Gain peaking in driven shield
in [Design]
|
From: George Herold on 7 May 2010 15:45 On May 6, 9:38 pm, George Herold <ggher...(a)gmail.com> wrote: > On May 6, 9:54 am, MooseFET <kensm...(a)rahul.net> wrote: > > > On May 5, 9:26 am, George Herold <ggher...(a)gmail.com> wrote: > > > It really matters that the gain be 1.00 and the phase be 0.00 at the > > frequency you need to measure. In my case, I have the option of > > changing > > the circuit between measurement bands. The circuit had a roll off at > > about F*100 and F/100 > > The phase shift at high frequencies is certainly troublesome... > Playing around with LTspice tonight it seemed that the peaking > happened when I made the Cig capacitance too small. (Cig is only > shown in Bill Sloman's correction to my circuit diagram.) > > Perhaps if the gain is less than 1.00 the gain peaking will be > smaller? I'm being too greedy by driving the shield too hard. > > George H. > Hi all, I had a bit of time this afternoon and I tried this, > +-----Cis--+ > | | > | |\ | > +-----+--+ \ | > | | >----+--+P +----+----+ > | +-- / OPA | O <-+ | | > Rmeas. | |/ 134 | T R1 Csg > Rmeas. | | | R1 | > | +---------+ GND | | > | GND GND > GND Where R1 was 100 ohms and the Potentiometer was also 100 ohms. I looked at the step response and was able to 'tune out' the over shoot by pulling the pot wiper off the top point. Thanks for all your help and interest, "mischief managed" George H.
From: MooseFET on 7 May 2010 20:52 On May 6, 6:38 pm, George Herold <ggher...(a)gmail.com> wrote: > On May 6, 9:54 am, MooseFET <kensm...(a)rahul.net> wrote: > > > > > On May 5, 9:26 am, George Herold <ggher...(a)gmail.com> wrote: > > [....] > > > > +-----Cis--+ > > > | | > > > | |\ | > > > +-----+--+ \ | > > > | | >----+---+----+ > > > | +-- / OPA | | | > > > Rmeas. | |/ 134 | R1 Csg > > > Rmeas. | | R1 | > > > | +---------+ | | > > > | GND GND > > > GND > > > If the cable is very long, the simple capacitance model of the > > cable fails. You need to start dealing with the inductance > > too. > > Yeah, I've only got about 0.5m of cable at a few MHz max. Once the > cable becomes a tranmission line you can't 'do' the driven shield > trick. Can you? > > > > > > > I have used the LSK170 to make a very low noise non-inverting gain > > to make this work a little better. The noise put onto the cable > > by your shield drive starts to win at some high frequency. > > > --------------- > > ! ! > > [R] U1 [V] > > ! !\ ! > > !---+--------! >----+--+---- > > in ---->! !/ ! > > !---+--[R]------------- > > ! R2 > > [R] > > ! R1 > > > IIRC R1=300, R2=30 U1=gain of 100 > > Wow, not sure I really get that circuit. You've got the Jfet as a > 'follower' and I assume I should hang the shield on the bottom of > R1... (Correct me when I'm wrong) along with some resistance (?) to > sink the current. The amp U1 is feeding the bottom of a voltage > source and R which looks like a current source. But I'm not sure what > R2 is doing? I forgot to note that the U1 gain is -100 not +100 Does that make more sense? The drain of the FET goes down, the output of U1 goes up, the source of FET goes up making it negative feedback. [...] > The phase shift at high frequencies is certainly troublesome... > Playing around with LTspice tonight it seemed that the peaking > happened when I made the Cig capacitance too small. (Cig is only > shown in Bill Sloman's correction to my circuit diagram.) When you get up in frequency, the phase of what you feedback starts to matter hugely. Just a little error makes for a lot of reduction of the impedance of the thing.
From: Nemo on 9 May 2010 07:11 George Herold writes >There are three layers. On the outside is ground, next is >the shield, the center of the coax is the signal line from the >resistor. Driving the middle shield, reduces the capacitance between >the inner signal line and outer ground. However it adds the >capacitance from the shield to the inner signal line. I'll look up >the AD8626. And try adding a bit of R in the output line. Do you >have a TI app note number? What sort of source impedance are you >looking at? 100k looks OK, but 10k ohms shows gain peaking. I am looking at a biased photodiode which is, I suppose, very high impedance indeed. It strikes me that my topology is fundamentaly different to yours - I am using two different coaxes, and driving the shields of both with images of the signals on their cores (which is why I use a dual op amp). That's why I assumed there is no capacitance from inner core to outer shield, because I have two independent shields. I then wrap a metal mesh round the pair of them to act as an RFI screen, I'll connect that to 0V. This is all try-it-and-see stuff, I can't say it works yet! Hope to improve leakage (mainly), noise (hopefully) and see if I can push bandwidth up to 1MHz. The points about phase inversion etc is very interesting. Eeek, hadn't thought of that! This is going to be fun 8) -- Nemo
From: George Herold on 9 May 2010 16:37 On May 7, 8:52 pm, MooseFET <kensm...(a)rahul.net> wrote: > On May 6, 6:38 pm, George Herold <ggher...(a)gmail.com> wrote: > > > > > > > On May 6, 9:54 am, MooseFET <kensm...(a)rahul.net> wrote: > > > > On May 5, 9:26 am, George Herold <ggher...(a)gmail.com> wrote: > > > [....] > > > > > +-----Cis--+ > > > > | | > > > > | |\ | > > > > +-----+--+ \ | > > > > | | >----+---+----+ > > > > | +-- / OPA | | | > > > > Rmeas. | |/ 134 | R1 Csg > > > > Rmeas. | | R1 | > > > > | +---------+ | | > > > > | GND GND > > > > GND > > > > If the cable is very long, the simple capacitance model of the > > > cable fails. You need to start dealing with the inductance > > > too. > > > Yeah, I've only got about 0.5m of cable at a few MHz max. Once the > > cable becomes a tranmission line you can't 'do' the driven shield > > trick. Can you? > > > > I have used the LSK170 to make a very low noise non-inverting gain > > > to make this work a little better. The noise put onto the cable > > > by your shield drive starts to win at some high frequency. > > > > --------------- > > > ! ! > > > [R] U1 [V] > > > ! !\ ! > > > !---+--------! >----+--+---- > > > in ---->! !/ ! > > > !---+--[R]------------- > > > ! R2 > > > [R] > > > ! R1 > > > > IIRC R1=300, R2=30 U1=gain of 100 > > > Wow, not sure I really get that circuit. You've got the Jfet as a > > 'follower' and I assume I should hang the shield on the bottom of > > R1... (Correct me when I'm wrong) along with some resistance (?) to > > sink the current. The amp U1 is feeding the bottom of a voltage > > source and R which looks like a current source. But I'm not sure what > > R2 is doing? > > I forgot to note that the U1 gain is -100 not +100 > Does that make more sense? Hmm, I'm not quite sure. I put thin in my notebook and if I ever need it I hope I can find it. I re-drew your circuit to look like this, --------------- ! ! [R] |\ V] +------- \ ! !---+ GND+ >----+--+---- in ---->! | / ! !---+--[R]------------- ! R2 [R] ! R1 (Sorry a bit hard to squeeze the ground in there.) But I'm not sure that's what you intended. > The drain of the FET goes down, the output of U1 goes up, > the source of FET goes up making it negative feedback. > > [...] > > > The phase shift at high frequencies is certainly troublesome... > > Playing around with LTspice tonight it seemed that the peaking > > happened when I made the Cig capacitance too small. (Cig is only > > shown in Bill Sloman's correction to my circuit diagram.) > > When you get up in frequency, the phase of what you feedback starts > to matter hugely. Just a little error makes for a lot of reduction > of the impedance of the thing.- Hide quoted text - > Yeah that's what I found, If I tried to 'tune out' too much capacitance I got gain peaking at high frequency. My solution is a pot on the output to adjust the 'feedback'. I still have to see how it works on a noise signal. George H.
From: George Herold on 9 May 2010 16:55
On May 9, 7:11 am, Nemo <z...(a)nospam.nospam.nospam.nospam.co.uk> wrote: > George Herold writes > > >There are three layers. On the outside is ground, next is > >the shield, the center of the coax is the signal line from the > >resistor. Driving the middle shield, reduces the capacitance between > >the inner signal line and outer ground. However it adds the > >capacitance from the shield to the inner signal line. I'll look up > >the AD8626. And try adding a bit of R in the output line. Do you > >have a TI app note number? What sort of source impedance are you > >looking at? 100k looks OK, but 10k ohms shows gain peaking. > > I am looking at a biased photodiode which is, I suppose, very high > impedance indeed. It strikes me that my topology is fundamentaly > different to yours - I am using two different coaxes, and driving the > shields of both with images of the signals on their cores (which is why > I use a dual op amp). That's why I assumed there is no capacitance from > inner core to outer shield, because I have two independent shields. I > then wrap a metal mesh round the pair of them to act as an RFI screen, > I'll connect that to 0V. This is all try-it-and-see stuff, I can't say > it works yet! Hope to improve leakage (mainly), noise (hopefully) and > see if I can push bandwidth up to 1MHz. > > The points about phase inversion etc is very interesting. Eeek, hadn't > thought of that! This is going to be fun 8) > -- > Nemo A photodiode is a whole different ball of wax. You can look at the current and run it into the inverting 'virtual ground' input of an opamp. The impdedance of a photodiode changes with the amount of current its generating. It's kT/e (25mV @RT) divided by the current. (I think? I'm sure someone will correct me if I'm wrong.) Since it's a current source you have to put the impedance in parallel with it. So at higher currents the impedance goes down.... which is not what you want for a current source. I was thinking I could also look at the current noise from a resistor (put it into the inverting input.) But then how to do the driven shield? Can I put TIA's in series? like this, |\ i--->--- \ | >--- to shield +--+ / | |/ | | |\ +--- \ | >--- to signal GND--+ / |/ The current noises may add, but with fets and 1Meg ohm you hardly care. George H. |