PMT(Photomultiplier Tubes) speed issue
Does anyone know about PMT(Photomultiplier Tubes), PMT has higher sensitivities than normal photodiode.
and I check the datasheet for some PMT ,they has an rise time around 5ns even to 1ns.
[B]Does this mean it can received 200Mbps data? can be used in wireless communication ? my input is 50 mbps.
why in hamamatsu's website, which has a list of application for PMT, and it seems PMT is not used for wireless communication.
thanks~
Maybe in theory you can, although I seem to think they have a faster rise time than fall time. However, an avalanche photodiode would probably be better (and smaller and cheaper) if you really need the gain. It is mainly APDs that I have experience with. You will find APDs available aimed at fibre optic communications.
Keith.
hi, keith12rs I remember we had this talk before....
APD is not suitable for my application , because they need high voltage around 100v.
And in my application it can not higher than 20v. that's the reason i can not use APD.
And for PMT. I just check the handbook of the PMT from hamamatsu. it says, for "metal channel" PMT, the rise time 0.65-1.5ns, fall time 1-3nm,
does that mean, PMT can at least receiver optical signal at around 100Mbps?
please take a look to the voltage required for a PMT to work: in general is at least 800-1200V!
So, if you can't have 100V, I think it will be impossible to get much higher voltage.
Why do you not want more than 20V? It is not difficult to generate the voltages required for APDs. As has already been pointed out, PMTs certainly won't reduce your high voltage problem!
Keith.
i checked this PMT
The input voltage is 11.5- 15.5 volts.
Do you mean inside this PMT there is high voltage generated?
---------- Post added at 13:00 ---------- Previous post was at 12:46 ----------
Because in my application,the photo detector APD , photodiode or PMT should be attached to the person ( contact the skin behind the ear) all day long.
So it would be dangerous to use APD which has a bias voltage around 100 volts.
anyhow, does PMT has a high voltage (1000v) inside ? it input voltage is 15 volts
The units you are looking at have the high voltage power supply built in - see the first sentence of the page you linked to.
I don't think high voltages are precluded from medical devices - you just need to comply with appropriate regulations.
I think an APD would be far more appropriate - PMTs are bulky and heavy compared to APDs. I wouldn't want a PMT behind my ear! I would look at trying to avoid them altogether, but that depends on why you have so little light to work with. You do need to look at the whole system though. You are talking about 100Mb/s communications but that means a low transimpedance for the amplifier. Why the high data rate?
Keith.
this optical link is for neural recording, the coming data is from 20Mbps to 100Mbps, let's say typically 50Mbps.
why high data rate means low transimpedcance amplifier?
and you think APD with 100volts bias is suitable for a person to carry it everyday, when you " comply with appropriate regulations" ?
Why is the amount of light so low that you need a PMT/APD? I assume the distance is very short?
If you want a high speed you often end up with a high input bias current (although you can make CMOS/HEMT designs with low impedance) so a high feedback resistor can be impractical. Also, input capacitance, photodiode capacitance and stray capacitance all limit the speed with high transimpedances. That is why you will find all high speed transimpedance amplifiers are low transimpedance - often just a few k ohms.
I haven't a copy of the appropriate regulations for medical devices here, but last time I looked I don't remember anything that would exclude high voltages. There will be plenty about how accessible any voltages are and whether they become accessible with fault conditions.
Keith.
1 Because the transmiter is beneath the skin(maybe even beneth the skull), and receiver is outside of the human body, like I said, behind the ear, contact to skin, because the skin there is relatively thin...
So, the skin will absorb and scatter a lot of opitcal power. only a little can reach the receiver. therefore I need a high sensitive and large size photodiode to collect optical power as much as possible.
the transmitter is inside the human body, that means you can not increase the optical power, because the power budget of the transmitter.
2 why high speed leads to high input bias current ? ( does input bias current means the current flow through the photodiode and feedback resistor?)
I understand the trade off between the speed , size , noise.
Large size photodiode will leads to large junction capacitance, this input capacitance will 1. decrease the speed , 2 increase the noise of the TIA. Am I right?
3 can you sent me some reference about this high voltage stuff? or some key words ?
Why use light? I would have thought RF would be a better transmission method under the circumstances.
Is there an optical wavelength that transmits better through the skin? IR is absorbed by water so is probably not a great choice.
You are right about the speed trade-offs, but even with a small photodiode you have amplifier input capacitance and the finite impedances you are dealing with. You need the input impedance of the transimpedance amplifier to appear to be low at your working frequency but it needs to be low due to the transimpedance, not because the signal is slugged by a large input capacitance. That makes the transimpedance amplifier design tricky. A lot of fast ones are bipolar.
I amm not sure offhand of the right regulations - it depends on the country you are in. IEC 60601 is the European one. The FDA regulations will be under Title 21 CFR (Code of Federal Regulations). The FDA ones are a good start because, unlike European ones, they are freely available I think.
Keith
1 RF is regulated... only 402-405M can be used for medical which called MICS.
2 850nm is the best choice, so many people done such experiments.
3 actually I am not designing an TIA, I am seeking for a TIA which is design to work with a large size photodiode (means large input capacitance).
all the TIA seems design works with 0.5-10pF input capacitor. But so far, the photodiode which currently seems the best choise (large size and 50M bandwith) S6967 has a 50 pF junction capacitance. do you know which TIA can deal with that?
4 I am in Germany, and I will check for this standards to see if APD suitable.
I DO NOT SUGGEST the usage of a PMT: if you will expose it to even very low light (room) levels you will destroy it, especially if you are using the typical power supplies for PMTs.
Please note that in the datasheet it is written they are used for PHOTON COUNTING: they are TOO SENSITIVE to being used in a not controlled environment!
I'm not particularly familar with MEDS (medical dataservice) applications, but I assume it's rather the other way around. The specific MEDS frequency 401-406 MHz, also a number of other frequencies listed in the ETSI documents, are reserved for it. General purpose SRD radio frequencies without special restrictions should be available as well.
actually, one of the purpose of the project is using infrared communication comparing with RF...
---------- Post added at 20:57 ---------- Previous post was at 19:42 ----------
I probably get it.
the input capacitor plus the capacitance of the amplifier together with the feedback resistermakes a RC network, that limit the bandwidth.
So for large size photodiode(large C), I should choose some TIA with low gain(small R) to maintain the speed.
am I right