[arnaud]

Very interesting post from KG about the differences between his and stax designs. Sad that this will be swamped into this 1000+ pages thread, isn't there a way to sticky such info?
 
In regards to the 727 amp, Birgir, I just dig an older post of yours where you did not see any particular issue with lack of feedback in the 727 output stage ( post #31 ). I guess that was before you could listen to it? 

 

[Headdie]

Thank Spritzer for just connecting the extra speaker wires to the leads on my SRD-7. Keep it simple is always a good idea and I might do that. But this way, I'd loose the advantage of the banana plugs behind the SRD-7. It wouldn't be easy to plug/unplug the energizer and move it from one system to the other. I could probably fit banana females on the leads, but reusing the SRD speaker outs seems simpler to me... So my question remains : would it be ok to short the leads to the speaker outputs of the SRD-7 ?​

 
Quote:

Gentlemen,
 
I have an SRD-7. I'd like to take it twenty feet away from my power amplifier, just beside my computer and listening chair. To do that, I wonder if I could run speaker cables from my amplifier to the SRD-7 loud speaker outs (yes the SRD outputs)... And to turn them into inputs, I would also connect the SRD inputs to the SRD loud speaker outs... Hoping that you follow me... Now, I don't understand the electronics inside, but reasoning it as a blackbox, it seems to me that it should work... The front switch would just become an on/off switch... Could someone tell me if I'm gonna fry everything inside or get sweet music if I do it ?
 
Thank so much,

 

 

[spritzer]

Quote:

Very interesting post from KG about the differences between his and stax designs. Sad that this will be swamped into this 1000+ pages thread, isn't there a way to sticky such info?
 
In regards to the 727 amp, Birgir, I just dig an older post of yours where you did not see any particular issue with lack of feedback in the 727 output stage ( post #31 ). I guess that was before you could listen to it? 

This was just after the amp was released so I hadn't heard it.  In theory there is nothing wrong with the non-nfb except it needs to be run harder to make up for the lack of feedback.  I said as much back then. 
 
I was also really into tubes back then... how things change since to me they are just a gimmick these days and I take sand over them any day.

 

[maverickronin]

Quote:

I'm curious where the idea that the regular Stax amps are insufficient to drive the O2 originates from. Are there any objective measurements around to verify this?

SR-007mk2:

Impedance : 170k (at f = 10 KHz)
Capacitance : 94pF (including cable)
Sensitivity : 100dB / 100V r.m.s. at f = 1 KHz

SR-507:
 
Impedance : 145k (at f = 10 KHz)
Capacitance : 110pF (including cable)
Sensitivity : 100dB / 100V r.m.s. at f = 1 KHz

There doesn't seem to be a great deal of difference here. Sensitivity is identical, and capacitance is actually lower on the O2.

I'd be interested in this too since I'm not really up on how or if the requirements of a 'stat amp are different from an amp for normal dynamics besides just needing to swing a lot more voltage.
 
Has anyone ever measured different 'stat amps different real loads attached to see how they compare?

 

[tyre]

Quote:

I'd be interested in this too since I'm not really up on how or if the requirements of a 'stat amp are different from an amp for normal dynamics besides just needing to swing a lot more voltage.
 

 
http://www.tubecad.com/november99/page3.html

According to the article, the impedance of a capacitive load decreases as frequency rises, and subsequently, more output current is needed. As far as how much music actually contains 20kHz content, I couldn't say.

It looks like the author uses 500V peak-to-peak (around 176V rms) as the basis for his calculations. I don't know if that would be sufficient to reach 120dB peaks with a 100dB / 100V rms sensitivity. Someone more knowledgeable can probably do the math, I can't seem to grasp logarithmic units of measurement.
 
 
Quote:

Has anyone ever measured different 'stat amps different real loads attached to see how they compare?

 
That's exactly what I would be interested in seeing.

 

[maverickronin]

Quote:

http://www.tubecad.com/november99/page3.html

According to the article, the impedance of a capacitive load decreases as frequency rises, and subsequently, more output current is needed. As far as how much music actually contains 20kHz content, I couldn't say.

It looks like the author uses 500V peak-to-peak (around 176V rms) as the basis for his calculations. I don't know if that would be sufficient to reach 120dB peaks with a 100dB / 100V rms sensitivity. Someone more knowledgeable can probably do the math, I can't seem to grasp logarithmic units of measurement.

Music with stuff up to 20kHz isn't uncommon or anything but I am unsure as to how important it actually is.
 
If I did the math right 176VRMS should be about 105dB SPL for a 'phone with 100dB/100VRMS sensitivity.

 

[jcx]

average power bandwidth estimates for recorded music are only 3-5 kHz, doesn't mean you want to skimp on high frequency for transients, but slew limiting distortion isn't the most audible
 
popular vocal mics may have only ~15 kHz fc, 20-25 kHz isn't unusual for small condensers used in close micing, the Earthworks 50 kHz is pretty rare for a recording mic
 
 
the audiophile's swearing by vinyl aren't getting more than that ~ 5kHz power bandwidth in the recorded music - the needle jumps in the groove or cuts its own track if the tracking force were high enough

 

[spritzer]

This is a fun read even though it is all in Japanese:
 
http://www7b.biglobe.ne.jp/~konton/STAX%20Drive%20Amp.htm
 
Calculations of power needed etc.

 

[Amarphael]

So what does the future of high voltage transistors holds in store for our exclusive needs? any other sources than the all but gone CRT TVs?

 

[livewire]

For the stuff that is still in production, mainstream sellers such as Mouser, Newark, Allied, Digikey, etc.
 
For the "unobtanium", old stock resellers such as B&D, various audio electronics userboards, or roll the dice - there's always ebay...
Ya takes your chances, lot of fake stuff popping up in the market as of late.
NTE - ECG also markets many replacement transistors of similar design, one would need to compare specs before using. 
 
Thanks to KG and others, IXYS mosfets are some of the newer stuff that is making inroads to HV stat CCS and PSU designs.
 
FWIW, there are some new production low voltage signal input dual fets being made by LIS.
Only recently, this part's analogue was only available as old stock Hitachi sand.
Unfortunately, the complimentary fets are currently not being produced.
Too bad, as these are critical parts for the source input end of some amps, and are very hard to find.
 

 

[kevin gilmore]

Turns out that ROHM is making some high voltage surface mount transistors rated at 600 volts
and they sound really nice for the driver and translation stages. They are surface mount however
and i'll be making a whole bunch of surface mount to to126 converters to handle this problem.
They are used in the telephone industry, so i don't see them going away any time soon.
ixys does make output fets that run at 1200 volts, and they also sound nice but are a bit
hard to drive. The 2sc4686 non A version is still in production, and is good for 1000 volts.
There will always be ways to make stacker amps with slighly lower voltage parts.
You just have to get creative.
 
One of these days linear systems may actually stop lying to people and sell the p-channel
input fet parts.  Until that day, THAT corp makes a very nice fully complementary bipolar
part (like the old motorola part that krell used in the past)
 

 

[DaveBSC]

It's here!

 

 

 

[Maxvla]

So I've always been curious... on Stax amps that only have 1 output bias.. why do they always have 2 headphone sockets? Seems like a waste of money in most cases.

 

[spekkio]

It may be for one to A/B phones quickly and easily. Also, sharing music with a friend. If I don't recall wrongly the way the Stax systems are run, it doesn't hurt the output stage much to include multiple outputs.

 

[El_Doug]

Some of us have friends
 
Quote:

So I've always been curious... on Stax amps that only have 1 output bias.. why do they always have 2 headphone sockets? Seems like a waste of money in most cases.