Discrete low noise wideband transimpedance amp

[Brian Howie]

Yep. I used 3pf accross a current source. I am using a simple current
source for the diode, with 100Meg accross it. I have a bleed of 100Meg
to gnd to stop the node floating. I also have 1Meg from the base of the
cascodes to the fet gates, with 400k to the negative rail. There may be
some blemish somewhere causing the decrepancy.

It might be possible to remove all the dc blockers and use the fact that
there is about 0.4 vgs and make a vbe multiplier with the feedback
resistor without effecting the noise much.


A "good" design should be limited by the inherent noise of the devices
used, not the feedback/bias resisters.

Found it!! I stuck a 470ohm protection resistor in series with the
photodiode; the noise from this doesn't affect the performance at the lower
bandwidth. I get just below 1.5pA/rtHz at 100MHz now.

The ATF551 works better than the P35 device (
http://www.bookham.com/rf/pdf/packaged-fet/1110.pdf) I'm using.

Brian.

--
Brian Howie
BAE SYSTEMS Avionics Limited
Sensor Systems Division
Crewe Toll Phase II, 1st Floor,
Edinburgh EH5 2XS
Phone +44 (0)131 343 8769
FAX +44 (0)131 343 8941
Email [email protected]

 

[Kevin Aylward]

Found it!! I stuck a 470ohm protection resistor in series with the
photodiode; the noise from this doesn't affect the performance at the
lower bandwidth. I get just below 1.5pA/rtHz at 100MHz now.

I have now tried a single ended input version (around 7.5ma). It gets
1pa at 100Mhz, flat to 100Mhz with a 33K feedback resistor. 3db about
190Mhz.

Usually a dif pair is worst for noise at low frequencies. It sqrt(2)'s
the noise. At HF, limited by capacitance, it could be better as the
input capacitance could be half, which when feed by a *current*, will
double the signal even though the noise goes up by sqrt(2), hence may be
better. Its whether or not Cbe is dominating or not. It seems it insnt
for these fets.

Kevin Aylward
[email protected]
http://www.anasoft.co.uk
SuperSpice, a very affordable Mixed-Mode
Windows Simulator with Schematic Capture,
Waveform Display, FFT's and Filter Design.

"That which is mostly observed, is that which replicates the most"
http://www.anasoft.co.uk/replicators/index.html

"quotes with no meaning, are meaningless" - Kevin Aylward.

 

[John S. Dyson]

Just for interest, are there other similar HEMT parts from other
manufacturers ? Just to can select if Agilent part will be inaccessible
here in smaller ammounts ...

I know of no other components that have the incredible mix of features.
(I just happened on the Agilent parts a couple of years ago, during
a search for low-UHF components.) If I hadn't looked closely at the
spec sheets, I would have ignored the parts as being the typical
finiky SHF GaAs FET. (I immediately ordered a handful of 'em (54143)
for my jellybean supply, given my interest in low UHF.)

Future/Active might be the best supplier -- give them a try.. Also,
Avnet RF might also have some in stock.

Infineon? used to have a line of large geometry fets, but I don't think
that they had the cool enhancement mode JFET design.

John

 

[Winfield Hill]

John S. Dyson wrote...

I know of no other components that have the incredible mix of features.
(I just happened on the Agilent parts a couple of years ago, during
a search for low-UHF components.) If I hadn't looked closely at the
spec sheets, I would have ignored the parts as being the typical
finiky SHF GaAs FET. (I immediately ordered a handful of 'em (54143)
for my jellybean supply, given my interest in low UHF.)

Future/Active might be the best supplier -- give them a try.. Also,
Avnet RF might also have some in stock.

Newark has 'em in stock. Is Martin Devera aware that unlike ordinary
MOSFETs, the ATF54143 can have high input gate currents (200uA max on
the spec sheet, creating a large DC offset, and a high 8pA/rt-Hz max
current-noise density)? Didn't he mention 2.5pA/sqrt(Hz), saying he
hoped for lower noise? The '54143 may be well suited for amplifying
50-ohm source-impedance signals (0.5dB noise at 2GHz), but is it such
a good choice for high-Z photodiode amplification?

Thanks,
- Win

whill_at_picovolt-dot-com

 

[John S. Dyson]

John S. Dyson wrote...

Newark has 'em in stock. Is Martin Devera aware that unlike ordinary
MOSFETs, the ATF54143 can have high input gate currents (200uA max on
the spec sheet, creating a large DC offset, and a high 8pA/rt-Hz max
current-noise density)?

That would depend upon the leakage current source, right? Also, that
would argue for running with lower Vdg... The 55143 has about 1/2 of the
leakage (max) as the 54143, but I do agree that the high leakage current
(max) would be worrisome for certain applications.

The parts are still amazing (but are so different than normal that
they require careful use.)

John

 

[Martin Devera]

Newark has 'em in stock. Is Martin Devera aware that unlike ordinary

MOSFETs, the ATF54143 can have high input gate currents (200uA max on
the spec sheet, creating a large DC offset, and a high 8pA/rt-Hz max
current-noise density)? Didn't he mention 2.5pA/sqrt(Hz), saying he
hoped for lower noise? The '54143 may be well suited for amplifying
50-ohm source-impedance signals (0.5dB noise at 2GHz), but is it such
a good choice for high-Z photodiode amplification?

Hello Win,

thanks for pointing this out. It seems that Spice3f5 doesn't simulate
leakage and it is not in the model I have. Probably I could place
diode with Is=100uA between G-S ? It might account noise correctly then.
In any case I got 1pA/rtHz up to 10MHz and 12nA to 70MHz with plain BF861
so that I can say I'm satisfied.
I might test that P35 (45mS, 10nA leakage) if I will be able to obtain
a few pcs but I don't hope for too much better figures under 100Mhz.

By the way is there some theorem setting lower physical limit on obtainable
noise from such amps ?

Martin