1. Home
  2. Electronics Design
  3. JFET amp noise
  4. Page 2

JFET amp noise

Hi Helmut,

Helmut Sennewald wrote:

Just a typical op-amp circuit:

formatting link

Make R4 open, and then make R3, R1, and R2 1 uOhm and you get fV/Hz1/2 noise. There seems to be no lower limit. Then make both R3 & R4 open while leaving R1 & R2 shorted and you get the opposite-- to much noise.

At 780 Hz my LTspice gave a gain of about 6000 for the NTE937M, but I'm sure open loop gains can greatly vary between parts. I just wanted to see a ballpark figure, so I didn't take the time to measure my NTE937M exact gain, but it seems the noise is definitely in the low two digit nV range. I was just surprised to see such high noise from an open loop LT1028N.

Now that I think about it, I guess an air gap would generate noise, even a vacuum gap. If you have potential then the electrons will accelerate in the vacuum, so the energy has to go somewhere. Perhaps an ideal noiseless resister is one that directs the energy somewhere besides current. I know radiation resistance is noise free. Here's another thought. Take a circuit with a DC voltage source and two resisters. Lets say there's 1 volt across resister B. That resister is generating noise. Now replace that resister with a 1 volt battery. Essentially at DC it's the equivalent circuit except less noise. :) I believe magnetic viscosity is another example of a noiseless resistance.

BTW, I'm now trying to measure power mosfet noise at 20+ pulsed amps. Am I wasting my time trying to get an accurate figure with LTspice? I tried adding KF and AF to the IRF530, but it doesn't seem to change the noise in any way. Also for some reason the input voltage to the mosfet doesn't seem to change the noise. Increasing the mosfet's input voltage equates to more output current. In real life I'd expect more current to equal more noise.

Thanks, Paul

Reply to
pmlonline
Loading thread data ...

schrieb im Newsbeitrag news: snipped-for-privacy@c74g2000cwc.googlegroups.com...

Hello Paul,

this test circuit with open R3 and R4 can never work. Even an offset voltage of 1uV would keep the opamp close to one of the power supply rails.

Do you seriously believe that? I don't! More likely you can put 10 camels through the eye of a needle than this theory above is true.

Best regards, Helmut

Reply to
Helmut Sennewald

Are you using the "typical" input offset voltage Helmut? What if you use the minimum? Typical offset for NTE937M is 3mV. So I'd bet my offset is about 1mV or less.

The circuit used to have the offset resisters and adjustable pot, but I had to remove those because they caused a lot of noise. I know that some pots can generate a ton of noise. When the offset parts were removed most of the noise went away.

You should note that the noise is random, which caused the signal to reach both positive and negative saturations. I could watch the signal drifting all over the place over time. Some times the signal was in saturation and sometimes it was not saturated. That provided plenty of time to study the signals typical swing per second. :-)

Actually I do. :) Consider the radiation resistance of a yagi antenna. It's not uncommon to have 1"+ diameter poles, so electrical resistance is insignificant, but the radiation resistance is typically designed for 50 ohms, but can be extremely high ... in the thousands of ohms. This is resistance, not reactance. Where would such 50 ohms noise come from Helmut? There is no radiation resistance noise. :-)

Think about it. If there's no di/dt then there's no field. If there's no field then there can't be any such noise coming from the field. When there is di/dt there's still no radiation noise, lol. Where are you proposing this magical radiation noise is coming from Helmut? Perhaps on a quantum level there might be some yocto zero point energy noise, lol.

Paul

Reply to
pmlonline

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.