In conjunction with our Hi-Fi free-space LASER communicator experiments my buddy designed a photodiode gun that received and amplified as audio whatever light sources it was pointed at. You could hear fluorescent lights flicker at 60 HZ and the pulses of IR remote controls. Fun toy.
I designed a few things like that for hearing research. The slowest was to monitor a strobe light for output stability. The fastest was part of LDV they were using to measure sound pressure levels inside the cochlea.
Excellent video! Maybe instead of switching gain resistors/capacitors, it switches between transimpedance amplifiers? Having a switch in the feedback loop feels somehow dangerous to me (although I´ve seen nice TIAs with this feature). For such huge photodiodes, it also make sense to have a TIA with a bootstrapped input to allow measurements at higher frequencies. Maybe one of those Opamps have it.
the front piece reminds me of the thick fiber bundles we used to use to illuminate objects. it seems to me that this unit could be connected to similar components like the ones we used or it is just what was available in the lab. still enjoying your content! great stuff. thanks for sharing!
That old HP LED Division still lives under Broadcom (Old Avago) today. The Burr Brown device is could be a hybrid pair of JFETs attached to a monolithic op amp with the input stage having a metal mask to bypass its own input stage. Thus lowo noise JFETs with high current could be used. A common application at the time for precision parts.
Well, I would agree that is was probably made under a time-crunch but 4-layer boards are really pretty standard in that they allow you to have a power and ground bus; which again allows you to not have to think about shielding and power distribution/decoupling when you're just using traces for the power and ground. I had to laugh on the DG201; that goes way-back! We switched over to using 4016's and 4066's, not too much later as they were cheaper and as I recall lower on resistance. 73...
@@IMSAIGuy Yeah but if you consider the burden-rate of even a small group of engineers, techs, plus a program manager could easily exceed a $1000/ hour waiting for this thing before they could all be back to full utilization, the cost of that 4-layer board that has the greatest potential to work the first time (involving low-level, noise sensitive signals you don’t want as part of the measurement) without diagnostic-time and rework (which may very well involve another cost-center) is nothing in comparison! Any signal integrity problems,have an immediate vertical cost and are the worst to sort out whether time or money. Also, the hardware may have been charged to another cost-account since not to skew labor costs. I never saw anyone catch-heat for an ‘elaborate’ JIT prototype unless it didn’t work…
@@SpinStar1956 Having worked in an R&D group back in the 1980's designing and building laboratory "black boxes" such as this, I agree with all of the points you raised. I usually had one shot to design and build a 1-off device and it had to work. Parts costs was considered insignificant compared to the labor cost of the group of engineers I worked with. So the choice of parts and the 4-layer board is just the sort of approach I would have used back in the day.
Long time ago I was involved with log amps also for optics application. 10pA is by far not the lowest input bias current for special op-amps. I was using intersil 8007 (I don't remember exactly the parts number) and this one was 0.5pA typical. Among these we found some that measured 0.1pA. Essentially you could express the bias current in thousands of electrons. Anyway your need these to measure very low light intensities as current on a shorted photo diode. This seems your little contraption doing. The switches might be for ranges. As I said we used a bipolar transistor in the feedback which was a different story...
I don't think transimpedance amplifiers are allowed in the state of Florida. I suspect the op-amp with heat sink is driving the cable that comes off the thing. If you are doing a very high gain transimpedance amplifier, you might use two opamps to implement it. This gives you more gainbandwidth to work with assuming you can make it stable.
Interesting video. The heat sink on the OPA101 is important to keep the JFET input stage as cool as possible given it's exponential relationship to temperature. The spec10pA max is only achieved with B Grade part with the heat sink at 25C ambient...
Do you know why they used mica capacitors? Today, mica capacitors are several times more expensive than NP0/C0G ceramics and polypropylene capacitors, and I imagine this was also true in the 80s. I don't see a reason to waste mica capacitors on a low-frequency circuit.
momma mia! what's this Lou-eejy? it's a in house pasta diner! pardon me sir, but it looks like a ''step'' counter to me!! my bruder, he use this when he want to know how many steps. he just made! Momma Mia!! Oh Lala...the 3rd one? it's the Oh God button! quick! some1 knock on da vood! it probably crawled in U'r pocket, like a little hungry kitty!! did U hear that Indiana? Shhhh...low noise!! QUIET! but it's just a little pcb! they no need 2 go to Kansas? just 2 back ally's and the main roads! walah!
In conjunction with our Hi-Fi free-space LASER communicator experiments my buddy designed a photodiode gun that received and amplified as audio whatever light sources it was pointed at. You could hear fluorescent lights flicker at 60 HZ and the pulses of IR remote controls. Fun toy.
I've seen a similar thing using a solar panel and a Bluetooth speaker to check out if you have flickering lights because of not enough smoothing.
I designed several test fixtures into those little Pomona boxes. Very rugged and well made!
a rugged price tag also
I designed a few things like that for hearing research. The slowest was to monitor a strobe light for output stability. The fastest was part of LDV they were using to measure sound pressure levels inside the cochlea.
Excellent video!
Maybe instead of switching gain resistors/capacitors, it switches between transimpedance amplifiers? Having a switch in the feedback loop feels somehow dangerous to me (although I´ve seen nice TIAs with this feature).
For such huge photodiodes, it also make sense to have a TIA with a bootstrapped input to allow measurements at higher frequencies. Maybe one of those Opamps have it.
Fascinating...
Great workmanship too...
Good memories
the front piece reminds me of the thick fiber bundles we used to use to illuminate objects. it seems to me that this unit could be connected to similar components like the ones we used or it is just what was available in the lab.
still enjoying your content! great stuff. thanks for sharing!
Love this! What a bizarre little board!
Red to mid range IR >600nm is ~red then IR 850 and 950nm covert IR etc
good job Sir
That old HP LED Division still lives under Broadcom (Old Avago) today. The Burr Brown device is could be a hybrid pair of JFETs attached to a monolithic op amp with the input stage having a metal mask to bypass its own input stage. Thus lowo noise JFETs with high current could be used. A common application at the time for precision parts.
Well, I would agree that is was probably made under a time-crunch but 4-layer boards are really pretty standard in that they allow you to have a power and ground bus; which again allows you to not have to think about shielding and power distribution/decoupling when you're just using traces for the power and ground.
I had to laugh on the DG201; that goes way-back! We switched over to using 4016's and 4066's, not too much later as they were cheaper and as I recall lower on resistance.
73...
the price of that 4 layer board in the 1990's was way more than today.
@@IMSAIGuy Yeah but if you consider the burden-rate of even a small group of engineers, techs, plus a program manager could easily exceed a $1000/ hour waiting for this thing before they could all be back to full utilization, the cost of that 4-layer board that has the greatest potential to work the first time (involving low-level, noise sensitive signals you don’t want as part of the measurement) without diagnostic-time and rework (which may very well involve another cost-center) is nothing in comparison! Any signal integrity problems,have an immediate vertical cost and are the worst to sort out whether time or money. Also, the hardware may have been charged to another cost-account since not to skew labor costs. I never saw anyone catch-heat for an ‘elaborate’ JIT prototype unless it didn’t work…
@@SpinStar1956 Having worked in an R&D group back in the 1980's designing and building laboratory "black boxes" such as this, I agree with all of the points you raised. I usually had one shot to design and build a 1-off device and it had to work. Parts costs was considered insignificant compared to the labor cost of the group of engineers I worked with. So the choice of parts and the 4-layer board is just the sort of approach I would have used back in the day.
ok, but remember we only got about $12/hr back then. @@SpinStar1956
Long time ago I was involved with log amps also for optics application. 10pA is by far not the lowest input bias current for special op-amps. I was using intersil 8007 (I don't remember exactly the parts number) and this one was 0.5pA typical. Among these we found some that measured 0.1pA. Essentially you could express the bias current in thousands of electrons. Anyway your need these to measure very low light intensities as current on a shorted photo diode. This seems your little contraption doing. The switches might be for ranges. As I said we used a bipolar transistor in the feedback which was a different story...
I don't think transimpedance amplifiers are allowed in the state of Florida.
I suspect the op-amp with heat sink is driving the cable that comes off the thing.
If you are doing a very high gain transimpedance amplifier, you might use two opamps to implement it. This gives you more gainbandwidth to work with assuming you can make it stable.
speed was not an issue, we always used just one opamp for photodiodes
@@IMSAIGuy Maybe it was all about accuracy. You don't want any current in the switches that set the gain.
Mica capacitors, first OP amp must be for power rail
I guess you had services you could call back in the day that would fax you the datasheets?
Interesting video. The heat sink on the OPA101 is important to keep the JFET input stage as cool as possible given it's exponential relationship to temperature. The spec10pA max is only achieved with B Grade part with the heat sink at 25C ambient...
Man, that photodiode is the size of a CCD sensor.
Do you know why they used mica capacitors? Today, mica capacitors are several times more expensive than NP0/C0G ceramics and polypropylene capacitors, and I imagine this was also true in the 80s. I don't see a reason to waste mica capacitors on a low-frequency circuit.
the real answer is: those are the parts that were in the stockroom.
momma mia! what's this Lou-eejy? it's a in house pasta diner! pardon me sir, but it looks like a ''step'' counter to me!! my bruder, he use this when he want to know how many steps.
he just made! Momma Mia!! Oh Lala...the 3rd one? it's the Oh God button! quick! some1 knock on da vood! it probably crawled in U'r pocket, like a little hungry kitty!!
did U hear that Indiana? Shhhh...low noise!! QUIET! but it's just a little pcb! they no need 2 go to Kansas? just 2 back ally's and the main roads! walah!