have already been proven in my SA-100 clone, and in this thread where EFB is installed in a bonafied Fisher SA-100, but new elements had to be developed to make the circuits universal. Those have been done, and are now incorporated and being proven in the circuits used in the final EFB documentation effort currently underway, wherein EFB is being installed in a Fisher 400 receiver (documented in a separate thread). With the completion of that effort, the article will again get underway, but it will also benefit from the added experience the current effort will supply. The article will in fact tie all the versions together, explaining what version works best in which application, and address what elements of original design can best be improved on with its implementation. Thanks so much for the interest!

Dave
 
dcgillespie said:
Hi Rust -- A great question, with the obvious answer that it's gotten sidetracked -- but still in the works. About the time I started on it, I also started getting barraged with questions and requests concerning the demise of Audio Regenesis, so I took some time to first deal with that concern. While doing so, it became clear that what was really needed was a universal EFB board, that could work in almost any application, with nothing but slight alterations to component values as required. That effort has yielded two basic designs, that can cover anything from cathode bias to fixed bias designs, employing triode, UL, or pentode output stages. The basic elements of these two designs have already been proven in my SA-100 clone, and in this thread where EFB is installed in a bonafied Fisher SA-100, but new elements had to be developed to make the circuits universal. Those have been done, and are now incorporated and being proven in the circuits used in the final EFB documentation effort currently underway, wherein EFB is being installed in a Fisher 400 receiver (documented in a separate thread). With the completion of that effort, the article will again get underway, but it will also benefit from the added experience the current effort will supply. The article will in fact tie all the versions together, explaining what version works best in which application, and address what elements of original design can best be improved on with its implementation. Thanks so much for the interest!

Dave
Click to expand...

I think I'm gonna need 5-6 of those universal EFB boards, Dave. :thmbsp:

-D
 
LMAO!!!


Now back to your regularly scheduled Highly technical Electronic VOODOO! :music:
 
dcgillespie said:
... it became clear that what was really needed was a universal EFB board, that could work in almost any application, with nothing but slight alterations to component values as required. That effort has yielded two basic designs, that can cover anything from cathode bias to fixed bias designs, employing triode, UL, or pentode output stages. Dave
Click to expand...

Would this new design be applicable to the Dynaco ST-70 and Mark III amps? :scratch2:

-DB
 
Darth -- Absolutely, and it would be a very installation as well. Plenty of real estate (unlike the Fisher 400 I am current working on) and UL operation, meaning that only a bias supply regulator is required, since the screens are powered by the UL taps. With the typical sag found in the power supplies of these units, EFB(tm) will make a significant improvement in their performance.

Dave
 
Dave; I've decided to put an EFB-II in my Sansui 1000A provided the stock circuit is acceptable. The presence switch has got me a bit scared tho. When ON, it bypasses the RC (C110-R172 & C143-R168)network on the NFB loop (at least i presume that's the NFB loop) see pic of amp section below.

I've got it set up like my 800-C with the 10ohm resistors on pin5, 100ohm screen resistors, and a stock output coupling R-C of 200K resistors and .33uf caps. Running EH7591's. I modified the BIAS Circuit to the 4 pot version that Sansui later installed. I don't see any reason it can't run the EFB like an 800-C or a 500-C except for the presence switch which boosts the mid bass-low mids a little. Switch is called presence. Between it and Loudness, I have 4 different levels of boost at low levels.
Wondering if this is going to be a problem with the EFB-II.


attachment.php
 
Last edited:
Larry -- The presence control has no bearing on installing EFB at all.

The universal EFB circuit I am installing in the Fisher 400 thread will work perfectly in your Sansui. It has all the flexibility needed to work with the Sansui circuits, and would only require some slight component value adjustments to operate correctly with the Sansui operating conditions.

NJ -- The Fisher 400 thread is the proving ground for the universal EFB circuit. The basic elements of the circuit have already been proven in the previous EFB installation documentations, but some additional development was needed to make it a truly universal design.

It is a parallel operation going on while transfer of the AR boards is completed. The idea is that when the AR transfer is complete, all the necessary work in developing and proofing a universal board would also be complete, so that they (there are two) would then be next up in the que. That will put them towards the latter half of the year.

Dave
 
dcgillespie said:
It is a parallel operation going on while transfer of the AR boards is completed. The idea is that when the AR transfer is complete, all the necessary work in developing and proofing a universal board would also be complete, so that they (there are two) would then be next up in the que. That will put them towards the latter half of the year.

Dave
Click to expand...

That's financial year...right, Dave? ;)

-D
 
Dave,
I may try this mod to my SA-100. The mosfet you used is no longer made, but this new replacement should work: http://www.fairchildsemi.com/ds/FD/FDP8N50NZ.pdf

My main question is this: what do you think about using a 5R4GYB instead of a 5ar4 to drop more of the B+ voltage instead of cooking those dropping resistors right next to the PT?
To compensate for any possible sag, I am going to bypass c1a with a decent sized polyproplene cap (actually, I may bypass all the filtering and decoupling stages.

-B
 
Not -- I responded to your questions in your PM, but keep us posted here if you do the modification. Happy to help if you should need any!

Dave
 
Thank you Dave.
I am getting my Mouser order together now. I will post back.

For others, here is the response about the 5R4:

dcgillespie said:
As for choice of rectifier tube, the 5R4 will reduce B+ voltage levels, reduce power output accordingly, and eliminate the slow warm feature of the 5AR4 tube. It will also require re-biasing the amplifier due to the lower B+ levels produced. It's certainly your call, and the 5R4 will certainly work, but personally, I would stay with the original device.

If you are installing the EFB modification, you can actually reduce the value of the initial 125 ohm dropping resistor from the rectifier tube to 50 ohms. This is possible because the screen regulator portion of the EFB circuits actually provides significant filtering, resulting in an amplifier that is at least as quiet as the original design, even with the reduced value dropping resistor.
Click to expand...

Thank you again. My idea of using the 5R4 was to dissipate heat above the chassis instead of inside using the dropping resistors. I could ad some CL-90 thermistors before the rectifier to help with the slow warm up, but it seems with the all of the screen bleed resistors removed, the heat under the chassis will be much less anyway.

I will aslo be addressing the grounding issue.

My amp has been working fine for a good 5 years, but it needs to have the electrolytics replaced so I figured I would look into any improvements.
I would love to sneak a choke in there, Triad and Hammond have a 1 henry 240ma choke that could fit under the chassis without drilling, so maybe I will try that instead of a 300ma choke that is over 2" tall.
http://www.mouser.com/ds/2/177/EDB158SA-222276.pdf
or
http://www.mouser.com/ds/2/410/C-24X-222135.pdf

I have several of the Triad C-24x's on hand anyway. They are about the same small size.
 
Also, is the Damping circuit the .47u/5w resistors (R59 and R60) from the speaker output COM to ground, or the small circuit that runs from the 16ohm speaker tap to ground (C13,14 and R23) and the correlating circuit on the B channel?
 
The Damping circuit consists only of the .47 ohm resistors in each channel. C13 and R23 (and the corresponding components in Channel B) make up the negative feedback circuit, which you want to leave installed. C14 and its mate C21 are solely to prevent any RF from entering the amplifier in areas where RFI is a problem, via the speaker leads, which could act as an antenna to pick the the signals up. You can leave them in, or take them out at your own discretion.

Using about a 50 ohm choke would make for a wonderful improvement. If you go this route, you can reduce the existing 125 ohm dropping resistor to about 30 ohms @ 10 W -- just enough to help protect the rectifier tube during hot switching events. Add the choke after this resistor and first cap. After the choke, B+ is then sent to the output tube plates, and on to the rest of the circuit. This approach will require an addition filter cap of about 50 uF after the choke, since the original one at the tap for the output tube plates is now positioned before the choke.

The problem with the little SA-100 is available room, which as you said is limited at best. When I built my own clone version of this amplifier, I used a 300 ma choke to allow the choke to be effective even under full power conditions in both channels. If you are interested you can see it in the pics of this thread:

http://www.audiokarma.org/forums/showthread.php?t=466229

These are really wonderful amplifiers. I find I use my clone more than any other system in my rotation. The addition of EFB raises performance, and drops the under hood temps remarkably.

Dave
 
Doh! How I didn't recognize that as NFB is beyond me :slapshead
I thought the .47 resistor was probably what you referred to, because I saw the rest of the components in most of your pictures; but wanted to be doubly sure.

I have two nice panasonic/matsushita 100uf/500v cans, very small, that will glue inside near the bias circuit and a JJ 100/100 can on top - all care of our dear Mr. McShane.
I also have some of his smallish flattish GE polypro's that are 5uf and will fit for bypassing the decoupling stages and I think I can fit a decent sized Solen after the choke, or use a 47u F&T axial that I have. Both will fit I think.

I truly love the amp, but it runs so hot that some of the original wiring looks burned up and their is a penciled note on the bottom of the chassis that says "overheated wires replaced 1984"

It still sounds great as it is with Mullards in the front and either 6p14p-eb or sylvania 7189's, but I want to be cautious while giving it some bionics for it's next 50 years :)
 
You must log in or register to reply here.
Back
Top Bottom