0.1Hz to 2MHz Amp circuit/bought equipment

Hi Wayne

If you just need a buffer without voltage gain to convert a hi-Z input to a low-Z output and you want to roll your own I'd look at 0033 and

0063 style buffers. Here is an app note about these:

Regards, Joerg

an

lm6152 is one op amp that springs to mind, 75mhz gain bandwidth product., should be easy to use, although might fall short of 100M ohm but theres plenty of others, to chose from, dont know what gain/input ofset you need tho.

Colin =^.^=

colin answered...

Wayne responded...

That's 400kHz, not 400Hz, and it's the slew-rate-limited maximum frequency for 5Vp-p output swing, not the gain frequency response.

You didn't specify your required gain or maximum rate of change. Slew rate is S = 2pi f A. For a full +/-10V output swing at 2MHz, that's a serious 125V/us. You'll also need a +/-15V supply opamp.

I like Burr-Brown OPA627's nice JFET input, but at 55V/us it's too slow. Their OPA637 does 135V/us, but you'd have to wire it in input-compensated mode for G = 1, which could compromise Zin.

The LM6321 and LM6325 were nice 800V/us G=1 buffers. The LM6313 works well, but is hard to get. I also liked the LM6362, R.I.P. Guess you'll have to consider Analog Device's bipolar AD817, with 350V/us, even tho it has 3.3uA of input current. Ouch! Other fast-slewing +/-15V opamps are LT1355, LT1357, LT1358 and EL2244. The LT1229 is an exceptionally-fast current-feedback opamp, rated up to 1000V/us and with a lower Ibias = 0.3uA typ for the + input.

Now, if you create an input attenuator to allow using +/-5V supply opamps, you'll need less slew rate, and you'll have many more ICs to choose from, like the elegant OPA655, a JFET rated at 290V/us. For this opamp you'd need to restrict your input signal to +/-2.5V, but then it would give you full-range swing capability up to 16MHz.

Thanks, - Win

(email: use hill_at_rowland-dot-org for now)

yes well spoted -)

i did notice it had rather low output curent, i shld of mentioned it but he didnt specify what gain or op load so i didnt consider that. i noticed the slew rate is 30v/us so i didnt condisder op swing a problem.

but somehow looks like this will struggle to give 1v op at 1mhz probably half this at 2volts strange ... ive used quite a few at 5volt and not noticed this, maybe ive not got anywhere near the 100pf load its speced at wich undoubtedly limits it but ive used it mainly at several hundred khz and the edges look sharp.

he specified 10volt/1uv so i asume the 10v is heavily atenuated like in a scope amplifier and this would have a bandwidth of 2mhz with a gain of 10 wich with 1uv input wld only be 10uv op so this may stil be suitable, but like i said theres 100s to chose from i chose this over others becuase of the low input bias current.

i just noticed the lmh6732 i think this wld be overkill tho as it has GBW of upto 1.5ghz and still with low input bias current but only 15Mohm, and

350mhz op at 2v. i bet it must be new or hard to get cos i could of done with one of these not long ago, dont know how available it is now though. youd beter be careful about layout though cos it cld easily oscilate at frequencies well over what you can see on most scopes, and wld just show up as very jitery behaviour and odd unexplained DC offsets.

or maybe the lmh6645 is better, i just looked for low input bias curent in this nice table from national ...

actualy i dont know why i didnt think of the opa655 as ive used this before too it would do well.

Colin =^.^=

This device tails off close to 400Hz at 5v. See image att. form datasheet (image sniped)

an

Z = 1/(2*pi*R*f). An input impedance of 100MEG, at 2MHz, corresponds to about .0008pF.

I don't believe you can construct a piece of instrumentation that meets that spec. (The leads of an IC have more capacitance between them than that, never mind what's inside the IC.) Or did you just want DC input resistance that high?

How about if you tell us what it is you're trying to use this preamp for? There's probably a better way to do what you want.

If you are looking for a complete solution and not a chip to build around, the Stanford Research preamp seems pretty close. See:

Response is +/- 0.5 dB to 1 MHz but they don't show any response curves so it's not clear if it has enough gain at 2 MHz for you.

-- Regards, Carl Ijames carl.ijames at verizon.net

As I pointed out, it wasn't the lack of gain above 400kHz that Wayne complained about the LM6152, intentionally or not, it was the LM6152's slew-rate limitation for 20Vp-p full-swing output.

Thanks, - Win

(email: use hill_at_rowland-dot-org for now)