dac I/V convertion with very low distortion

[dirkwright]

Originally Posted by EssB

Hey Dirk,
I think Sergio is working up to TO-3, he's going for TO-126 (BD139/140) first ;-)

Good work Sergio!

It's only a matter of time my friend, only time. Eventually, everyone goes for a TO-3. It's the biggest hammer in the tool box.

[smms73]

Originally Posted by Shinja

hi smms
I considered CFP but dismissed.
the CFP itself is 2 stage feedback amplifier, then it has slew rate limit, moreover asymmetrical rise and fall.
I think the charactaristics are not suitable for our application.

I will investigate that. In simulation it gives good results even with 50 ohms input impedance.
One thing is for sure . the bias of the input transistors have to be between 60ma and 100, for increase linearity, we have to use some to126 devices with heatsink or paralleling some pairs of BC327/337.

Using feedback with this type of dac are very tenting

[regal]

Originally Posted by smms73

To increase flexibility and compatibility with more dac chips, I have included a reference voltage input , for use with dacs that need a offset voltage at the input, like ess9018 , ess9012. The new input is the one named "SET".

Have you tried this I/V with the TDA1541, I imagine the SET is where one would inject the 2mA offset it requires?

[staccatiss]

No problem with 60-100mA! Plenty of good devices to use. Size TO220 is also possible. For ex the audio drivers
2SC4973+complementary
2SC5171+complementary (very linear, fast)

/S

Originally Posted by smms73

I will investigate that. In simulation it gives good results even with 50 ohms input impedance.
One thing is for sure . the bias of the input transistors have to be between 60ma and 100, for increase linearity, we have to use some to126 devices with heatsink or paralleling some pairs of BC327/337.

Using feedback with this type of dac are very tenting

[jcx]

modern CFA op amp do much better than descrete Q at their appropriate power, V levels

pwr pad CFA op amp like TPA6120 do single digit W, >200 mA, and have ppm audio distortion specs - and work up to 100 MHz - in closed loop

why would anyone today lust after big Q in DAC I/V?

[smms73]

jcx. This circuit is not a cfa. It has no feedback loop, but it uses some techniques to make it very linear.

[zenelectro]

Originally Posted by smms73

I was not happy with the solution of paralleling 4 board for the es9012 / 9018. I have thought in using a different input for this type of dac , it has less than 500mohms of input impedance, it is biasing at 60ma . what you guys think ?

Q1 and Q2 will be bd139/bd140.


meanwhile i am waiting for the components to arrive to try my test boards.

Now you are getting a little lazy, using CFP connected transistors to handle
the high current swing / low OP Z of Sabre.

You can get -super- low distortion fron Sabre using very simple open loop
I-V , even without the small + feedback loops, but you must think very
laterally

ZZ

[zenelectro]

Originally Posted by Shinja

hi smms
I considered CFP but dismissed.
the CFP itself is 2 stage feedback amplifier, then it has slew rate limit, moreover asymmetrical rise and fall.
I think the charactaristics are not suitable for our application.

WRT CFP, I agree.

In fact I consider the small + feedback loops a violation of the open loop theme
- but that point could be argued forever, life is too short

[dirkwright]

Originally Posted by smms73

I will investigate that. In simulation it gives good results even with 50 ohms input impedance.
One thing is for sure . the bias of the input transistors have to be between 60ma and 100, for increase linearity, we have to use some to126 devices with heatsink or paralleling some pairs of BC327/337.

Using feedback with this type of dac are very tenting

Yes, oh yes, soon will come the day when you give up on those wimpy little transistors and embrace your true manhood with a pair of TO-3 monsters.

[zenelectro]

Originally Posted by dirkwright

Yes, oh yes, soon will come the day when you give up on those wimpy little transistors and embrace your true manhood with a pair of TO-3 monsters.

And the bigger the bjt the slower / more non linear it is.

There is an art to picking the right device for a particular place in this
circuit and running it at the right current.

Z