EEVblog

Another mystery item from the storage bunker.
Forum: www.eevblog.co...
Subscribe on Library: lbry.tv/@eevbl...
EEVblog Web Site: www.eevblog.com
The 2nd EEVblog Channel: / eevblog2
EEVdiscover: / eevdiscover
Support the EEVblog through Patreon! / eevblog
AliExpress Affiliate: s.click.aliexpr...
Buy anything through that link and Dave gets a commission at no cost to you.
Donate With Bitcoin & Other Crypto Currencies!
www.eevblog.co...
T-Shirts: teespring.com/s...

[@SeanBZA]

The big transformer is needed for the peltier element heaters, so likely there is a model which not only can heat the sample, but also cool it. You need a beefy power supply to enable fast heating, as you might have a sample barely defrosted, and need to get it to the reaction temperature for whatever reagents you are using, typically anything from 30C to 70C, plus you will have a cleaning cycle, where the heaters will need to get the cleaning solution up to 90C to properly scrub the cell windows clean. Same for the pump, massive with the need to pump the cleaning solution through turbulently for cleaning scrub action, but then has to flow through the analysis solution slowly. Quick test to see how narrow the filters are is to shine light through the 2 close ones, you will probably find nothing out the other side. Pump uses the motor, controlling motor speed and using the microswitches as limit sensors, so it runs one way till it hits an end, then runs the other way to the other switch, so you can have a calibrated volume of fluid per stroke, and also adjust motor speed to do flow control from fast to slow. Valve allows the sample chamber to be isolated, so the volume reacting will not move. Adjustable filters probably there so you can choose a band for the photocell to get the fluorescing of the sample, as these typically use a reagent that either absorbs light depending on the reaction, or which emit light when excited, so that you need to remove the excitation light to get a result that is the emitted light from the sample only. Thus 2 ranges, depending on the selected excitation, so the filter response does not allow light through giving an offset. Going to guess the entire sample volume is made from ground quartz glass plate and tube, that has been sintered together, and the sample tubes are PTFE that has been press fitted into ground openings, and backfilled with an epoxy for rigidity.

[@PedroDaGr8]

Looking at the design, this is very much a colorimetric device. First off, flourescent detectors are almost always mounted at a 90 degree angle to the excitation source. This perpendicular arrangement serves to minimize the amount of excitation light which reaches the detector. As no filter is perfect, a linear setup will always has some excitation passthrough. Additionally, as flouescence "always" absorbs at a shorter wavelength than it emits, the second filter doesn't make sense as an emission filter. At least not for any dyes other than ones which absorb below 500nm and emit above 500nm. That's pretty much flouescein-type and Alexa 488-type dyes. In which case, the other filters are essentially useless. That secondary filter is likely designed to handle the wavelengths where the dichroic mirror filters do not behave as a filter. Just search for 533nm FWHM filter on Google. You will see that filter has a strong peak at 380ish nm. This second filter would handle that. Also, these are typically for measuring blood enzyme levels. Due to the highly temperature dependent nature of the enzyme kinetics, they are going to need to have very strict controls around temperature. As even small amounts of temperature change can significantly throw off the results for biological enzymes.

[@spykillergames8402]

id be taking that beefy power transformer for audio purposes

[@playaspec]

Came to comment on the pelier too. I'm surprised he missed those as the heater.

[@6or7breadsticks]

@@spykillergames8402 Id just use it for a usb power boost

[@pietpaaltjes7419]

@@PedroDaGr8 That's correct about blood analyses. That was the core business of this company at the time I had my internship there.

[@PedroDaGr8]

These instruments were used for running assays to measure enzyme levels in the blood. The prism method you mention for selecting wavelengths is called a monochrometer. On that note, monochromters no longer use a prisms. Almost all of the monochrometers I have seen in recent years use diffraction gratings as they are smaller in size, cheaper to make, and have more consistent output. Filters are almost never used for spectrophotometers anymore either. You pretty much only see them in the cheapest of cheap models. That being said, filters are still heavily used in fields like flow cytometery and flourescent microscopy. The filters are normally dichroic mirrors, bandwidth is often +/- 5nm or larger. The filter material isn't too expensive US$200ish (though for a medical device it will be much pricier).

[@notstarman]

Filters still very common in this type of equipment since they are easier to validate than a monochromator since you can use NIST traceable glass in their construction. This is why they are used so commonly in validated hardware in regulated environments like Medical, GLP, and GMP facilities. Filters are used in fluorescent imaging applications since these systems are generally exposure limited. Filters can be designed with wider bandwidths than monochrometers and have a better transmission rate within the selected range. This means they can handle both brighter white light sources and dimmer EM signals than monochrometers can. Monochrometers also have limitations that filters do not for example monochrometers always allow a little straylight through which is why they are almost always in filter/monochrometer or monochrometer/monochrometer arrangements. It's really about designing the right light conditioning method for the measurement type, the application, and the price of the device.

[@norullzz]

The filters look like thin film filters en.wikipedia.org/wiki/Thin-film_optics

[@briangoldberg4439]

I used other brand spectrum analyzers in the chemical engineering industry. We used them to measure grain size for materials after doing batch particle filtering. Are you saying this particular model is only for blood work?

[@st_us]

German Translation here: ELKO is not the capacitor Brand, it stands for ( Elektrolyt Kondensator) which means electrolytic capacitor. The Brand is ROE, Roederstein capacitors.

[@onlyrgu]

Yes i seen them before!!

[@max_kl]

And the "rauh" after "Elko" means "rough". Does anybody have an idea what a rough capacitor is?

[@Troppa17]

@@max_kl The english equivalent is 'raw'. This refers to the aluminium foil inside the capacitor. A raw foil has a lager surface than the plain foil. In an nutshell this makes either smaller capacitors or capacitors of the same size but with higher capacitance rating possible at slightly worst electric properties. Nowadays almost all aluminum electrolytic capacitors are raw capacitors. Plain aluminum electrolytic capacitors are still used in hi-fi and scientific equipment.

[@max_kl]

@@Troppa17 Why would a larger surface area increase the voltage rating? I thought that the voltage rating depends on the thickness of the electrolyte and its breakdown voltage?

[@Troppa17]

@@max_kl Oh, I'm sorry I messed up. I actually meant higher capacitance not voltage rating. (I will correct that in the first comment.) But the dielectric strengh of the raw aluminium foil is higher so it could withstand a higher voltage at a high frequences (800 kHz and up if I remember correctly) compared to a plain aluminum foil capacitor.

[@electronicsNmore]

Great teardown Dave! Love 80's electronic equipment. :-) 14:48 board discoloration is from the heat of the halogen lamp, you can see it at 13:50 exposing the exact area. The other areas of the board were probably protected by mounting brackets, etc.

[@davidwillmore]

Heat or light output. If it is putting out reasonable amounts of UV for years, that woul do the job as well.

[@TheEPROM9]

The proccessor board is also a keeper with all those lovely vintage chips on in.

[@mrnmrn1]

I would say that the whole thing should be reassembled and preserved. Maybe replace the AMD 8080 with a more common one, and sell this one on ebay. Then repair the instrument and sell it on ebay, too. Probably it just has a shorted tantalum, so every single tantalum should be replaced in it. You can see at 8:34, the last in the row is sooty. The EPROMs also need to be re-programmed to prevent loss of the firmware. After 30 years they can develop amnesia. I bet this one with all the filters would sell for a pretty penny on ebay if repaired. Or it could be donated to a not too wealthy school.

[@electronash]

TheEPROM9 - I see you still have good taste in vintage chips, mate. ;)

[@tad2021]

The dark region on the PCB is likely from the light source. That would explain the unsharp shadow lines, it's from literal shadows.

[@redsquirrelftw]

Was thinking that too, maybe it's actually UV discolouration. Would be interesting to see if it lines up with any shadows.

[@maxtorque2277]

Could also be from the heat of the large transfomer too!

[@VicHayden]

Might be connected to the amount of heat those halogen bulbs release, it is concentrated on the bulb side of the board. I have seen similar discoloration in lab equipment in the past.

[@TzOk]

I think that too... such halogen bulb could be easily used for soldering, but I don't think it operates there at full power.

[@imark7777777]

I comment the same thing knowing somebody else probably mentioned it.

[@ManuelHefti]

I worked in the lab for many years. What you basically can do is to measure concentrations of specific substances in liquids from the UV into the visual (sometimes also near infrared) range of light. Let's say your substance is red, it will block light from the blue and green spectrum range. The more concentrated your solution is, the less light of this range will pass. You'll need to do a calibration curve with different concentrated solutions in prior and you measure your sample then. Like this you can compare the transmission of the unknown sample against your calibration curve of the known samples and get the concentration of the unknown sample.

[@matmay]

Exactly

[@redtails]

Dave I really like this teardown. I work in biomedical sciences and this technology is something that I've used before. I'm puzzled by the 'flow' nature of this machine. Normally sample is loaded into a cuvette (a square glass/plastic/quartz container with a known lightpath and volume like 0.1ml) or a flat-bottom plate (like a 96 well plate). Light goes through it and transmission is measured, from which absorbance is calculated and expressed in log space. The use of plates may be too recent for this machine's age. Dyes specific to a compound are used on samples, which have a linear response to the absolute concentration of that compound, to be able to calculate the concentration from a standard curve. Though the 'flow' nature of this machine may be an indication of its age, as that would definitely be too time-consuming nowadays seeing the "2 minute" rinse cycle after each sample, jeez. A modern absorption/transmission UV/VIS plate reader can measure a 96-well plate (which contains the standard curve, samples with known concentrations, negative controls, and around ~30-40 biological samples in duplicates) in under 2 minutes, which would be 100-fold faster than this machine.

[@omniryx1]

Dave: "This is what I expected." Translation: "I have no clue."

[@wilcostienezen9403]

Hi Dave, yes I know vitatron. It was a company 20 km from my home. There where 2 company's. One who is made pacemakers the other one manufacturer and distributor of clinical laboratory instrumentation. I worked there for 6 month.

[@terminsane]

the Applied Science channel guy recently did a video on creating these filters. its worth checking out

[@kris2501100]

This is Vitatron vitalab 21. Manual is available from here. However you need to fill a form. www.labexchange.com/manuals/vitatron/vitalab-21

[@vladsinger]

Renamed in 1987 according to this pubs.rsc.org/en/content/articlelanding/1987/AP/ap9872400282#!divAbstract The makers of Vitatron instruments announce the re-naming of their range of small scale photometers and clinical analysers. The Vitatron MCP, Vitatron IFPM and the Vitatron ISP-M become the Vitalab 10, Vitalab 20 and Vitalab 21, respectively. The Vitatron FPS-A range of photometers is re-named the Vitalab 30 series and consist? of the Vitalab 30 with plug-in filters, the Vitalab 31 with automatic filter wheel and the Vitalab 32 with monochromator. All of the Vitalab 30 series can be converted into the Vitalab 100 series of batch analysers by the addition of a sample turntable and electronic diluter. Vital Scientific Ltd., Huffwood Trading Estate, Partridge Green, Sussex RH13 8AU.

[@Ispike73]

I work in a bioresearch facility repairing instruments, I've worked on a couple spectrophotometers. Be careful playing with old lab equipment, some of this stuff is very difficult to fully decon. There should be a biohazard sticker on this...

[@harrysmbdgs]

Very good point, God only knows what might have been through those tubes! 🤷🏻‍♂️

[@vidasvv]

TOO LATE LOL

[@SixTough]

Decon from what?

[@Ispike73]

@@SixTough Pathogens, not just from the sample but from general use and handling in laboratory conditions. I've opened enough equipment up for service to find contamination on the inside even after being 'decontaminated' that I treat it all as still being a biohazard, even if it's been VHP'd.

[@SixTough]

@@Ispike73 have you actually done cell cultures from this?

[@Ellipsis780]

Those filters are definitely a keeper. A set of matched ones no less.

[@Slide100]

Hmmmm. “Tron” in their name, and a model called “MCP”? Coincidence? 😀 Great tear down Dave. Thanks!

[@jaycee1980]

and these devices are now, of course, END OF LINE :)

[@papaalphaoscar5537]

I worked with a more primitive single wavelength model for colorimetric assays. We used to put the sample manually inside a "tes-tuby thing" thing called a cuvette. Those were made of quartz way back then and we were constantly reminded that one would cost about half a month's salary. :-D

[@gleggett3817]

For colorimetric in my work fortunately the absorbance was at a wavelength where plastic was fine. Used to go through boxes of plastic cuvettes each of which was made of expanded polystyrene. Between that and disposable pipette tips i generated a lot of plastic waste in that job....

[@maxtorque2277]

You can so tell that Dave is not a mech eng watching this ;-) 1) Worm and wheels really only transfer torque in one direction, because there large reduction ratio and effective tooth angle prevent back driving 2) The micro switches and the shaft on the valve assy clearly read the VERTICAL position of the valve! ie the shaft moves up and down and not round and round!

[@joefowble]

You can also tell he didn't take chemistry lab courses very far.

[@swp466]

14:46 -- The PCB is maybe discolored from exposure to the halogen bulb?

[@arthurnonimus]

Don't have a manual but these instruments usually use a reagent that is added to the sample which makes the analyte under test absorb the specific wavelength output as selected by the monochromator or filter. It's a UV-Vis/IR combined device and I believe was an instrument used for body fluid analysis. The functions on the front Factor, Standard, Reagent, Sample, and Blank are used for setting the parameters within the software - for example 20 times dilution factor, reagent 3 for analysis of blood enzymes (don't know the actual reagents), a set of known concentration standards and a blank for instrument calibration, and then the sample. The other functions are likely for titrimetric analysis as well as noisy/difficult to read samples.

[@jamesbrowne1004]

As you were posting this, I am working on a circuit board for a simple single wavelength version of one. No need to use our expense bench version for a one wavelength reading in a wet messy algae growing operation. Could use a couple of those filters for my next project, a more sophisticated but hardened field instrument. BTW the blank setting is to get a base reading for calculating absorbance.

[@protonjinx]

First you run a 'sample' of distilled water through to get a baseline of the lamp/filter performance which creates the "blank" dataset. Then you run through the sample and the difference between the "blank" dataset and the "sample" dataset is the actual measurement you want. In the 90s I helped my uncle (chemistry teacher) to control a spectrophotometer via rs232 by means of some borland pascal for windows, to turn a manual single-value device into a scanning (sweeping the whole wavelength range) device. I was young and didnt really know what I was doing, but we eventually got it working fine.

[@chemputer]

Chemist here, just by the look of the thing at 0:44 I'm fairly sure it's a UV/Vis spectrophotometer. (Giveaway is the UV range 330-500nm and visible to near IR range 500-1000nm, as well as standard and reagent. Definitely specialized. I could see this constantly measuring what comes off a HPLC or something.

[@eismeister]

Elko (Elektrolyt Kondensator) in German is short for electrolytic capacitor

[@maxtorque2277]

Sample vial Heater block looks like dual peltier assemblies to me!

[@andymouse]

Yep

[@pietpaaltjes7419]

That is correct.

[@spgoo1]

Surely a photometer, definitely not a spectrometer, technically just a colorimeter with manual filters!

[@pugnate666]

you want the vial to be small, to reduce heating time and allow small sample sizes (getting a bigger sample can be hard). thanks for the teardown! interesting as always ^^

[@largepimping]

After reading some of the comments, all I can think about is how Elizabeth Holmes could probably create a better version of this same device.

[@michaelfuchs1467]

🤣🤣🤣

[@mrnmrn1]

8:34 The last tantalum in the row has blown up :-) . These green tantalums are made by Roederstein, Germany. Just as bad as any other dry tantalum cap. I used some NOS ones from the late '80s for prototyping about 5 years ago. 1 out of 2 shorted within a few hours, and it was a 35V one on a regulated 15V rail...

[@SilverGreen93]

Those green tantalum caps were all over the place in Romania. Very familiar, as they were locally produced.

[@tiavor]

Elko is just a short for Elektrolytkondensator

[@paulcohen1555]

And they had EXCELLENT capacitors.

[@user-rs4iy4nh9z]

I worked at Vitatron for 15 years. This was still real manual work. It could be that I was one of the technicians who made one of the circuit boards for this device. Printed circuit boards were sandblasted and dolder paste was applied by hand. The parts were placed on by hand and soldered using a microscope. The entire device was assembled by hand (both software and hardware), which is why production numbers were low. The best work I've ever done, actually. With a great employer until it was taken over by M.

[@electronash]

I love teardowns like this. It's the sort of stuff you don't normally get to see. 16:53 - Looks like the shaft moves up and down as the motor "winds the wick", so it will operate either switch at the limit.

[@JerryBiehler]

The filters are dichroic, layers of material with different refractive indexes stacked and depending on the thicknesses and how many layers you can tune to filter whatever you want. Pretty cheap, the manufacturers cost probably 10-20 a piece and then they put in in a housing and sell it for a lot more. Still way cheaper than putting in a grating and the mechanism to control it, and you don't need calibration with filters.

[@colinswift2656]

Multilayer dielectric stack filters. Quite complicated to describe but layers of different refractive index half a wavelength thick....not overly expensive....

[@CoolMusicToMyEars]

Hi David, I have something similar, & it has a heater & optical sensor in a chamber lots of pumps in my unit Analox instruments P-GM7 All blood test instruments work on similar principle Philip from Cheshire 🇬🇧

[@SomnolentFudge]

that AMD chip maybe worth a fair bit to collectors on eBay

[@bubblynubs]

I use a a Hach DR 5000 spectrophotometer in work to find the exact concentration of chlorine solutions to use in calibrating analysers. I'm sure the chemist uses it for a lot more than that though. It uses removable vials instead of piping it directly into the machine. A reagent is added into the chlorine solution which turns it a shade of pink and the machine uses that to detect the exact concentration.

[@pietpaaltjes7419]

When I saw that keyboard it immediately did ring a bel. I have a simmilar one lying around from the time I did an internship at Vitatron I think in about 1990ish. They made verry interesting products with not only the electronics but also the nice mechanics, physics and optics. I had a good time there.

[@deelkar]

the range switch is necessary if the filter inserts used are interference filters, because a 300 nm filter will also pass 150 nm and lower fraction of the stated wavelength, so you need to select the range where it is sensitive with a wider filter.

[@detalite]

20:27 There are two small Peltier modules for heating and probably cooling too.

[@sedsberg77]

We had some equipment at work with similar filters, just mirrors instead of windows. These filters costs thousands and thousands of dollars.

[@TzOk]

I have an old car exhaust gas analyzer, which is very similar in construction. It also uses a bulb as a light source, but this bulb is just barely glowing (it operates in the IR range). My analyzer has 2 sensors, with fixed filters.

[@tvsinesperanto7446]

Board looks just like my Vic-20 when I took it apart at age 10. It worked when I put it back together...eventually.

[@michaelslee4336]

As soon as I saw the the video thumbnail I thought that those look very similar to the colour standard that I just used to check our PFX190 Lovibond tintometer. We use our one to check colour of Chlorinated Paraffin expressed in Hazen units.

[@jimtron66]

Obviously an output port. Nope, obviously some sort of test button. Continues the investigation by pulling and poking random controls. Next week I'll be doing a tear-down on an old WW2 landmine I found in the bunker...

[@rayceeya8659]

The discoloration of the PCB board is probably from exposure to the lamp itself. Those bulbs are very bright and extremely expensive.

[@mikemike7001]

Just reading about the Am9080 the other day. An unlicensed clone of the 8080, reverse-engineered from photographs of an 8080 chip. Wonder if CPU Galaxy has one.

[@paulcohen1555]

Exactly not. Look here: en.wikichip.org/wiki/amd/am9080

[@mikemike7001]

@@paulcohen1555 Yes, I knew but didn't point out that while it was initially unlicensed, Intel and AMD entered a cross-licensing agreement the following year. I didn't know about Wikichips, which looks like a cool resource. Thanks for the link!

[@chuckp3986]

You're giving this thing a bit more credit than it is due. For the photosensor, you don't really need to do much characterization. The calibration is typically going to be done in situ at the start of every day. Generally you'll use Beer's Law (no really) to assume absorption is linearly related to concentration of whatever you're looking for. It is still good stuff. Spectrophotometers haven't changed a whole lot over the years. Light sources are where it's at. The common ones are going to be tungsten lamps, if you're doing UV then deuterium, xenon arc is also fairly common.

[@hempbear]

Nice teardown. Green tagged tantalums - so they burn in nice green color

[@nonsuch]

I have something I got from an Audiologist I'd love to send you but, it would probably cost me over $100 at least to send it. I think it would make a great teardown. Maybe I can get a collection going.

[@giokiborg]

16:00 probably some volume metering device Optical sensors can be calibrated in field as well (at least evaluated using filters) It most likely measures absorption as well, since Absorption = 2 - log10 %Transmittance Pretty strange configuration , usually you want to separate liquid handling parts from electronics, due to probable hazards in case of leak I do not remember seeing such mixture of liquid tubes and electronics in any spectrometer

[@audioradiostereo8214]

Still many failures are caused by liquids reaching electronic parts. Liquids climb thanks to the capillarity and gas flows.

[@kalhana1]

Waiting for the day that Dave gets his hands on the instrument that I helped design in the previous company I worked at. Hopefully he will say his high praises like "what a bobby dazzler!", "a thing of beauty is a joy forever", "some grey bearded.... had to tweak this", "spared no expense". And not things such as "crusty burger!", "dodgy as", "flapping around in the breeze". But considering it has a typical life of around 15-20 years, we will all have grey beards by the time it appears in a dumpster.

[@brucejones2354]

Kalhana, you nailed it ! ! !

[@nickademuss42]

I have worked on a lot of gas analyzers at power plants, almost all of them are set up this way with a light source and sample path going into and out of the equipment. Most were calibrated by injecting a gas that contains none of the gas the analyzer can see. this "zero's" the system, then a low and high-level gas containing the target gas would be injected to give you a coefficient to set in the software. The EPA regulated gasses would be tested on many different scales. The major differences between this one and gases like CO and CO2 is to sample they need a long chamber for the source light to pass into and out of. this is because the target gas is not very dense. This one is very short, so its looking for something dense, like iron, lead, or other metals. Or the concentrations are very high. Fun stuff, wash your hands very well if you handled the sample chamber and notice how the tubing is discolored on the output side of the pump. Also, analyzers like this need a long warm up time to get a stable reading. some several hours.

[@artursmihelsons415]

These caps, resistors and some IC remind me old Blaupunkt car radio.. And yes, all unit was only radio.. 😆

[@OsmosisHD]

Goede morgen Dave. Greets from the Netherlands!

[@bososz]

The sample/ vial in the scheeeematic is external (see the second picture in the schematic is a side view of the front panel) to the unit. What you were looking at was the flowcell, 99% sure anyway.

[@TheEPROM9]

I would love to experiment with these filters in my photography

[@derkeksinator17]

I'd love to drop out the sodium streetlamps for astrophotography, but now they started using LEDs. You can't see shit with all that light!

[@skipfred]

You can easily do literally the exact same thing that these filters would do in Photoshop.

[@derkeksinator17]

@@skipfred Except you can't. If there's too much light from the streetlighting you won't see any stars, even if you remove that particular wavelength in Photoshop.

[@skipfred]

@@derkeksinator17 I was talking to the OP lol. Edit: oh wait I see what you meant now. I was only thinking about filtering for stylistic reasons. Fair point.

[@TheEPROM9]

@@skipfred Photoshops the boring way to do things. I spend way to much time in front of computers.

[@vegapiratradiovpr425]

20:27 this is elements peltier

[@arraybytes]

I would love the eeproms, would love to read the program for that. Would also love the processor.

[@ernestneijenhuis2494]

My first job was at Vitatron as software QA engineer, in the mid 90's. However I never saw this device before. But I do remember they knew how to select the right components. I tested some of their 6502 based pacemakers and their programmers.

[@jensschroder8214]

I think the electronic components are all first class. Every hobbyist would like to have such a circuit board in the box to get parts here and there. Today you can get everything cheaply from China, but what do you do if you need that component quickly?

[@admirerofclassicalelectron2858]

These are cheap interference bandpass filters costing just some hundreds of bucks per filter. They shouldn't be so crusty. Their transmission bandwidth depends on design, but in this case I would assume 10-30 nm with a slope width of 2-3 nm. A wavelength delta of 6 nm is a barely visible color difference, but you should not judge by the reflection.

[@TrevorClarke]

the filters are likely a Fabry-Perot style filter. A stack of λ/4 stacks like in a standard thin film filter but with a cavity between them which is n-λ thick. The light reflects around the cavity and constructively interferes creating a very narrow passband. The thin film layers create a stop band around it for a 100 or so nm. The remaining light passes and is either block by a wide-band filter or ignored because the detector isn't sensitive to it.

[@iamiswitchspace6751]

Really wished you had done a power-up and saw if it still worked before the teardown... :(

[@Wilson84KS]

Dave, please take that thing further apart, I just love this old devices, how they developed devices, on the one side extremely smart and kinda barbaric on the other side at the same time. A while ago me and my dad took apart an old stereosystem, this thing had so many solenoids and motors inside, it was actually a robot, if you love technology, you need to love this old stuff. Man I just love the fact that you still have this "childish" curiosity in your age, never give up on that and infect your kids with it.

[@samsonofdan]

Filters are probably UV fused silica with thin film filters on them. ~20nm bandwidth or so. They're not too expensive. Thorlabs and Newport and Edmund Scientific sell em around $100 or so for 1" diameter

[@MatthewSuffidy]

Those 2 filters probably let almost the same colour of light through, just they have some different surface treatments. Your eyes really see 3 em colour bands, and any other colour than red green or blue is just your mind responding to there being more than one.

[@Darphi01]

I think the filters could be sputter coated silicon. Ben on Applied Science has done a video on similar filters.

[@luipaardprint]

Hi from all your Netherlands viewers!

[@chris746568462]

20:12 Pause. The "heater" looks like its two mini thermoelectric devices on either side of the sample container. The left one looks like it is held in by a grub screw, you can see it has a separate block behind it so the grub screw can clamp the whole lot together.

[@imark7777777]

The discoloration on that circuit board is most likely due to the UV put off by that internal light. Surprised they didn't Shield it.

[@zebo-the-fat]

Many years ago I used a similar device to measure pharaceutical products, paracetamol absorbs UV light (about 243 nM if I remember correctly)

[@paulcohen1555]

So it is good as Sun UV protection, right?

[@zebo-the-fat]

@@paulcohen1555 Probably not, only absorbs at one specific wavelength

[@Knight8365]

Great video & teardown. I'm just wondering why the sensor has a co-ax attached? Is this simply to isolate very small voltage changes, or is there high frequency data involved? Would the ADC be able to handle high frequency data? Cheers Dave have a good one!

[@pietpaaltjes7419]

I guess becourse of the low signals since the source is just a photo diode. All the processing of that signal occurs in the analoge part of the motherboard.

[@Mrflash222006]

A quick dig via google the ISP it is a smart colorimeter aka photometer it passes light through a liquid, used in labs for water testing and in medical fields

[@Akbar_Friendly_in_Cherno]

Looks like the inside of my La Marzocco espresso machine

[@MoraFermi]

IH401A's features: 35 Ohm Rds(On)! Amazing performance! Those are dichroic filters, solid state interference filters: en.wikipedia.org/wiki/Dichroic_filter

[@randomelectronicsanddispla1765]

I wouldn't think the photo sensor is heated, I would think it is cooled, to reduce thermal noise in the signal output

[@PaulSteMarie]

Those look to be dichroic filters: a thin film with surfaces a half-wavelength apart, such that the target wavelength reinforces itself through the reflections, and other wavelengths cancel out. en.wikipedia.org/wiki/Dichroic_filter?wprov=sfla1 for details

[@MC-Racing]

shine some light through those filters for us to see :-)

[@cristianchisbora8289]

Yeeey. A filter photometer for use in the clinical lab!

[@williefleete]

11:51 burnt tantalum cap? near the rear of the machine under a TO-220 device

[@fpgaguy]

I don't want to mention companies by name but just search for "bandpass optical filters", I've run into these before when playing with tunable laser cavities (in my case argon ion) and spectrometry and now lidar - they are also very useful for playing blocking an excitation light source for fluorescence microscopy- and BW can be below 1nm, basically it all depends on thickness of deposited metal and uniformity of the thickness

[@no_you_cant3210]

Can’t find any good info on it but ISP is an acronym for Infrared Spectrophotometer. Found that in an acronym dictionary

[@akhurash]

Curious to know if they needed that large of a transformer, that's huge! Interesting to see an AMD processor.

[@PedroDaGr8]

Heating blocks and cooling peltiers can take a decent amount of power. Add in a 20-60W halogen bulb (the light source) and you can easily have a couple hundred watts of power draw.

[@akhurash]

@@PedroDaGr8 Ah. Didn't pay attention to the halogen bulb. Makes sense.

[@oliverthane2868]

Another way to "do it" might be using a diffraction grating like this www.thorlabs.com/newgrouppage9.cfm?objectgroup_id=9026 ... That's how most optical spectrum analyzers I've seen work

[@albinblad1832]

Is it possible to disassemble the sensor housing :)?

[@jwenting]

gosh, I grew up not 20km from the town this thing was made. Never knew they had high tech medical equipment companies there.

[@Orbis92]

Green Tantalums are normally for military electronics, so they can camouflage on green PCBs ;)

[@frankowalker4662]

Cool bit of kit with some intresting spare parts Great tear-down.

[@emielv7677]

Oh wow. That thing was made 10km from my home!

[@audioradiostereo8214]

16:55 the metallic endless screw rotate the horizontal white plastic gear; coaxial at the plastic gear there is a disc cam or a peristaltic pump or a vertical shaft or something. The switches are the upper and lower position limit switches of the mechanics. 20:33 the smaller the flow cell, the less sample quantity needed. Nobody likes to have liters of blood taken for analysis or bring liters of strange and toxic liquids to the laboratory 2:23 with only one hand it is possible to insert the sample test tube into the little (verticalized) teflon tube and press the switch. It is called sipper.

[@admirerofclassicalelectron2858]

The partial discoloration of the board may be a result of the nearby light source. Probably it is a Xenon flash light creating a nearly homogeneous spectrum between 220 to 1000 nm. The UV part (below 400 nm) of the spectrum can bleach the green dye of the PCB.

[@pietpaaltjes7419]

I think they were pretty regular halogen bulbs. They were tested (caracterised?) In quantities at the Vitatron lab when I did my internship there. They were handy at home :-) l think I still got some.

[@admirerofclassicalelectron2858]

@@pietpaaltjes7419 Good old halogen bulbs? Interesting. The eightys were a different time. Indeed, even halogene bulbs create a small amount of UV-A light. And the duty cycle of an incandescent bulb is much longer.

[@davidchang-yen1256]

Those solenoid Neptune Research valves last forever if you don’t overheat them. Still made under the Cole-Parmer name.

[@MrMaxeemum]

That machine would have been super expensive. The machine itself would have been expensive, way more than the cost of the parts due to low production volumes but the real cost is in the know how & software behind the scenes. It would not surprise me if the cost of the machine would only account for 5-10% of the final cost. The electronics required would cost buttons these days.

[@DevilishDesign]

Tantalum capacitor (C3) near the voltage regs on the main PCB looks like it might have let out the magic smoke :)

[@ShadowPantherRus]

The waste disposal port is right next to the power plug at the back. There's a tube going that way from the pump.

[@sdttnkara]

Give it away to Mike! Mike will explain its theory of the operation better.

[@HugeWolf1]

Not sure if anyone asked this, but can this be for sampling gas instead of liquid?

[@onlyrgu]

Its band-pass filter right?? Its funny whole PCB can be replaced with a Arduino with relays!!and maybe some precision ADCs

[@cfb33774]

I wish you had turned it on before taking it apart!