Third order IMD in photomultipliers--weird behaviour

I'm working on an interesting project for a biotech customer, and I could use some wisdom from the assembled multitude.

They're building a scanning fluorescence microscope that scans several lines at once with a single photomultiplier, using a clever scheme to make each line come out at a different RF frequency.

The frequencies are inherently evenly spaced, so third-order IM products cause image artifacts. So they asked me to come up with a way to reduce the IMD.

I've been doing some measurements, and have found the usual cubic dependence of the IM3 products up above about 50 uA of anode current. That's generally a combination of voltage sag in the dynode bias string, plus some space-charge effects. It can be reduced by changing the dynode voltage distribution so that the last few stages have more bias.

Interestingly, though, below ~50 uA there's a broad range of anode currents (20 dB or so) where the IM3 products go up linearly with the signal, so that the SFDR is a nearly constant 60 dB, plus or minus a few. This is without changing ranges on the spectrum analyzer or anything like that.

Ahah, you say, the IM3 is in the illumination source, which runs at a constant drive level. So I thought, but it ain't so.

My illumination source is a pair of blue LEDs, each driven from a separate amp with separate power supplies. To get the spurs big enough to see, I'm using two synthesizers locked together--one is just a frequency doubler hung off my 10 MHz rubidium reference, and the other is a PTS1000 locked to the same reference, running at 20.001 MHz. The analyzer is an HP 8566B, locked to the same reference. That way I can use a 10-Hz resolution bandwidth to see the spurs.

Using a photodiode, I've verified that the spurs in the light sources are below -80 dBc.

There's mention in the literature of a space-charge effect at the photocathode, which would vary less with gain than in the last stages. If that's it, the -60 dB would be set at the very beginning, so it should depend mostly on the actual light intensity at the photocathode.

However, the linear behaviour occurs regardless of whether I'm using a fixed light level and adjusting the anode current with the bias voltage pot or leaving the voltage alone and using a variable optical attenuator.

It also doesn't depend on the attenuation of the signal in the RX chain--adding a 10 dB pad doesn't change the SFDR in that anode current range.

This is, in other words, weird as hell.

Any ideas?

Thanks

Phil Hobbs

--
Dr Philip C D Hobbs 
Principal Consultant 
 Click to see the full signature
Reply to
Phil Hobbs
Loading thread data ...

Crazy idea: electrical resonances inside the PMT. Long wires running to big plates.

formatting link

But the numbers don't look good.

--
John Larkin         Highland Technology, Inc 
picosecond timing   precision measurement  
 Click to see the full signature
Reply to
John Larkin

My tube is a small side-looker, about the size of a 6SN7GT. Why they did that, I don't know, but there's no changing it now.

One point is that the two beams don't illuminate the photocathode in quite the same place. To let me use it in room light, there's a couple of two-inch tubes in a row between the light and the tube. There's a white paper diffuser toward the outside, a 1-inch space, a pinhole, and a 1.6--inch space from there to the PC. The patches of light from the LEDs overlap mostly but not completely on the paper, so parallax makes them illuminate the PC a bit differently.

I'm going to wire up a variable-tap dynode supply and see what that does, but first I'll try changing the frequency offset to make sure it isn't something to do with the Cockroft-Walton generator. 500 kHz should do.

Cheers

Phil Hobbs

--
Dr Philip C D Hobbs 
Principal Consultant 
 Click to see the full signature
Reply to
Phil Hobbs

Two wacky ideas (probably no good): phase-shift the illuminators so the 1F signal's 3F harmonic generates sine, and 3F signal generates cosine. Fourier analyze to separate the contributions (alternately, phase-modulate an illuminator and look at the correlation of the modulation with PM output).

As an alternative, consider regulating the photocurrent: a secondary source of light shines onto the photocathode, driven by the difference of the output current and a fixed reference. Then , instead of analyzing the photocurrent, you analyze the drive to your secondary source (and instead of nonlinearities in a PM tube, it's the nonlinearities in a laser or LED). One variant would be to drive the secondary source from an arbitrary generator, and use monte carlo techniques to train the ARB for minimum deviation at the PM output. Obviously, this would work better at audio frequencies rather than real RF.

Reply to
whit3rd

Phil, I've only got questions... which you should feel free to ignore. To check my perhaps flawed understanding. You are shinning two leds, but modulating one at f1 and the other at f2. (on-off modulation or do you keep them always on and put a sine wave on the top, so to speak?)

You then find that spurs at 2*f1-f2, 2*f2-f1, 2*f1+f2 and 2f2+f1. (What's SFDR?)

These are the second order spurs? 2*f1 and 2*f2? (you can see I'm confused.)

Assuming my understanding is not too far off, would it help to change the intensity of just one led and see how that changed things?

George H.

Reply to
George Herold

Like a 931A?

Add another diffuser to mix them up more? Maybe the electron streams are elbowing oneanother sideways?

--
John Larkin         Highland Technology, Inc 
picosecond timing   precision measurement  
 Click to see the full signature
Reply to
John Larkin

Is not the relationship between incident optical power and output RF power INHERENTLY non-linear.

I seem to recall an interesting disconnect bwtween optical dBs and RF dBs...ie. increading the light 3 dB did not increase the detected RF by 3 dB.

If so isn't the third order IM inherent in the detection process?

Mark

Reply to
makolber

It's an R10699.

Maybe. It's the third order terms showing up without a third order amplitude dependence that has me confused. Possibly some sort of threshold effect, or a C-W interaction.

Cheers

Phil Hobbs

--
Dr Philip C D Hobbs 
Principal Consultant 
 Click to see the full signature
Reply to
Phil Hobbs

Thanks. I'm just doing the PMT module on this one, so large scale changes to the instrument aren't in the cards.

Cheers

Phil Hobbs

--
Dr Philip C D Hobbs 
Principal Consultant 
 Click to see the full signature
Reply to
Phil Hobbs

How stiff are the bias voltages on the dynodes? My first guess is that sin ce the "photocurrent" has to come from someplace, you'd see a lot of nonlin earity if the source impedance is high. That's probably the first thing yo u checked as well, but I'm a bit surprised this method works at all.

Might be able to compensate in software, if you characterize the light-vers us-current transfer function carefully enough.

-- john, KE5FX

Reply to
John Miles, KE5FX

Yes, I'm looking at 2f1-f2 and 2f2-f1. Third order spurs often land in-band, whereas second order ones are near 2f or near DC, where you can usually get rid of them, except for the amplitude compression of the fundamental components. (When a single-ended amp starts clipping, you get lots of second harmonic and the fundamental sees less gain.)

Spurious-free dynamic range. I'm misapplying it slightly to denote (signal power per tone)/(power in worst spur).

If it were the light sources talking to each other, e.g. a bit of 20.001 coupling into the 20.000 drive, the light from each LED would have IM3 components, but they don't.

Maybe. I just don't know what physics could generate cubic products without cubic amplitude dependence.

Cheers

Phil Hobbs

--
Dr Philip C D Hobbs 
Principal Consultant 
 Click to see the full signature
Reply to
Phil Hobbs

It's square law--photocurrent is linear in optical power, and P_RF = I**2 R. You can write that in terms of the fields, but with two separate LEDs, all the cross terms sum to zero.

Detecting the LED light with a photodiode gives me a clean result.

Right, it's 6 dB. Electrical SNR is the square of optical SNR for the same reason.

Nope, it's pure second-order. The photocurrents from the two LEDs should just add.

Thanks

Phil Hobbs

--
Dr Philip C D Hobbs 
Principal Consultant 
 Click to see the full signature
Reply to
Phil Hobbs

Well not exactly, I was thinking that if one light source had a decent amount of

2*f1 then it would mix linearly with f2 to give something at you 2*f1-f2.

George H.

Reply to
George Herold

Huh, I think we use that same tube, and hamamatsu's C-W HV generator.

George H.

Reply to
George Herold

Ah, I see, thanks. That may well be part of the story--it would have a quadratic dependence instead of cubic, and maybe there's a phase cancellation in there someplace. I'm seeing about 50% modulation depth from the LEDs, but it might not be that accurately sinusoidal. I'll have to balance it with a dark-emitting diode to cancel the even harmonics. ;)

Cheers

Phil Hobbs

--
Dr Philip C D Hobbs 
Principal Consultant 
 Click to see the full signature
Reply to
Phil Hobbs

Is the light output proportional to current? Then drive the LED's with linear current drive. ...Jim Thompson

--
| James E.Thompson                                 |    mens     | 
| Analog Innovations                               |     et      | 
 Click to see the full signature
Reply to
Jim Thompson

It is pretty nearly, but I have to drive them fairly hard to get them to be fast enough. (High level injection speeds up recombination a lot.)

Cheers

Phil Hobbs

--
Dr Philip C D Hobbs 
Principal Consultant 
 Click to see the full signature
Reply to
Phil Hobbs

How hard would it be to change the frequencies so that the 3rd order products fall between the bands, rather than on top of them?

Allan

Reply to
Allan Herriman

Not possible. In the customer's scheme, the frequency deltas are harmonically related.

Cheers

Phil Hobbs

--
Dr Philip C D Hobbs 
Principal Consultant 
 Click to see the full signature
Reply to
Phil Hobbs

What frequency and P-P current requirement? ...Jim Thompson

--
| James E.Thompson                                 |    mens     | 
| Analog Innovations                               |     et      | 
 Click to see the full signature
Reply to
Jim Thompson

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.