Marco Reps - 2021-07-09
Flex or FR4, get your advanced PCBs from https://www.pcbway.com/setinvite.aspx?inviteid=65003 5% off Brymen & Siglent w/ coupon 'marcoreps5' https://www.welectron.com/ Miniware LCR tweezers https://www.banggood.com/custlink/vDvho1IbtN Thermal imager https://www.banggood.com/custlink/KvmyouSNJT Stereo microscope https://www.banggood.com/custlink/KmGRl79QFJ Exclusive videos sometimes on https://www.patreon.com/marcoreps IanScottJohnston transformer anatomy https://www.youtube.com/watch?v=BgYZnbLIwc8 Massively simplified 732A Falstad sim https://tinyurl.com/yzcvd2pn LTZ1000A Falstad sim https://tinyurl.com/yj8tvqjo Has cool stuff other than feedthroughs https://twitter.com/GigaBecquerel Alternative hermetic feedthrough pins: Tusonix / CTS 4300 Mu metal cans https://de.rs-online.com/web/p/audio-transformatoren/6675962/ 0.2 ppm/°C VSMP resistors https://de.rs-online.com/web/c/passive-bauelemente/festwiderstande/smd-widerstande/?applied-dimensions=4294097052 R-Core transformer https://s.click.aliexpress.com/e/_Aq6SMY Wavetek 7000 EEVblog forum https://www.eevblog.com/forum/metrology/(3300)-wavetek-7000-the-hidden-gemstone/ Schroff RatiopacPRO AIR https://schroff.nvent.com/en-de/products/enc24571-911 LTZ1000 tests with various 'above' resistors https://snapshot.raintank.io/dashboard/snapshot/y6f9BqnEAGTXXd3IyctaI4C8PiKgosUk A few days in the life of 732A https://snapshot.raintank.io/dashboard/snapshot/FdXwDqNPFZMvbtzDFIvHqIOubY1S2T4c Fluke 5700A cal check logs https://github.com/marcoreps/5700A/tree/master/logs LTZmu stats w/ imbalanced batteries https://snapshot.raintank.io/dashboard/snapshot/DNbQoTWw6M4Rm1AbSv6CCZc9SKTMOZPW LTZmu project WIP https://github.com/marcoreps/LTZmu 00:00 Intro 02:01 Fluke 732A 02:55 732A battery 04:35 732A guarding 06:00 732A pre-regulator 07:32 732A ref assembly 09:11 732A thermals 10:42 732A 'fixes' 12:06 732A heater circuit 12:42 732A ref circuit 14:31 LTZmu 10V standard 15:33 LTZmu-metal cans 16:08 LTZmu SMD resistors 17:00 LTZmu flex PCBs 18:49 LTZmu assembly 20:23 LTZmu concepts 23:20 LTZmu surroundings 26:46 LTZmu circuit 27:48 Wavetek 7000 32:16 732A results 34:07 LTZmu no results yet
To calibrate a voltage reference you need two additional voltage references that you rub together in an alternating pattern until they are all with in spec.
im pretty sure i can make that circuit bend a little worse if you want me too hahaha. :D.
Awesome to see you here! I have been watching reps and look mum no computer for years now! The reference made me smile. Hope to see you here again!
They should make VFD displays that have 16-segment Pagan runic font... Just for you, my friend.
The "look mum no computer" reference at 10:50 got me good 😁
@Douglas Fish Ha! We noticed that too!
Still trying to recover from that 😂🤣
Yesss
My gf..wtf you laughing at? Me..can't explain
hahaha yeah that was funny. mainly because its very accurate :D
As always, a stellar job, Marco! More to come pls! 🎉
By far the most interesting YouTube channel regarding electronics, with a great sense of humour! I always learn something new and actually you gave me a wonderful idea for my next project with the flexible PCBs! Keep it up Marco!! 😉
It's amazing that you put so much work into these videos while consistently demonstrating your sense of humor. This is some of the best content on Youtube.
I fucking love your content
you can make a 365 hour video and I would still see every second of it
It would be interesting to attach a high precision temperature probe to each resistor, run a simulation falstad-style over the spec temperature dependence, and use that to correct the readings digitally. Maybe a company could even sell precision resistors with integrated thermocouples?
my dear god the 9v and the 1.5v battery in series killed me
@Power Max maybe Ikea batteries built different because I directly measured with an ammeter.
@TissuePaper you probably need to learn how to read your multimeter then. 😉
If you have a fancy lithium battery, those do have lower ESR. I can look no further then the datasheet to show the battery can't deliver more than a couple amps shirt circuit. Just look for the ESR rating and do ohms law.
@Power Max I also measured the B-field around the coil I was driving with the 9V and it was consistent with a 7-10A current.
@TissuePaper So I found duracell datasheet for their MN1604 battery, same industry standard 6LR61 alkaline 9V battery.
It provides constant current discharge curves and shows the voltage droop to nearly half with only 500mA draw. 9V batteries are no good for currents higher than this. Best suited for current in the range of single to double digit mA's.
The datasheet also provides a figure of 1700 mOhms or 1.7 ohms of impedance measured at 1kHz. A fully charged 9V cell will start out at 10V, 10V/1.7Ohms = 5.9A. that is a peak short circuit current. With zero volts at the terminals the battery does almost no useful work. (No power is delivered, as most of it is burned in the internal resistance.)
10A seems very optimistic / unlikely, but I guess it isn't as infeasible I had thought. Perhaps you had a rechargable cell? NiMh has a lower ESR than primary alkaline cells. Or one of the newer Lithium chemestry unit? IDK.
@Power Max I think the error bars on the ESR are pretty large. I could have just gotten a -3σ or -4σ fluke that had a freakishly low ESR.
Speaking of low-capacitance galvanic isolation (like the batteries you propose) - how much power does the circuitry consume, on average? I may have a very nice work-around for <1W systems :)
at the moment I am trying to lower the power consumption as much as possible to reduce self heating. <1W might be possible! what's your idea? LEDs and solar cell? :)
@Marco Reps A bit of both. Blue LEDs ~420nm act as excellent photo-collectors provided they are excited by wavelength matched violet LEDs or lasers (405nm). It is possible to get a photonic efficiency of up to 40% (!) and end-to-end efficiency (including the photonic efficiency of the LED of ~ 10%. In other words, you have a violet (visible, safe) "light source" consuming 10W of electrical input at one end and a Blue LED array outputting (ideally) 1W of electrical power (and unfortunately 1.5W of radiation-heat, ideally) at the far-end. https://www.researchgate.net/publication/273416580_All-optical_power_and_data_transfer_in_catheters_using_an_efficient_LED
@आदित्यAditya मेहेंदळेMehendale #BringBackDislikes Wow, fantastic info! Thank you for sharing. I vaguely remember seeing 20% being possible with optimal LEDs and modern solar panel (this might just be a theoretical value, not sure if anybody tried) but your method would be a lot cheaper and more compact. Love it!
@Marco RepsA few nice things about the wavelengths:
- The "royal blue" LEDs have an MPPT-voltage ~3V that is enough to power most MCUs without needing a dc/dc converter
- As the photonic efficiency is 40%, you "only" need to dump 60% of the power as heat at the receiving end (so 1.5W/W instead of 4W/W as in a solar panel)
- Not only can the violet power transmitter be modulated to send data forward, but also the blue "receiver" can be load-modulated to send data backwards (as demo'd in the video); the paper reports a few100kbps to be possible in the backwards direction without separate optics
- veri smol (1mmx1mm) Tx and Rx for ~ 100mW (excluding space needed for collimators/fibers/etc)
- 395-405nm violet lasers (no collimators needed!) are cheap and easy to find from blu-ray stuff
- 430-450nm blue InGaN LEDs are commonly used in remote-phosphor lamps
- Photo electro luminescence is frickin' cool (more info on this in the paper)
It basically scratches every itch I had with galvanic isolation for oscilloscope-probes, bio-measurements, high-side gate-drivers, HV-stuff, contactless stuff, and more... Only 'issue' is the required line-of-sight (or fiber).
Yes, the 3458A self cal would be an interesting topic for a video! I also noticed these (very small) jumps and it would be interesting to learn more about them :)
Memories!! I used to sell Fluke test equipment... and I very much remember the launch of the Fluke 5700 calibrator. That was 1988... man! How time has flown! I remember selling a 5700 to Siemens after I went to their office in Bayswater, Victoria on a "fishing trip" with a Fluke and Philips catalog prospecting for names and opportunities and six months later got this massive order for Fluke Calibration equipment.....
Your genius videos teach me a lot about electronics and humility ! As both qualified master electronic technician and master electrician your knowledge exceeds mine by far.
Very interesting! My vote for the next project is resistance standards :-)
(How good are those Leeds & Northrop standard resistors you can find on eBay? I guess a major challenge with them would be just determining the actual value as-received…)
Finally, my favourite volt-nut is back.
I wish I understood more of this but it's so interesting to see such high precision engineering
I love the aesthetics of those old resistors, they remind me of the ones in my old guitar amps.
They're just retro and funky and cool.
Using ceramic insulation would give you more stable temps because of the low thermal transfer. You can use brazing to hermetically seal those feet and it will draw solder in. Just use plenty of solder paste around the side of the feet and stick in the hole. Heat the housing and paste and as it cools it will draw in the solder as it hardens.
I just bodge up microcontroller circuits on stripboard.... so it's such a delight to watch an artist at work like this.
Great. Such a niech subject but so well thought through and executed and video produced perfectly. Thank you, well done.
As someone who doesn't care about any number after the dot in his £10 Chinese voltmeter, I applaud your dedication to getting the 10th decimal of your instruments right
Just an amazing video!!!! I can’t wait to make my own hermetic reference. Thank you so much Marco.
Sometime in the 1960's (I believe) Philips published a design brief about an ovenized reference voltage source that was built around two zener diodes. Or actually one avalance type (positive tempco) and one zener type (negative tempco). Additionally there were several resistors. In short their idea was based on utilizing an avalance type (higher voltage) diode as the pre-regulator and the lower voltage (zener) as the output stage. Besides selected nominal voltages, they trimmed the resistors to provide an optimal current and thereby a good compesation. I myself remember building a 10 V reference from a 1N825 temperature compensated zener that to my understanding consisted of one zener and two silicon diodes in in series inside an ordinary (DO-35?) package plus my added op-amp. I also at first made a mistake that prevented a start-up. Needed an additional bias resistor for correcting that.
I particularly liked the part where you mentioned ppms.... 🙂
And its always nice to de-stress after a particularly challenging project by playing with some playdough...
The "re-cycling" watersplash & mention of eels was nice.... I'm assuming they were electric eels ? 🤔
From the Emerald Isle
😎👍☘️🍺
I know Marco, your knowledge about electronics scares me, sometimes.
"look mum no computer" was very funny tho.
Very nice video.
Excellent. I enjoy this (these) projects. I vote for a repair video, but I enjoy all them. ☺️
That capacitor tweezer is cool :D
They're called the Miniware DT71. They look great but they have a couple design oofs
Really have enjoyed the deep-dives on the voltage references.
Great work on the flex-PCB.
I’d like to see a resistance standard.
Looks like you were having a problem with the ‘Uden-Pucky’ at the end there! 😅🤣😅
Those censored holes, really made my day... Love the gems you put in your videos. All the info is serious, also love it.
I work at an spectroemter company. We actually use these red heating stripes to heat our optics :)
I actually took the time today to put a Like on every one of your videos, because they are WORTH IT. You put out very very decent videos filled with entertainment and factual, professional work that is both interesting and accurate. Good on ya, mate. Here's a few ppm's for your effort. . . . . . . . . . . .
@19:25 I was wondering if you had that precut metal sheet manufactured or did you use a grbr machine to do it?
Also was interested on the capacitance measuring pincers, they seemed quite handy, but for now a bit too costy for me.
I love high engineering detail video like this. It's good for my engineering mind.
I wonder how the future annual voltage plot will look like, now that you upgraded from coal to metal resistors.
Thank you so much for yet an entertaining and informative video. I love your strive for excellence.
Technically fascinating, hysterically funny. Bravo my friend.
I realize this video is from a year ago, and I have no where near the experience you have.
But in trying to minimize noise, have you considered the mains circuitry? In America, we usually have a step down transformer supplying 2 - 120V ,180° out of phase lines to the fuse panel. This is derived from a single phase high voltage line on the light pole. I would imagine noise between the two phases in the house might cancel if you device transformer utilized a winding connected to each side of the panel. I haven't investigated the situation in Germany.
May I ask what software you use when showing circuit behaviour in your videos? It’s such a great way to visualize what’s happening in the circuit. Danke schoen!
I was just thinking about this channel. Hope the updates are further apart due to you being good busy and not anything else. We all got day jobs and I appreciate the extra effort to post these vids.
Are there any standards built on superconductors? E.g. could we load up a small superconducting torus with a known amount of current, and then measure the magnetic field of it, using this as the reference? As superconductors like that have been left running with the same initial current for 25 years, and after 25 years the amount of current flowing through it had not changed at all, it was still either well below the best equipment that could measure it, or it literally hadn't changed. I've seen papers that suggest the half life of current flowing in a superconductor is on the order of 1e300 years, although obviously the error bars are insane because we just haven't detected it.
A Josephson junction voltage standard is a superconducting device.
Marco, I have zero, I repeat, ZERO idea what's going on here but your content, your knowledge and humor is GREAT! I can only wish I knew half of electronics as you do 🙌
I wonder what happens if you get a stray high-energy particle pass through your zener junction. Could happen once in 30 years?
Perhaps redesign for an SMT only approach (TH part leads bent to solder smt) using thin 2 layer flex PCB on aluminum (attached after soldering) designed to offset and make heat sources consistant per area, a sort of heat gradient proof approach.
Thanks for explaining leakage current across "isolation" transformers.
You know I'm a mechanical engineer so most of this just goes over my head, but it's just so cool to look at. Also, where else will I get my German humor fix?
TPAI (The postapocalyptic inventor) has his moments too...
36 minutes of marco reps and some fabulous tech, nice :D
Thank you for my daily dose of German humour and test instrument repair and rework.. Always good
20:23 Bad idea: having the metal base resting on top of a metal case which in turn is placed on a large aluminium panel... What you want is to heat the metal base evenly and keep it hot so that all the metal legs get soldered simultaneously. Having a large heat conductor in contact with the base is doing the opposite, effectively causing you to apply heat to the base for much longer than necessary.
That "aluminum plate" is probably a reflow plate.
@loligesgame I didn't realize it was - he was using a normal plate in the shot prior to that one so I couldn't tell the difference. But if it is, why is he still using the soldering gun? And not only that, in the next shot he rests the whole metal legs holder on top of metal can used as a prop. Surely he'd be better off leaving it on the reflow plate?
Electrical engineer: The leakage current through these six layers of top grade insulation material is surely negligible.
Metrologist: Can we talk about that for a minute?
The metrologist wins this one hands down.
Klaus Luger - 2021-07-09
I'm sure that jump in voltage in 2020 was just a coincidence, it was otherwise such an excellent year
Yeeter the noob deleter - 2021-07-11
It was probably just a... Fluke 😎
transkryption - 2021-07-12
@Yeeter the noob deleterparasites? flukes? in the liver? ewwww!
Jens Knudsen - 2021-07-14
2020: The Year the Voltage Standards Went Crazy!
john shaw - 2021-08-21
@Jens Knudsen 2020: The year everything went crazy.