Mix sample: FET Engl + 5W FET PA + 4x12 cabinet
http://milas.spb.ru/~kmg/files/projects/Engl530/fet/reamp/5W4x12/530c1.MP3

In range > 100W the best choice on my opinion is LME49830+power mosfets.
http://www.national.com/ds/LM/LME49830.pdf
http://www.national.com/an/AN/AN-1850.pdf
http://www.national.com/an/AN/AN-1849.pdf

2 all, be careful.
All "official" Engl schematics is fake, interstage part drawn incomplete.
For examle E530:
"Official"
http://milas.spb.ru/~kmg/files/schematics/Engl/530/engl-e530-amplifier-schematic.pdf
Redrawn from real amp
http://milas.spb.ru/~kmg/files/schematics/Engl/530/1/530_full.gif

Problem of power scaling can be solved in transformer coupled SS PA with keeping all "tube" overdrive features.
It can be made simply by changing voltage at trahsformer center tap.

In this case, the point of beginning PA overdrive for 5/15/40W differs by only 1 digit at master volume

QuotePower transformers work very well in low frequencies (they are designed for 50/60Hz with over 95% efficiency, at full power), they couldn't care less above anything over 400 Hz ; it's a miracle that they have any highs at all.

In my FET PAs I used thin ribbon winded cores from power transformers (but with interleaving primary/secondary windings)
5W loss ~3dB@20kHz

15W loss ~4dB@20kHz

90W loss ~3dB@20kHz

QuoteWell, a 110+110V primary will probably work.

In this case you have twice reduced primary coil current limit, due to thinner primary coil wire.

QuoteYou maybe able to source 100volt line transformers as they have been used with success in Valve rigs and it is likely they could be made to work with KMG's concept.

110V transformer with center tap (2x55V) seems to be good for versions with 80V supply.
Just select one with suitable secondary voltage.

Quoteby the way i found an old monitor and found many things there transformer too. . . but the thing is i dont know how to check the transformers specs . . but thanks anyways!

If the transformer has a ferrite core, it is not suitable as the output transformer.

Quotewhere can i find transformer for micro PA? the 5watts PA? i really want to make one but i dont really understand how the thing works . .  about amps!

For micro PA you can try to use transformer from old transistor radio.
Also maybe this one will match for loading on cabinet.
75mW rating
http://eu.mouser.com/ProductDetail/Xicon/42TL008-RC/?qs=sGAEpiMZZMv0IfuNuy2LUcl0UojJLOKICZc25y8dmyo%3d
200mW rating
http://eu.mouser.com/ProductDetail/Xicon/42TM008-RC/?qs=sGAEpiMZZMv0IfuNuy2LUcl0UojJLOKIC2kkfS0HPnQ%3d

5W PA transformer
Core cross section 20x16 mm
Each half bobbin:
120 turns 0.25 mm (1/4) primary
75 turns 0.5 mm secondary (for 8 Ohm load)
120 turns 0.25 mm (1/4) primary

Which model do you mean?
For some models I use cores of standard transformers from this company
http://www.transled.ru/products/open/
5W - ТПК-15
15W - ТП-30
90W - ТПК-190
40W - old transformer core (don`t know type)
micro PA - TOT60/TOT57
Parameters of these transformers (TOT series) you can find there (unfortunately in russian)
http://milas.spb.ru/~kmg/files/literature/Malogabaritnye_transformatory_i_drosseli.(1985).zip

Another one sample of fet JCM800 & micro fet PA
http://milas.spb.ru/~kmg/files/projects/fetpa/micro/reamp/12A.WAV

The main problem with other PA versions is availability of transformer cores, suitable for sound applications.
5W version uses a standard core, which easy to order.

You will be laughing, but it is the sound of this baby:
http://milas.spb.ru/~kmg/microfetpa_en.html
I simply changed transformer to different one, to match 4x12 cabinet impedance.
Output power is about 70mW, but it really hear with the correct sound of full powered amp head.
It is very suitable for "night playing" when you do not want to disturb sleeping children.
Sample recording path:
E-MU Tracker Pre->Fet JCM800->Micro Fet PA->4x12 Cabinet->Sennheiser E906->E-MU Tracker Pre

QuoteAre you planning on selling these amps at all?  Or kits to build them?  At least a preamp maybe?

Only one item is now planned for production - 5W fet PA

I already received boards & transformers for it.

How about this sound?
http://milas.spb.ru/~kmg/files/projects/fetpa/micro/reamp/1.mp3

Yes, complex (voltage/current) feedback solves speaker dumping problem.
But DIYers rarely use this type of feedback.

I told about not overdriven PA.

About overdrive - when SS PA begins to overdrive, feedback path begins broken and amplification factor rises up to hundreds thousands. This lead to very sharp edges of clipping.
In tube PA amplification factor rises by about 3-7 dB (if feedback exists), so edges of clipping remains defined mostly by output tubes V/A transfer curve.

Width of flat tops of SS PA clipping increases proportionally input signal rising.
In tube PA, two things happen - in part increases the width of flat tops, and partly the signal "grows inside", forming an increasing step. This leads to signal compression.
In the case of overdrive, the intelligibility of the sound of tube PA is obtained higher than that of SS PA.
Conclusion - do not overdrive SS PA.

The old battle - SS against the Tubes.
Let's leave aside the preamp. It is different story.
Let's talk about not overdriven PA.
The  main difference is the negative feedback depth.
Due to deep feedback depth SS PA has low magnitude of long "tail" exponentially decreasing harmonics, also it has low output impedance.
Tube PA with feedback depth 3-7 dB has output impedance about 10-30 Ohms.
Also it has big enough magnitude of nearest harmonics, so its output spectrum occupies wider area.

These harmonics are perceived by the human ear is not as unpleasant overtones, but how the coloring the sound.
The high output impedance of the tube PA leads to producing higher sound pressure at speaker resonances.
So the tube PA sounds different and louder (at some frequences) against SS PA at the same power rating.