On Wed, 3 Aug 2016 07:05:18 -0700 (PDT), George Herold <gherold@teachspin.com> wrote:>On Wednesday, August 3, 2016 at 4:46:31 AM UTC-4, piglet wrote: >> On 03/08/2016 04:43, T Obulesu wrote: >> > On Tuesday, 2 August 2016 08:58:42 UTC+5:30, T Obulesu wrote: >> >> Hello guys.. >> >> >> >> I want to design a trans-impedance amplifier to detect currents of the APD (in the range of 200 nA to 100 micro A). >> >> >> >> Could some one suggest me the best IC with less bias current, single supply (+5 V only) and how to design a high sensitive TIA..? (typically, for every 100 nA change in the current, 100 mV voltage should be displayed). >> > >> > Well....Bandwidth required is 1 MHz.. >> > APD biasing voltage is 26-30V. >> > Photodiode capacitance is 5pF >> > >> >> How about the OPA380 ? But my fastest opto TIA is only few kHz. >> >> piglet > >Huh, yeah something like that. >To the OP, >Putting in "rough" numbers. >Say 1 Meg of feed back R, and another ~5pF of stray and opamp input C. >So 1 Meg ohm*10 pF = 10^-5 sec, ~10^4 Hz (2*pi =~ 10) >You'll need about a 100MHz GBW opamp. >piglet's suggestion look good. > >George H.The 1 meg feedback resistor and the 10 pF of capacitance put a 10 microsecond tau, 16 KHz, lowpass filter INSIDE the opamp feedback loop. That will make it ring like hell or oscillate, and will make a gigantic noise spike. Adding a cap across the 1M feedback resistor is the classic/clumsy/slow way to avoid the stability and noise crisis. Phil's paper and book discuss this. A good TIA will have 100x the bandwidth of the obvious one. APDs have a lot of internal gain, and are inherently noisy, so it might be OK to use a fast low-transconductance stage followed by a voltage amp or two. I'm playing with a couple of new TIA circuits now, just for fun, one being a cascode+MMIC thing. There are some amazing MMICS around lately, TIAs that aren't identified as such. Of course, that circuit will be AC coupled, so I'd need to embellish it if I want DC response. -- John Larkin Highland Technology, Inc lunatic fringe electronics
Need help to design a transimpedance amplifier
Started by ●August 2, 2016
Reply by ●August 3, 20162016-08-03
Reply by ●August 3, 20162016-08-03
On Wednesday, August 3, 2016 at 10:32:54 AM UTC-4, John Larkin wrote:> On Wed, 3 Aug 2016 07:05:18 -0700 (PDT), George Herold > <gherold@teachspin.com> wrote: > > >On Wednesday, August 3, 2016 at 4:46:31 AM UTC-4, piglet wrote: > >> On 03/08/2016 04:43, T Obulesu wrote: > >> > On Tuesday, 2 August 2016 08:58:42 UTC+5:30, T Obulesu wrote: > >> >> Hello guys.. > >> >> > >> >> I want to design a trans-impedance amplifier to detect currents of the APD (in the range of 200 nA to 100 micro A). > >> >> > >> >> Could some one suggest me the best IC with less bias current, single supply (+5 V only) and how to design a high sensitive TIA..? (typically, for every 100 nA change in the current, 100 mV voltage should be displayed). > >> > > >> > Well....Bandwidth required is 1 MHz.. > >> > APD biasing voltage is 26-30V. > >> > Photodiode capacitance is 5pF > >> > > >> > >> How about the OPA380 ? But my fastest opto TIA is only few kHz. > >> > >> piglet > > > >Huh, yeah something like that. > >To the OP, > >Putting in "rough" numbers. > >Say 1 Meg of feed back R, and another ~5pF of stray and opamp input C. > >So 1 Meg ohm*10 pF = 10^-5 sec, ~10^4 Hz (2*pi =~ 10) > >You'll need about a 100MHz GBW opamp. > >piglet's suggestion look good. > > > >George H. > > The 1 meg feedback resistor and the 10 pF of capacitance put a 10 > microsecond tau, 16 KHz, lowpass filter INSIDE the opamp feedback > loop. That will make it ring like hell or oscillate, and will make a > gigantic noise spike.Grin.. yeah, killing any oscillation and measuring the noise will be his problem. I was just trying to flesh out the details of how piglet picked the OPA380> > Adding a cap across the 1M feedback resistor is the > classic/clumsy/slow way to avoid the stability and noise crisis. > Phil's paper and book discuss this.Classic, clumsy and slow, pretty much describes my approach to circuit design. :^) I haven't done a TIA in a while but the stray ~0.1pF of the 1 Meg FB resistor may be enough to stabilize it.> > A good TIA will have 100x the bandwidth of the obvious one. > > APDs have a lot of internal gain, and are inherently noisy, so it > might be OK to use a fast low-transconductance stage followed by a > voltage amp or two.I've never used an APD... I did buy a few from DK. I buy fun parts, and then don't have time to play with them. George H.> > I'm playing with a couple of new TIA circuits now, just for fun, one > being a cascode+MMIC thing. There are some amazing MMICS around > lately, TIAs that aren't identified as such. Of course, that circuit > will be AC coupled, so I'd need to embellish it if I want DC response. > > > > > -- > > John Larkin Highland Technology, Inc > > lunatic fringe electronics
Reply by ●August 3, 20162016-08-03
On Wed, 3 Aug 2016 07:59:03 -0700 (PDT), George Herold <gherold@teachspin.com> wrote:>On Wednesday, August 3, 2016 at 10:32:54 AM UTC-4, John Larkin wrote: >> On Wed, 3 Aug 2016 07:05:18 -0700 (PDT), George Herold >> <gherold@teachspin.com> wrote: >> >> >On Wednesday, August 3, 2016 at 4:46:31 AM UTC-4, piglet wrote: >> >> On 03/08/2016 04:43, T Obulesu wrote: >> >> > On Tuesday, 2 August 2016 08:58:42 UTC+5:30, T Obulesu wrote: >> >> >> Hello guys.. >> >> >> >> >> >> I want to design a trans-impedance amplifier to detect currents of the APD (in the range of 200 nA to 100 micro A). >> >> >> >> >> >> Could some one suggest me the best IC with less bias current, single supply (+5 V only) and how to design a high sensitive TIA..? (typically, for every 100 nA change in the current, 100 mV voltage should be displayed). >> >> > >> >> > Well....Bandwidth required is 1 MHz.. >> >> > APD biasing voltage is 26-30V. >> >> > Photodiode capacitance is 5pF >> >> > >> >> >> >> How about the OPA380 ? But my fastest opto TIA is only few kHz. >> >> >> >> piglet >> > >> >Huh, yeah something like that. >> >To the OP, >> >Putting in "rough" numbers. >> >Say 1 Meg of feed back R, and another ~5pF of stray and opamp input C. >> >So 1 Meg ohm*10 pF = 10^-5 sec, ~10^4 Hz (2*pi =~ 10) >> >You'll need about a 100MHz GBW opamp. >> >piglet's suggestion look good. >> > >> >George H. >> >> The 1 meg feedback resistor and the 10 pF of capacitance put a 10 >> microsecond tau, 16 KHz, lowpass filter INSIDE the opamp feedback >> loop. That will make it ring like hell or oscillate, and will make a >> gigantic noise spike. > >Grin.. yeah, killing any oscillation and measuring the noise >will be his problem. I was just trying to flesh out the details >of how piglet picked the OPA380Its input bias current is high. It will rail with a 1M feedback resistor. Input current noise is correspondingly high.>> >> Adding a cap across the 1M feedback resistor is the >> classic/clumsy/slow way to avoid the stability and noise crisis. >> Phil's paper and book discuss this. > >Classic, clumsy and slow, pretty much describes my approach to >circuit design. :^) I haven't done a TIA in a while but the >stray ~0.1pF of the 1 Meg FB resistor may be enough to stabilize it. > > >> >> A good TIA will have 100x the bandwidth of the obvious one. >> >> APDs have a lot of internal gain, and are inherently noisy, so it >> might be OK to use a fast low-transconductance stage followed by a >> voltage amp or two. > >I've never used an APD... I did buy a few from DK. I buy fun parts, and >then don't have time to play with them.They have a lot of gain... approaching infinite as you crank the voltage up... but they are noisy and temperature sensitive. -- John Larkin Highland Technology, Inc lunatic fringe electronics
Reply by ●August 3, 20162016-08-03
On Wednesday, August 3, 2016 at 11:36:21 AM UTC-4, John Larkin wrote:> On Wed, 3 Aug 2016 07:59:03 -0700 (PDT), George Herold > <gherold@teachspin.com> wrote: > > >On Wednesday, August 3, 2016 at 10:32:54 AM UTC-4, John Larkin wrote: > >> On Wed, 3 Aug 2016 07:05:18 -0700 (PDT), George Herold > >> <gherold@teachspin.com> wrote: > >> > >> >On Wednesday, August 3, 2016 at 4:46:31 AM UTC-4, piglet wrote: > >> >> On 03/08/2016 04:43, T Obulesu wrote: > >> >> > On Tuesday, 2 August 2016 08:58:42 UTC+5:30, T Obulesu wrote: > >> >> >> Hello guys.. > >> >> >> > >> >> >> I want to design a trans-impedance amplifier to detect currents of the APD (in the range of 200 nA to 100 micro A). > >> >> >> > >> >> >> Could some one suggest me the best IC with less bias current, single supply (+5 V only) and how to design a high sensitive TIA..? (typically, for every 100 nA change in the current, 100 mV voltage should be displayed). > >> >> > > >> >> > Well....Bandwidth required is 1 MHz.. > >> >> > APD biasing voltage is 26-30V. > >> >> > Photodiode capacitance is 5pF > >> >> > > >> >> > >> >> How about the OPA380 ? But my fastest opto TIA is only few kHz. > >> >> > >> >> piglet > >> > > >> >Huh, yeah something like that. > >> >To the OP, > >> >Putting in "rough" numbers. > >> >Say 1 Meg of feed back R, and another ~5pF of stray and opamp input C. > >> >So 1 Meg ohm*10 pF = 10^-5 sec, ~10^4 Hz (2*pi =~ 10) > >> >You'll need about a 100MHz GBW opamp. > >> >piglet's suggestion look good. > >> > > >> >George H. > >> > >> The 1 meg feedback resistor and the 10 pF of capacitance put a 10 > >> microsecond tau, 16 KHz, lowpass filter INSIDE the opamp feedback > >> loop. That will make it ring like hell or oscillate, and will make a > >> gigantic noise spike. > > > >Grin.. yeah, killing any oscillation and measuring the noise > >will be his problem. I was just trying to flesh out the details > >of how piglet picked the OPA380 > > Its input bias current is high. It will rail with a 1M feedback > resistor. Input current noise is correspondingly high. >Oops.. my spec sheet says 50 pA.. that seems fine. http://www.ti.com.cn/cn/lit/ds/symlink/opa380.pdf> > >> > >> Adding a cap across the 1M feedback resistor is the > >> classic/clumsy/slow way to avoid the stability and noise crisis. > >> Phil's paper and book discuss this. > > > >Classic, clumsy and slow, pretty much describes my approach to > >circuit design. :^) I haven't done a TIA in a while but the > >stray ~0.1pF of the 1 Meg FB resistor may be enough to stabilize it. > > > > > >> > >> A good TIA will have 100x the bandwidth of the obvious one. > >> > >> APDs have a lot of internal gain, and are inherently noisy, so it > >> might be OK to use a fast low-transconductance stage followed by a > >> voltage amp or two. > > > >I've never used an APD... I did buy a few from DK. I buy fun parts, and > >then don't have time to play with them. > > They have a lot of gain... approaching infinite as you crank the > voltage up... but they are noisy and temperature sensitive.I really want to bias an APD above the breakdown voltage and see if I could use it as a SPAD. (Geiger-Muller mode) It will probably stink. George H.> > > -- > > John Larkin Highland Technology, Inc > > lunatic fringe electronics
Reply by ●August 3, 20162016-08-03
On Wed, 3 Aug 2016 08:43:00 -0700 (PDT), George Herold <gherold@teachspin.com> wrote:>On Wednesday, August 3, 2016 at 11:36:21 AM UTC-4, John Larkin wrote: >> On Wed, 3 Aug 2016 07:59:03 -0700 (PDT), George Herold >> <gherold@teachspin.com> wrote: >> >> >On Wednesday, August 3, 2016 at 10:32:54 AM UTC-4, John Larkin wrote: >> >> On Wed, 3 Aug 2016 07:05:18 -0700 (PDT), George Herold >> >> <gherold@teachspin.com> wrote: >> >> >> >> >On Wednesday, August 3, 2016 at 4:46:31 AM UTC-4, piglet wrote: >> >> >> On 03/08/2016 04:43, T Obulesu wrote: >> >> >> > On Tuesday, 2 August 2016 08:58:42 UTC+5:30, T Obulesu wrote: >> >> >> >> Hello guys.. >> >> >> >> >> >> >> >> I want to design a trans-impedance amplifier to detect currents of the APD (in the range of 200 nA to 100 micro A). >> >> >> >> >> >> >> >> Could some one suggest me the best IC with less bias current, single supply (+5 V only) and how to design a high sensitive TIA..? (typically, for every 100 nA change in the current, 100 mV voltage should be displayed). >> >> >> > >> >> >> > Well....Bandwidth required is 1 MHz.. >> >> >> > APD biasing voltage is 26-30V. >> >> >> > Photodiode capacitance is 5pF >> >> >> > >> >> >> >> >> >> How about the OPA380 ? But my fastest opto TIA is only few kHz. >> >> >> >> >> >> piglet >> >> > >> >> >Huh, yeah something like that. >> >> >To the OP, >> >> >Putting in "rough" numbers. >> >> >Say 1 Meg of feed back R, and another ~5pF of stray and opamp input C. >> >> >So 1 Meg ohm*10 pF = 10^-5 sec, ~10^4 Hz (2*pi =~ 10) >> >> >You'll need about a 100MHz GBW opamp. >> >> >piglet's suggestion look good. >> >> > >> >> >George H. >> >> >> >> The 1 meg feedback resistor and the 10 pF of capacitance put a 10 >> >> microsecond tau, 16 KHz, lowpass filter INSIDE the opamp feedback >> >> loop. That will make it ring like hell or oscillate, and will make a >> >> gigantic noise spike. >> > >> >Grin.. yeah, killing any oscillation and measuring the noise >> >will be his problem. I was just trying to flesh out the details >> >of how piglet picked the OPA380 >> >> Its input bias current is high. It will rail with a 1M feedback >> resistor. Input current noise is correspondingly high. >> >Oops.. my spec sheet says 50 pA.. that seems fine. >http://www.ti.com.cn/cn/lit/ds/symlink/opa380.pdfI probably typoed the part number. I never learned to type. That will just about do what he wants in a single stage. There is still a voltage swing crisis at 100 uA. Switching the feedback resistor has its own issues! -- John Larkin Highland Technology, Inc picosecond timing precision measurement jlarkin att highlandtechnology dott com http://www.highlandtechnology.com
Reply by ●August 4, 20162016-08-04
On Tuesday, 2 August 2016 08:58:42 UTC+5:30, T Obulesu wrote:> Hello guys.. > > I want to design a trans-impedance amplifier to detect currents of the APD (in the range of 200 nA to 100 micro A). > > Could some one suggest me the best IC with less bias current, single supply (+5 V only) and how to design a high sensitive TIA..? (typically, for every 100 nA change in the current, 100 mV voltage should be displayed).Let me say why am looking for new design.. We are currently using the receiver module which has three stages: 1. Trans Impedance stage 2. High Pass filter (HPF) stage 3. Unity gain amplifier (just for inverting the output of the HPF) Well...we used LM2662 as a -5V supply.. Here on we could see the hell noise below 100kHz.. I couldn't get rid of this noise by using bypass caps..but I could just reduced it.. Yet there is a lot of noise all across the circuit ranging from few kHz to hundreds of MHz... https://drive.google.com/open?id=0B_khpIBuCMVdb18ySmZubGsyRUU is the link where I have uploaded couple of documents.. I really don't know what sort of noise it is and from where it is coming....
Reply by ●August 4, 20162016-08-04
On Tuesday, 2 August 2016 08:58:42 UTC+5:30, T Obulesu wrote:> Hello guys.. > > I want to design a trans-impedance amplifier to detect currents of the APD (in the range of 200 nA to 100 micro A). > > Could some one suggest me the best IC with less bias current, single supply (+5 V only) and how to design a high sensitive TIA..? (typically, for every 100 nA change in the current, 100 mV voltage should be displayed).Let me say why am looking for new design.. We are currently using the receiver module which has three stages: 1. Trans Impedance stage 2. High Pass filter (HPF) stage 3. Unity gain amplifier (just for inverting the output of the HPF) Well...we used LM2662 as a -5V supply.. Here on we could see the hell noise below 100kHz.. I couldn't get rid of this noise by using bypass caps..but I could just reduced it.. Yet there is a lot of noise all across the circuit ranging from few kHz to hundreds of MHz... https://drive.google.com/open?id=0B_khpIBuCMVdb18ySmZubGsyRUU https://drive.google.com/open?id=0B_khpIBuCMVdbUVRWW85MHFmSXM https://drive.google.com/drive/u/0/folders/0B_khpIBuCMVdMmplSUc3U2FNdjA https://drive.google.com/open?id=0B_khpIBuCMVdeVMxZkRGR1NvM2M are the links where I have uploaded couple of documents.. I really don't know what sort of noise it is and from where it is coming...
Reply by ●August 4, 20162016-08-04
On 08/04/2016 02:31 AM, T Obulesu wrote:> On Tuesday, 2 August 2016 08:58:42 UTC+5:30, T Obulesu wrote: >> Hello guys.. >> >> I want to design a trans-impedance amplifier to detect currents of >> the APD (in the range of 200 nA to 100 micro A). >> >> Could some one suggest me the best IC with less bias current, >> single supply (+5 V only) and how to design a high sensitive TIA..? >> (typically, for every 100 nA change in the current, 100 mV voltage >> should be displayed). > > Let me say why am looking for new design.. We are currently using the > receiver module which has three stages: 1. Trans Impedance stage 2. > High Pass filter (HPF) stage 3. Unity gain amplifier (just for > inverting the output of the HPF) > > Well...we used LM2662 as a -5V supply.. Here on we could see the hell > noise below 100kHz.. I couldn't get rid of this noise by using bypass > caps..but I could just reduced it.. Yet there is a lot of noise all > across the circuit ranging from few kHz to hundreds of MHz...> > https://drive.google.com/open?id=0B_khpIBuCMVdb18ySmZubGsyRUU > > https://drive.google.com/open?id=0B_khpIBuCMVdbUVRWW85MHFmSXM > > https://drive.google.com/drive/u/0/folders/0B_khpIBuCMVdMmplSUc3U2FNdjA > > https://drive.google.com/open?id=0B_khpIBuCMVdeVMxZkRGR1NvM2M > > are the links where I have uploaded couple of documents.. > > I really don't know what sort of noise it is and from where it is > coming... > Well, if you're connecting a photodiode directly to a SMPS, what did you expect? Hint: Taking a spiky supply rail, differentiating it with a small capacitor, and sticking the resulting current into a high-gain current amplifier may not be what you want to do. Of course, three lines of algebra would have told you what the results were going to be, and you'd have learned something instead of just throwing spitballs. This theory stuff really works. ;) We've had several threads about capacitance multipliers recently, including one where I posted an LTspice schematic of a 2-stage cap multiplier that will knock that stuff down by about 180 dB, assuming that you have adequate RF shielding. Cheers Phil Hobbs -- Dr Philip C D Hobbs Principal Consultant ElectroOptical Innovations LLC Optics, Electro-optics, Photonics, Analog Electronics 160 North State Road #203 Briarcliff Manor NY 10510 hobbs at electrooptical dot net http://electrooptical.net
Reply by ●August 4, 20162016-08-04
On Thursday, August 4, 2016 at 8:16:51 AM UTC-4, Phil Hobbs wrote:> On 08/04/2016 02:31 AM, T Obulesu wrote: > > On Tuesday, 2 August 2016 08:58:42 UTC+5:30, T Obulesu wrote: > >> Hello guys.. > >> > >> I want to design a trans-impedance amplifier to detect currents of > >> the APD (in the range of 200 nA to 100 micro A). > >> > >> Could some one suggest me the best IC with less bias current, > >> single supply (+5 V only) and how to design a high sensitive TIA..? > >> (typically, for every 100 nA change in the current, 100 mV voltage > >> should be displayed). > > > > Let me say why am looking for new design.. We are currently using the > > receiver module which has three stages: 1. Trans Impedance stage 2. > > High Pass filter (HPF) stage 3. Unity gain amplifier (just for > > inverting the output of the HPF) > > > > Well...we used LM2662 as a -5V supply.. Here on we could see the hell > > noise below 100kHz.. I couldn't get rid of this noise by using bypass > > caps..but I could just reduced it.. Yet there is a lot of noise all > > across the circuit ranging from few kHz to hundreds of MHz... > > > > https://drive.google.com/open?id=0B_khpIBuCMVdb18ySmZubGsyRUU > > > > https://drive.google.com/open?id=0B_khpIBuCMVdbUVRWW85MHFmSXM > > > > https://drive.google.com/drive/u/0/folders/0B_khpIBuCMVdMmplSUc3U2FNdjA > > > > https://drive.google.com/open?id=0B_khpIBuCMVdeVMxZkRGR1NvM2M > > > > are the links where I have uploaded couple of documents.. > > > > I really don't know what sort of noise it is and from where it is > > coming... > > > > Well, if you're connecting a photodiode directly to a SMPS, what did you > expect? > > Hint: Taking a spiky supply rail, differentiating it with a small > capacitor, and sticking the resulting current into a high-gain current > amplifier may not be what you want to do. > > Of course, three lines of algebra would have told you what the results > were going to be, and you'd have learned something instead of just > throwing spitballs. This theory stuff really works. ;) > > We've had several threads about capacitance multipliers recently, > including one where I posted an LTspice schematic of a 2-stage cap > multiplier that will knock that stuff down by about 180 dB, assuming > that you have adequate RF shielding.He could start with a simple RC filter on the bias supply... that might be enough to show him where the problem is. (My hint is to look at the noise on the bias voltage and see if it looks the same as your signal noise.) It is weird how this started as a thread to find an opamp. To the OP, do you have any thermal regulation on the APD? I've read that the gain is a strong function of the bias voltage and temperature. George H.> > Cheers > > Phil Hobbs > > -- > Dr Philip C D Hobbs > Principal Consultant > ElectroOptical Innovations LLC > Optics, Electro-optics, Photonics, Analog Electronics > > 160 North State Road #203 > Briarcliff Manor NY 10510 > > hobbs at electrooptical dot net > http://electrooptical.net
Reply by ●August 4, 20162016-08-04
On 08/04/2016 09:44 AM, George Herold wrote:> On Thursday, August 4, 2016 at 8:16:51 AM UTC-4, Phil Hobbs wrote: >> On 08/04/2016 02:31 AM, T Obulesu wrote: >>> On Tuesday, 2 August 2016 08:58:42 UTC+5:30, T Obulesu wrote: >>>> Hello guys.. >>>> >>>> I want to design a trans-impedance amplifier to detect currents of >>>> the APD (in the range of 200 nA to 100 micro A). >>>> >>>> Could some one suggest me the best IC with less bias current, >>>> single supply (+5 V only) and how to design a high sensitive TIA..? >>>> (typically, for every 100 nA change in the current, 100 mV voltage >>>> should be displayed). >>> >>> Let me say why am looking for new design.. We are currently using the >>> receiver module which has three stages: 1. Trans Impedance stage 2. >>> High Pass filter (HPF) stage 3. Unity gain amplifier (just for >>> inverting the output of the HPF) >>> >>> Well...we used LM2662 as a -5V supply.. Here on we could see the hell >>> noise below 100kHz.. I couldn't get rid of this noise by using bypass >>> caps..but I could just reduced it.. Yet there is a lot of noise all >>> across the circuit ranging from few kHz to hundreds of MHz... >> > >> > https://drive.google.com/open?id=0B_khpIBuCMVdb18ySmZubGsyRUU >> > >> > https://drive.google.com/open?id=0B_khpIBuCMVdbUVRWW85MHFmSXM >> > >> > https://drive.google.com/drive/u/0/folders/0B_khpIBuCMVdMmplSUc3U2FNdjA >> > >> > https://drive.google.com/open?id=0B_khpIBuCMVdeVMxZkRGR1NvM2M >> > >> > are the links where I have uploaded couple of documents.. >> > >> > I really don't know what sort of noise it is and from where it is >> > coming... >> > >> >> Well, if you're connecting a photodiode directly to a SMPS, what did you >> expect? >> >> Hint: Taking a spiky supply rail, differentiating it with a small >> capacitor, and sticking the resulting current into a high-gain current >> amplifier may not be what you want to do. >> >> Of course, three lines of algebra would have told you what the results >> were going to be, and you'd have learned something instead of just >> throwing spitballs. This theory stuff really works. ;) >> >> We've had several threads about capacitance multipliers recently, >> including one where I posted an LTspice schematic of a 2-stage cap >> multiplier that will knock that stuff down by about 180 dB, assuming >> that you have adequate RF shielding. > > He could start with a simple RC filter on the bias supply... that might be enough > to show him where the problem is. (My hint is to look at the noise on the > bias voltage and see if it looks the same as your signal noise.) > > It is weird how this started as a thread to find an opamp.The OP doesn't know what he's doing, and wants a quick fix. There is one, of course--your RC or my cap multiplier--but understanding it does sort of require doing those three lines of algebra to see just how vulnerable a TIA is to bias supply noise. It never ceases to amaze me how many people chicken out of doing that, even when their livelihood (or their thesis) is on the line. C'mon, T. Obulesu, you can do it. All you need is the maximum slope of the ripple waveform, the diode capacitance, and I=C dV/dt.> > To the OP, do you have any thermal regulation on the APD? > I've read that the gain is a strong function of the bias voltage and > temperature.Yup. It's really easy to turn them to lava. Cheers Phil Hobbs -- Dr Philip C D Hobbs Principal Consultant ElectroOptical Innovations LLC Optics, Electro-optics, Photonics, Analog Electronics 160 North State Road #203 Briarcliff Manor NY 10510 hobbs at electrooptical dot net http://electrooptical.net
