with the power consumption of less than 2 mA there will be no heater:
so most probably the compensation is similar to the Geller SVR-T with AD587LQ.
(which I think is the better choice for long term stability).
http://www.gellerlabs.com/Voltage_References/SVRTH/SVR%20THR%20Sch.jpg
I bought this eBay unit, and I also bought Doug Malone's Voltage Standard REF-003 as well, to compare against each other. I don't have the Gellar one yet.
I have taken the eBay unit apart, and sketched out the schematic. It is indeed very similar to the Gellar version that is temperature compensated, linked by Andreas. There is no heater.
Like the Gellar unit, it's a REF102 with an NTC thermistor and copper tape for temperature bonding the thermistor to the REF102 body. It differs from the Gellar version in that it has a 1M resistor to the trim pot center wiper from pin 5 of the REF102 (Gellar does not), and in place of Gellar's R01, the eBay version has a fixed R then the NTC thermistor to Vout, and parallel to the thermistor there is a 50k trimmer for trimming the tempco. There is also a reverse voltage protection diode on the negative voltage input. This diode will actually float the reference ground output about 0.6V above the input ground reference. All GND connections are star-connected inside.
It works about as good as you can expect from a REF102. The seller nulls it to close to 5uV of 10V (< 1ppm). It will drift with humidity as said already. According to the the seller it's been burned in 1000 hrs. For the price, I was expecting much better construction inside. It's just a perf-board with point-to-point wiring. I wish it had been shipped with a drop of Loc-tite on the trimmers, but there was none.
photos:
https://carousel.dropbox.com/photos/cc/TzYblf4jDLha7Or#/(In the schematic below, C2 and C3 might be wrong. They were under the IC inside the socket and I couldn't read them. I did not want to de-socket the chip and risk stressing it after it was trimmed.)
If external humidity affects plastic packages, why not encapsulate the package to keep the moisture out? Or enclose the board in a sealed box with some silica gel?
Codeboy2K,
Those are great images that you have. Anybody considering buying the D-105 should certainly have a look at them.
Spoiler alert
I am sure I could solder better than that when I was a young teenager.
The adjustments are some Bourns knock-off potentiometers. They are marked BENTEx I can not read the last letter, it is obscured by a wire. It inspires confidence.
Has anybody considered using a LT6655 in the LS-8 ceramic package? It is claimed to reduce the sensitivity to humidity variations.
I have a Fluke 732A that has been powered up 10 years. I also have a Datron 1281 ( 2x LTZ1000) they agree within 6ppm, but I don't know which one is correct. I did find I had to buy the special low thermal emf cables.
Jay_Diddy_B
Wow, after seeing those pictures, I am going to pass on this one. That is not going to last very long and at this price he should have a proper circuit board. This is just selling prototypes that all will have variable characteristics in the end.
For the price, I was expecting much better construction inside. It's just a perf-board with point-to-point wiring. I wish it had been shipped with a drop of Loc-tite on the trimmers, but there was none.
photos: https://carousel.dropbox.com/photos/cc/TzYblf4jDLha7Or#/
For the price, I can't believe the seller puts that much effort into it. He would have to sell a huge pile of these things to make any business sense at all. Perf board construction is soooooooooo slllloooooowwwwww and prone to mistakes. After materials, construction time, and the eBay effort - why? How big is the market for hobby level voltage references? Anyone that needs the precision would not likely be using an experimental device, they would need a commercial reference.
Moving on.....
Expectations (caused by feelings, not by some scientific theory upfront) met [X]
Hello,
In short: there is really no way out of the maze of humidity.
Putting additional plastics around will at maximum change the time constant.
The LS8-package itself is hermetically tight. So it has no problems with humidity.
But you should never ever solder it on a EPOXY board .....
Try to read between the lines of the datasheet. At least a hidden hint is in the 1236LS8 datasheet.
Its more important to know what they do not write (or omit in later versions) into the datasheet.
With best regards
Andreas
Edit: the hint is also in the 6655 datasheet:
"However, PC board material may absorb water
and apply mechanical stress to the LTC6655LS8. Proper
board materials and layout are essential."
And the main problem of LTC6655LS8 is the large hysteresis
(in any package) compared to other (buried zener) references.
See measurements:
https://www.eevblog.com/forum/projects/general-purpose-%27precision%27-voltage-reference/msg546623/#msg546623
Ovenizing is the way out of the maze of humidity, or at least a way to be minimally affected by it.
And holy shit, those photos are scary as hell. You can buy a working Fluke standard (from the 1960's) for $200. Why would anyone pay this goober $100 for a few components poorly hand soldered together? (Not that everyone has a place for a 3U 19" 60lb piece of gear, but still.)
Sent from my Smartphone
@Terabyte2007,
have you opened your box and can you share what the inside looks like?
I truely hope it is not like in the pictures from codeboy2k.
I would be really shocked if this like out of a garbage can is what is actually sold. But if it is, I wonder how any buyer could accept it and not force the seller to refund all the money. And no way could anyone give positive feedback.
But the seller records show 100% positive feedbacks. Therefore I am looking forward to see more details.
BTW the Geller SVR is/was a great piece of gear, beautifully made and is not using a REF102. Unfortunately Joe decided to go out of this business.
http://www.gellerlabs.com/
I would like to see what's inside Terebyte's unit too.
The eBay seller probably has good feedback because it does work as described, and maybe no one ever wanted to open it up before. Mine also works well, and is as stable as I can measure it under short term 24-hour testing. It's spec'd at 0.002% but it's actually 0.0002% accurate and it's stable beyond my ability to measure it. I can't measure less than 100uV on the 20V range of the meter, and it's not moving at that decade, so if it's moving at all I can't tell. Based on my ability to measure, I can only say it's stable to < 10 ppm but I suspect that it's actually better than that.
GELLER is using the AD587. Too bad he's stopped selling it, I didn't get one.
BTW the Geller SVR is/was a great piece of gear, beautifully made and is not using a REF102. Unfortunately Joe decided to go out of this business.
http://www.gellerlabs.com/
Yes, I missed out. But Joe Geller kindly sold me a couple of blank boards which I've not yet had time to make up.
He may well be still willing to supply these if you contact him.
BTW the Geller SVR is/was a great piece of gear, beautifully made and is not using a REF102. Unfortunately Joe decided to go out of this business.
http://www.gellerlabs.com/
I never understood how he made money from this. Building, burning in and hand calibrating each unit is a lot of work.
BTW the Geller SVR is/was a great piece of gear, beautifully made and is not using a REF102. Unfortunately Joe decided to go out of this business.
http://www.gellerlabs.com/
I never understood how he made money from this. Building, burning in and hand calibrating each unit is a lot of work.
Hello Dave,
as far as I can tell, he probably never planned to make money (for living) with it.
For me he is and was a very generous guy who tried to support hobbyists like me/us, because he shared the same interests.
Because he is also member of your eevblog, maybe he will respond himself.
bye
quarks
The reference ICs are burned in for ~1,000 hours prior to use. But that is to produce good "long-term" stability. The +-2ppm figure is the maximum error. To get that we use temperature-compensation. It took about three years to develop the device. I have a scientific background. So, I only had to learn the application side.
It took him 3 whole years to develop
that?
I wonder how long it would take him to do it, if he didn't have a scientific background.
It's basically just inspiration + education + resource + experience + planning + trial and error + persistence + good equipment + common sense + hard work + prayer = good results.
There is so much innovation in this product that it appears to be magic.
Two red flags right there.
It's basically just inspiration + education + resource + experience + planning + trial and error + persistence + good equipment + common sense + hard work + prayer = good results.
There is so much innovation in this product that it appears to be magic.
Two red flags right there.
See update to my forum signature. XD
Sent from my Smartphone
100$ ... Give it a try.. If you're able to verify it.
Personally, I doubt the performance , once from the capabilities of these refs, 2nd because the description is too enthusiastic, and there's a strange list of compensation features... Like these guys, which claimed supercaps from nanoparticles, and Free Energy, and so on.
You know, they all simply use common terminus technicus / pseudo scientific terms in such descriptions , without really being able to implement these...
At least, if this community would be able to prove this gadget to not fulfill the claimed specs, you may blame him afterwards.
I also offer to check this circuit to ppm uncertainty.
Frank
You've already revealed your bias and your intent. Your evaluation is worth your objectivity. Presumption is the child of arrogance.
I bought this eBay unit, and I also bought Doug Malone's Voltage Standard REF-003 as well, to compare against each other. I don't have the Gellar one yet.
I have taken the eBay unit apart
The pots are friction-fit into the plastic shell. Disassembly causes the pots to move, ruining the calibration. Until the unit is recalibrated, it will probably not meet the spec of 10VDC +-2ppm.
I wish it had been shipped with a drop of Loc-tite on the trimmers, but there was none.
If we used Loctite, the trimmer stems would not fit in the case holes. The unit is not designed for end-user disassembly.
In the schematic below, C2 and C3 might be wrong.
C2 and C3 are each in the 100pf to 2uf range, but if you measure one you've measured them both together. Thanks for the nice expose'.
Hi,
The capacitors are marked 224 which 0.22uF.
Jay_Diddy_B
It took him 3 whole years to develop that?
I wonder how long it would take him to do it, if he didn't have a scientific background.
Let's see you make one.
]prayer = good results.
appears to be magic.
Two red flags right there.
Let's get one thing straight! Criticizing other
people does nothing to advance one's own cause, and it is the bastion of the weak-minded. If you can do better, why don't you? Because you really can't. I say, let's see the results! You might trivialize my efforts, but you trivialize yourself even more, because you
haven't accomplished what I
have. Your publishing private email messages on a public blog, thereby committing violence to the implicit expectation of decency among men, and revealing a mischievous nature. Maybe if you prayed you'd be able to accomplish remarkable things too.
And, to be quite honest, the D-105
does appear to be magic, simply because it is so small and simple, it defies what people currently sense as believable. This thread is
proof of that. Suspicion is the device of a guilty mind!
The pots are friction-fit into the plastic shell.
Are you saying that you've deliberately coupled the mechanical stresses in the housing into the voltage setting? Is that part of the "magic"?
Wow, after seeing those pictures, I am going to pass on this one. That is not going to last very long and at this price he should have a proper circuit board. This is just selling prototypes that all will have variable characteristics in the end.
How truly it is written, "The people shall revile what they do not understand."
And holy shit, those photos are scary as hell. You can buy a working Fluke standard (from the 1960's) for $200.
Those standards are +-10-20ppm. Why pay more for less? Why would anyone pay this goober $100 for a few components poorly hand soldered together?[/quote] How truly it has been said, "Do not throw pearls before swine." and again, "They will see but not understand. They will listen but not hear."
Clearly I possess that which you do not know. Otherwise your mind would be opened, and you would see the truth in front of your eyes . The old must pass away to make room for the new. Everything is not as it appears. Else, my work could be stolen.
I originally posted this in the wrong thread.....
I have a few notes to share.
I don't know anything about Awesome14 and only know a little about the D-105 from what I have read and seen pictures of. I am in the market for a DC reference and looked at this before the conversation started. I am not specifically looking for the bargain basement unit, but rather something that I can trust. Since I would like to have a few references to cross check I was hoping to find units that are affordable. For $100, there is very little risk in trying.
While I have neither the skill or the equipment to confirm the specs, I would have to trust the unit works. Thankfully there is an active group of people that love to test specs and have the means and skills to do so. While I am NOT a scientist, I have worked with many of them. A key part of the scientific community is cross checking claims. Any good scientist should welcome others to cross check for anything out of place and confirm the results are repeatable. I hear a very defensive conversation here where the scrutiny should be welcomed. There is also a "magic" claim that will always have people running away. There are claims that this is the holy grail of voltage references that overcomes the challenges that have vexxed any and all engineers that have attempted to tackle this level of precision and stability.
Awesome14, surely you can expect this community of engineers to pick your claims to the edge of death. The images that were posted show a construction that most of us would have last seen in the 7th grade. I work with far less precision circuits that absolutely require a well considered PCB layout for them to function well and have any sort of consistency from unit to unit.
I do not doubt your sincerity, integrity, or intelligence. I hope that this unit is proven to be what you claim - if so I will buy 4 of them. Let those with the questions and the curiosity verify everything you say. When/if they do maybe you can add that to the eBay listing and raise the price. Independent verification is far more powerful than anything you could possibly type yourself anywhere on the internet.
Everything is not as it appears. Else, my work could be stolen.
If you were at all concerned with the value of your work why are you selling them for $100? The claim of "magic" from someone in the scientific community is over the top. There is more than enough brain power and experience here to analyze the claims and you are talking down to us as closed minded sheep that cannot comprehend your scientifically sound circuit. Every word you type has me cringing. I was pulling for the product until I started seeing these snake oil responses which are typically from someone that could not possibly understand the topic well enough so they avoid trying to form an answer. A reference is a CRITICAL device that someone like me would be counting on to ensure my circuits properly interface with other circuits designed by others. I cannot rely on magic to get there. Are you really afraid that someone will take this secret and start making $100 reference instruments?
The pot screws are friction fit into the case? That is NUTS! Open the cover an the calibration is instantly destroyed? A normal instrument has a sticker for that purpose.
Criticizing other people does nothing to advance one's own cause, and it is the bastion of the weak-minded. If you can do better, why don't you? Because you really can't. I say, let's see the results! You might trivialize my efforts, but you trivialize yourself even more, because you haven't accomplished what I have.
This is simply not true and an obvious red herring.
I am sorry to say that in addition to my personal opinion on the quality of the construction, I am now even more put off by the reaction and statements made by the seller.