New Stasis front end

[Zen Mod]

and that is asking for different OS pcb(s)

 

[6L6]

Discussing the cascode stage with Nelson, he said it was done because at the time there were no single devices that could handle the voltage and the current simultaneously.

As soon as it could be done in one transistor, it was.

 

[gary s]

So as you say Jim, with BJT's available these days, it would be madness to build a cascoded O/S.

 

[2 picoDumbs]

Forget cascode.
Unnecessary

 

[Zen Mod]

Pa didn't tell that he remembers time when toobz were made for 6V battery anode circ

that's the reason why common name for rail is Ub



( I knew that reason for cascode stage, believe info came from Pa, but wasn't even contemplating about cascode OS, having other occupations)

 

[mpmarino]

Cascode output stage? NP was obviously on Motorola's payroll in those days

 

[Nelson Pass]

Well, they didn't break.

The Stasis 1 had a triple series arrangement....

 

[gary s]

Obviously very conservatively rated - so will last a lifetime.

Mr Pass, that leads me to a question on the selected output devices, I see you have used the On Semi MJ21193/94 types with an fT of 4MHz and Cob of 500pF. I see Nakamichi for the PA5 (same O/S as the Stasis) used Sanken 2SC3263/2SA1294 with an fT of 60MHz and Cob of 250pF. Their voltages, currents, HFE and power ratings are similar.

Do you see any reason not to use the Sanken devices for the Stasis output stage- just asking as I have these here, thanks.

Gary..

 

[gary s]

Based on Zen Mods output stage and pcb layouts, one can shoot for the following output power's with the DC rails as shown, when biased roughly as per the original Threshold Stasis biasing instructions - or by setting the bias to achieve a maximum of 50 deg C at the hottest part on the OS heatsinks - give or take a few watts.

42 watts = +/-30VDC rails.
50 watts = +/-35VDC rails.
57 watts = +/-40VDC rails.
65 watts = +/-45VDC rails.
70 watts = +/-50VDC rails.
78 watts = +/-55VDC rails.
85 watts = +/-60VDC rails.
93 watts = +/-65VDC rails.
100 watts = +/-70VDC rails.

 

[6L6]

Power in W is V^2/R load

V are RMS.

50V rails will probably hit 42V maximum swing at high power because of the 1R emitter resistors.

42*.707 = 29.69

29.69^2 = 793

793/8 = 99 W

The same with 65V rails (57V swing, 40.3Vrms) gives an output of 200WPC into 8R...





Again, it is worth considering my previous admonition -

As rails go up and current increases, it becomes easier and easier to let the magic smoke out with excitement and vigor.

 

[gary s]

Thanks Jim for the input, I was basing my rough calc's on the published Stasis Biasing specs from some time ago. Was just a rough guide for members. I will be sticking to around 40VDC rail voltage - plenty of power for me.

For instance the SA/3 had +/-34VDC rails for a power rating of 50 watts with a bias voltage of 200mV hot across the emitter resistors.

 

[Zen Mod]

FE + OS pcbs package

OK, wasn't that much alleluia moody, but I know Pa have these days some other things to think of, so here it is ....... maybe some work shaven from his pile

(Pa - remember that scene from War Lover book - two old guys moving load of planks from vehicle to ground ...... and bunch of eager young guys did that for them in 10 mins.... )

anyway , to cut story short, here is first hand of FE pcb made ditto for OS pcbs posted few days ago

mounting of FE to OS with 2 spacers (holes 3.2mm) , all connecting pads aligned

resisted Babelfishing urge, added one cap and few base/gate resistors

shoot me

few minor important changes from Pa's schm (post #1) - no output signal return from OS to FE, thus protection diodes and Boucherot cell are placed on OS pcb; price is small - separate wire from PSU central GND point must be wired to OS pcb

no SMD parts on FE, even if I started with few ....

I'll optimize everything in next few days, but simple Yay (or Nay) nod could save my day

be well, stay safe, all

edit: oh yes - didn't resisted putting bottom footprint for my fave 2SK2145

 

[mpmarino]

You Didn't!

 

[gary s]

Well Zen Mod, you are doing a sterling job, thanks again for all the variations. Papa needs to pay you more money.

Now all we need is a sign off on all of this from the ultimate amp guru.

By the way ZM, what Idss grade do you use for your fave 2SK2145?

 

[Zen Mod]

Moi - BL

here, GR will do

already written in post #63, same applies for 2SK170

 

[gary s]

Thanks again ZM, you must have been up early this morning there. Nelson has replied to me that 6 to 8mA Idss will be fine here. I do have some tightly matched 2SK170BL in that range, although Nelson also said matching to 1mA will be fine.

I assume the 2 JFets in the SMD package are already matched to some degree by Toshiba?

 

[Zen Mod]

nope, I'm late

 

[6L6]

This is very cool.

I assume the 2 JFets in the SMD package are already matched to some degree by Toshiba?

Yes, of course. (they are on the same die)

 

[Zen Mod]

.......

I assume the 2 JFets in the SMD package are already matched to some degree by Toshiba?

good enough

and because I'm using them (two) in parallel, effectively as one JFet, matching of two in same case/die is effectively irrelevant

 

[gary s]

I agree Jim, this thread is very worthwhile and thanks again ZM. It will be interesting to see how many new Stasis builds are done - all thanks to Nelson - yet again.