Taming the ADA4817

Minus R2 in general. It increases the dropout voltage, but you want to minimize that unless you have a lot of ripple to contend with.

Cheers

Phil Hobbs

--
Dr Philip C D Hobbs 
Principal Consultant 
ElectroOptical Innovations LLC 
Optics, Electro-optics, Photonics, Analog Electronics 

160 North State Road #203 
Briarcliff Manor NY 10510 

hobbs at electrooptical dot net 
http://electrooptical.net
Reply to
Phil Hobbs
Loading thread data ...

But there's also a place for low performance cap multipliers, especially in HV supplies. You can cut down the line frequency or switch frequency ripple without a big HV cap and dangerous stored energy.

Reply to
whit3rd

ncy

I published a little paper on doing that back when I was a postdoc. It's a different case because you probably aren't running the amp off the HV.

Cheers

Phil Hobbs

Reply to
Phil Hobbs

When making a TIA with a high cutoff frequency, fc, high GBW bandwidth op-amps are indicated, especially to deal with high input capacitance, Cin, and high- value gain resistors. Applying f_T > fc^2 / f_RCin (see AoE III, page 539), can quickly push the f_T requirement to a high value, like 1GHz or higher, even for a low desired TIA bandwidth, 1 MHz, etc. But this only means you need a high loop gain in the region of the fc cutoff. It certainly does NOT mean you need full-fledged gain out to 1GHz!!

You can get fT above 1GHz, without 1GHz op-amps, using a composite configuration, see Figure 8.78.

Sometimes high-bandwidth op-amps are chosen to get high slew rates. But there are fast slewing op-amps that don't make you suffer from RF woes. E.g., parts using the CFB + VFB configuration, like the LM7171. You get over 4kV/us slewing, with a benign well-controlled 200MHz bandwidth.

Anytime you use an op-amp with fT above say 50MHz, and especially to 1GHz, you MUST treat it as an RF circuit at those frequencies. That means NOT using 5M, 1M or 0.1M feedback resistors! Yes, that can be the impedance of your network at low frequencies, but with parallel capacitors and other parts, you must make the feedback impedance sensibly low above 10 to 100MHz, so your GHz op-amp can control itself.

At high frequencies the entire PCB setup needs to be GHz ready, short wiring paths, immediately-located small 0603 bypass caps to a ground plane, low signal impedances, etc. And no cap load is allowed on the op-amp, sheesh!! Add a 50-ohm resistor if you want to connect to an output connector. Better, isolate the gain op-amp with a second lower-bandwidth part, 50-ohm back terminated, to drive cable connectors.

--
 Thanks, 
    - Win
Reply to
Winfield Hill

I first saw one in a car radio, around 1970.

--
Subject: Spelling Lesson 

The last four letters in American.........I Can 
The last four letters in Republican.......I Can 
The last four letters in Democrats.........Rats 

End of lesson.  Test to follow in November, 2016 

Remember, November is to be set aside as rodent extermination month.
Reply to
Michael A. Terrell

That's the year I got a Weller 8200 soldering gun as a birthday present and started ripping apart old TVs for parts. Eventually I learned how to build stuff that worked without using a book, but it took quite awhile, roughly from age 10 to 16.

Cheers

Phil Hobbs

--
Dr Philip C D Hobbs 
Principal Consultant 
ElectroOptical Innovations LLC 
Optics, Electro-optics, Photonics, Analog Electronics 

160 North State Road #203 
Briarcliff Manor NY 10510 

hobbs at electrooptical dot net 
http://electrooptical.net
Reply to
Phil Hobbs

I bought mine in 1965, when I was 14. I mowed lawns to buy my first real tools. I ordered it from Allied, Radio. I splurged and bought the kit, with the carrying case. I still have some of the tools, 50 years later. :)

Reply to
Michael A. Terrell

ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.