Ns pulse detection circuit understanding to modify

Thread Starter

Lakey009

Joined Apr 30, 2017
8
Hello all forum newbie,

I have been trying to understand part of a larger circuit, pictured. The circuit is designed to detect a small pulse produced by a detector. The aim is to understand and upgrade the 30year old circuit hoping for a small increase in efficiency. It is part of a work based side project of mine and I'm at the point where I'm assessing the viability.
IMG_20170426_101816~2.jpg

My ongoing/learning interpretation so far is, the 47pF cap near TP4 (far left picture) is present to remove the 5kV DC voltage? A voltage divider lowers the potential at the inputs, possibly helps reduce ringing too, terminated to Gnd through divder? The op amp seems to be running as a differentiator. Could you please help me understand if my analysis thus far is correct and what are the two diodes for across and Vin on the op-amp?

Many thanks

Dan
 

Attachments

Thread Starter

Lakey009

Joined Apr 30, 2017
8
A little more to help which I missed.

The circuits main aim is to detect the low signal pulses while minimising propagation delay to accurately measure number and the time they occurred. The pulses are around 5 ns. The pulses have a frequency of 1hz to 5Mhz.

This part of the circuit detects the small fast pulse and converts it to a TTL where the pulse is later shaped/ elongated for counting/ timing.

Hope this helps you help me ☺
 

AlbertHall

Joined Jun 4, 2014
8,184
Yes to the function of the 47pF capacitor.
The VV100B is fast amplifier with a gain of 10, so the two of them cascaded amplifies the input pulse by 100.
 

Thread Starter

Lakey009

Joined Apr 30, 2017
8
Thanks Albert, do you think the voltage divider is to reduce ringing + keep the amp in the small signal bandwidth for greater bandwidth of the amp vv100B is 400Mhz, which a quick look by me seems v.good for its age.

Reason for this line of thinking is, if the above is true I could use a 'better' new op amp, with greater bandwidth. So removing the divider R17 the op amp gets a larger pulse to detect, which should improve its ability to detect smaller + faster pulses.while R4 is adjusted but remains to sink the pulse.

Thanks for your help

Dan
 

MrChips

Joined Oct 2, 2009
19,154
R1, R2, C1, C2, C3 are low pass filters to remove HF noise from the HV supply.
R3 is the load resistor on your detector.
CL couples the signal to the amplifier. You cannot remove this.
R17 and R4 is a voltage divider. R4 presents a 100Ω input impedance.
Diodes CR1 and CR2 are your protection diodes to prevent high amplitude signals from ruining your amplifier.
The opamps are not differentiators. They are AC coupled to amplify the pulses only.

What is your meaning of "small increase in efficiency"?

I do not see any way of removing any of the components.
 

AlbertHall

Joined Jun 4, 2014
8,184
I am nobody's expert of fast pulse amplifiers but remember that it is not only the bandwidth you need to consider but the slew rate. The bigger your output voltage swing, the more important is the slew rate.
 

bertus

Joined Apr 5, 2008
19,939
Hello,

The pulse signal seems to com from an E.M. ( Electron Multiplier ).
These type of multipliers work with high voltages.
Electron multipliers are often used in mass spectroscopy instruments.
I also found a datasheet of the VV100B.

Bertus
 

Attachments

Thread Starter

Lakey009

Joined Apr 30, 2017
8
R1, R2, C1, C2, C3 are low pass filters to remove HF noise from the HV supply.
R3 is the load resistor on your detector.
CL couples the signal to the amplifier. You cannot remove this.
R17 and R4 is a voltage divider. R4 presents a 100Ω input impedance.
Diodes CR1 and CR2 are your protection diodes to prevent high amplitude signals from ruining your amplifier.
The opamps are not differentiators. They are AC coupled to amplify the pulses only.

What is your meaning of "small increase in efficiency"?

I do not see any way of removing any of the components.
Yes I'm aware of the function of those. Do you think a I could remove R17? Ah that is what the diodes are for!

Ah yes there is no feedback resistor so the output wouldn't be the differential of the input.

Some pulses are smaller than others, so if R17 is removed I would get an improvement of the signal to noise ratio of the signal across the op-amp, which inturn equates to output. I should be able to detect smaller pulses = small increase in efficiency of pulse counting.

Thanks for you help!
 

Thread Starter

Lakey009

Joined Apr 30, 2017
8
Hello,

The pulse signal seems to com from an E.M. ( Electron Multiplier ).
These type of multipliers work with high voltages.
Electron multipliers are often used in mass spectroscopy instruments.
I also found a datasheet of the VV100B.

Bertus
Thanks for the datasheet, I could only find a limited poor website link.

Yes you are correct this circuit is from an electron multiplier, which is in operation on a secondary ion mass spectrometry machine.
 

Thread Starter

Lakey009

Joined Apr 30, 2017
8
Ah yes Bertus! The data sheet shows I don't need R17. So I think could select its size/possibly remove it, within the limits of the op-amp I replace it with in a "modernised" design?
 

MrChips

Joined Oct 2, 2009
19,154
RC circuits can become differentiators if the RC time-constants are much shorter than the pulse-width of your signal. In this case, C5 and C22, 0.1μF into 51Ω has a long time constant.

Any differentiation comes from CL and R17. Removing R17 is going to alter the shape of your pulses.
If you need larger signals, you need to increase the gain of the amplifiers. SNR is not going to improve.
 

bertus

Joined Apr 5, 2008
19,939
Hello,

The resistor R17 will limit the current for the protection diodes in case the EM has a flash over, wich might happen when the vacuum is not optimal.
Most EM do require a certain vacuum level.
I know that our multipliers work at about 2000 Volts, the powersupply for the EM can be used upto 3000 Volts.

You might want to have a look at this site:
http://www.shimadzu.com/an/electron_multiplication.html

I also attached a PDF about EM and Ion multipliers.

Bertus
 

Attachments

Thread Starter

Lakey009

Joined Apr 30, 2017
8
Amazing, Thanks yes your correct with the time constants, it's been a while since I've delved into electronics.

Yes I agree it will change the shape of the pulse too much.

From what I've read I can achieve a much greater bandwidth using lower gains + cascade. So yes I can see a higher gain is useful but, it will amplify the noise too and will not improve SNR as you correctly say.

Thanks for everyone's help, you've answered my questions ☺

Before someone jumps on me mentioning bandwidth again, I've been looking into possible substitutions. AD8009 seems like a contender 5.5kV slew rate + other properties bonus is removing some of those power supplies for the VV100B amps should help a little too.

Dan
 

Thread Starter

Lakey009

Joined Apr 30, 2017
8
Hello,

The resistor R17 will limit the current for the protection diodes in case the EM has a flash over, wich might happen when the vacuum is not optimal.
Most EM do require a certain vacuum level.
I know that our multipliers work at about 2000 Volts, the powersupply for the EM can be used upto 3000 Volts.

You might want to have a look at this site:
http://www.shimadzu.com/an/electron_multiplication.html

I also attached a PDF about EM and Ion multipliers.

Bertus

Ah Thanks for the handbook + link, yes you are correct it limits the current. Our EM run at 4.5kV, vacuum is really good but it could happen.I probably shouldn't reduce R17, just ensure everything's ok for a new op-amp.

Dan
 
Top