EEVblog Electronics Community Forum
Electronics => Projects, Designs, and Technical Stuff => Topic started by: ali_asadzadeh on December 10, 2019, 07:27:30 am
-
Hi,
I want to design a data logger with a 170C working parts for a customer, they have a sensor with a 0-250mV full scale output, so I should read the sensor data with a 0.05% full scale and store it in a RAM and later I should send it to the PC, I need a 16-24Bit ADC, a MCU and about 4-8MB of RAM, a RTC for time tag and the whole system should work with a single 3.7V Li-ion battery for 3 days(72hours), it would be used in a high temperature environment, so the parts should with stand the environment (170C).
I'm trying to find the parts for the system, The parts seems pricey, so I came here to ask for advise and better priced parts. so far I could only find some parts for the ADC, something like ads1278,
But it's pricey, I need to find the MCU (I prefer ARM Cortex) and RAM and RTC too, But digikey does not seems helping me in here. do you have any parts in mind?
There seems to be no RAM and RTC in this working temp |O
DO we have chinese parts?
-
I often see parts rated to 150C, but not often anything more than high-power discretes rated above that. I suspect it's a limitation of the IC processes/chemistries commonly used for memory & dense logic.
Sadly I don't think automotive parts are rated that high either, which knocks out a big possible market. Hopefully you're not stuck with space-rated level parts.
Ordinary search engines might be a better idea for researching this. There are some companies that claim to provide such parts, but I am going to assume they're $$$ and slow/clunky ordering.
https://www.voragotech.com/products-main/#arm-microcontrollers (https://www.voragotech.com/products-main/#arm-microcontrollers)
EDIT: 3 days of operation. Hmm, I wonder how much thermal mass and insulation you would need to get away with more standard 100-150degC parts.
-
You might need to think (literally) outside the box.
You could run the sensor cables to a monitor outside the environment. If the length is too long, devise a primitive modulator.
An IR thermometer could perform measurements from outside the environment through the correct type of glass.
You could use sacrificial cooling, e.g. water with a steam vent. An insulated device containing a sizeable water bath would be able to stay at 100° for a long time.
Etcetera ...
-
Small-ish thermos bottle filled with water, boiling keeps the temp at 100C
-
It would be used for deep down hole drilling for oil! so these solutions would not apply! Now I need to find a RAM and the RTC :-X
-
Erh, Li-Ion for 170C°? :scared:
-
As others have said /hinted at - LiIon highly unlikely to work unless you get specially made Lithium cells (usually primary/non-rechargable) - they do exist, but very specialised. Look at what companies like Steatite, Creasefield, ElectroChem etc.. recommend for high-temp down bore applications.
Alot of the silicon will work at 170C, although you'll have to try it on a case by case basis (testing will be the only way to proceed). Most modern IC packaging will also work, for a limited period at least, high temperature lead-free solder, and probably find somesort of silicone potting compound to hold everything in place.
Tg170 or higher PCB substrate, forget the cheap PCB services.
Old school RAM will probably work, FLASH probably won't. Build in error detection/correction when storing your data. Also old school processors with as little going on as possible - run them slow. RTC - be really simple and record samples at known intervals, don't overcomplicate design. What is the bandwidth of what you are recording? Temperature/pressure?
Understand the failure mechanisms of your design, avoid thermal shock/uneven heating and look up electromigration in silicon due to thermal effects. Automotive passives (capactiors) will be less prone to cracking due to the flexiable design of the terminations. Even solder flux contaminants could kill your design. Design for low power, use big junctions - cause any extra heat might tip you over the edge.
Good luck & TEST.
-
The specs are unrealistic. You won't make it work reliable at 170*C.
Either place the electronics far away from the heated area, or in a cooled enclosure.
-
It would be used for deep down hole drilling for oil! so these solutions would not apply! Now I need to find a RAM and the RTC :-X
Your customer who wants drill for oil has certainly enough $$ to pay for a high-end professional solution. No need to ask eevblog for an advice, imho.
-
AFAIK oil drills continuously pump mud as lubricant/coolant for the drilling head, but I'm not very sure this is always the case. Maybe it could be used to cool the electronics, too.
-
I want to design a data logger with a 170C working parts for a customer...
...the whole system should work with a single 3.7V Li-ion battery for 3 days(72hours)
...
DO we have chinese parts?
Do you do ANY work yourself, or do you outsource everything to this forum?
The only components made that are likely to tolerate operation at 170C are resistors and SiC MOSFETs and diodes. Molten salt batteries might be okay at 170C (might actually need to be hotter than that) but certainly nothing with a Li-ion chemistry.
No MCUs are rated for >125C that I am aware of, but if there are any then they certainly won't be some cheap-ass Chinese parts like you are so fond of demanding we find for you.
-
I want to design a data logger with a 170C working parts for a customer...
...the whole system should work with a single 3.7V Li-ion battery for 3 days(72hours)
...
DO we have chinese parts?
Do you do ANY work yourself, or do you outsource everything to this forum?
The only components made that are likely to tolerate operation at 170C are resistors and SiC MOSFETs and diodes. Molten salt batteries might be okay at 170C (might actually need to be hotter than that) but certainly nothing with a Li-ion chemistry.
No MCUs are rated for >125C that I am aware of, but if there are any then they certainly won't be some cheap-ass Chinese parts like you are so fond of demanding we find for you.
Try hard-er
https://aspencore.us.janrainsso.com/static/server.html?origin=https%3A%2F%2Fwww.embedded.com%2Ffloating-point-microcontroller-for-extreme-temperatures-up-210-c%2F (https://aspencore.us.janrainsso.com/static/server.html?origin=https%3A%2F%2Fwww.embedded.com%2Ffloating-point-microcontroller-for-extreme-temperatures-up-210-c%2F)
-
...
Try hard-er
...
Why should I? I'm not the one trying to design something that works at 170C in an oil well bore hole for as cheap as possible.
-
The battery is going to be a major problem unless you use one with a special chemistry and construction intended for high temperature operation.
Parts typically derate to 0 power by the time they reach 150C so be as conservative as possible with power and voltage derating. Watch out for leakage in precision circuits.
What I would do is qualify parts by manufacturer and perhaps even lot for extended operation at 170C or actually a little higher. Destructive testing should give some idea of the reliability over temperature.
-
Sorry to say that people who know how to design circuits that can run at 170'C will already be working for companies making borehole logging equipment or similar. They will have access to a great deal of internal company hard-won knowledge and experience.
As an 'outsider', you have very little chance of pulling it off successfully.
-
It would be used for deep down hole drilling for oil! so these solutions would not apply!
Good luck with that. It's not a gentle environment ...
RoGeorge is right: the drilling head should be being cooled by the flow of mud. And perhaps you can use the temperature differential to generate power via a Peltier device, thus avoiding the battery problem.
-
Oil well logging equipment also has incredibly difficult sealing and pressure resistance requirements, not to mention mechanical ruggedness. I don't know if the (potential) customer has all of that taken care of, if not, then temperature may be the least of your problems.
-
Guy thanks for the answers, But why panic so much! :) somebody out there has done it, And I'm a professional too, do not worry I can do it too! I managed to find almost 95% of the parts that I needed including passives, ADC, MCU,Flash,RTC and osilator, my main problem for now is the RAM, I have found some parts, But I do not want to limit my choices to only a single source!
The MCU has only 32KB of internal RAM so either I should find a way to compress the DATA or find an external SRAM (I prefer SPI, But I'm fine with the parallel bus too)
This is the RAM
https://www.voragotech.com/products/hs512k16 (https://www.voragotech.com/products/hs512k16)
Though I wonder how it should be used, all the pins seems to short on the outer frame! should I cut it out? any other source for the RAM is highly appreciated. I do not want to limit my choices to a single manufacturer.
-
Guy thanks for the answers, But why panic so much! :)
I just don't want to see you ending up with your ass in a sling. ;)
P.S. Just a quick one. Don't forget that most Vreg ICs go into die over-temperature shutdown at around that temperature or below.
-
Good luck sourcing the parts if you live in Iran. Most of them are probably under one or another export limitation.
-
Guy thanks for the answers, But why panic so much! :) somebody out there has done it, And I'm a professional too, do not worry I can do it too! I managed to find almost 95% of the parts that I needed including passives, ADC, MCU,Flash,RTC and osilator, my main problem for now is the RAM, I have found some parts, But I do not want to limit my choices to only a single source!
The MCU has only 32KB of internal RAM so either I should find a way to compress the DATA or find an external SRAM (I prefer SPI, But I'm fine with the parallel bus too)
This is the RAM
https://www.voragotech.com/products/hs512k16 (https://www.voragotech.com/products/hs512k16)
Though I wonder how it should be used, all the pins seems to short on the outer frame! should I cut it out? any other source for the RAM is highly appreciated. I do not want to limit my choices to a single manufacturer.
;D
If you seen the request for such a device in some auction job, just be careful what contract you sign, and how much damage you will pay if your device won't meet the specs and the deadline. It's obvious you had no idea what you are trying to do, and chances you can build that in time are zero. Find some other less demanding product to build. Please don't get this as offensive, it is just a realistic evaluation. Good luck anyway.
-
Good luck sourcing the parts if you live in Iran. Most of them are probably under one or another export limitation.
I wouldn't be surprised actually, if discussing this topic violates most others countries' embargo/sanction laws. :(
Tim
-
If you seen the request for such a device in some auction job, just be careful what contract you sign, and how much damage you will pay if your device won't meet the specs and the deadline. It's obvious you had no idea what you are trying to do, and chances you can build that in time are zero. Find some other less demanding product to build. Please don't get this as offensive, it is just a realistic evaluation. Good luck anyway.
Actually I can build it, ;) the problem for now is beating the hell out of the pricey parts! :palm:
-
This is the RAM
https://www.voragotech.com/products/hs512k16 (https://www.voragotech.com/products/hs512k16)
Though I wonder how it should be used, all the pins seems to short on the outer frame! should I cut it out?
Errrrmmm, is that a legitimate question? Or a troll? For operation at 170C ambient you'll be using highly specialized assembly techniques. Most customers will be bonding directly to a bare die mounted on an inorganic substrate, or putting the encapsulated part into a cavity and welding the leads. If welding they want the intact lead frame, before bending and trimming.
-
Batteries are available toat least 270C (but probably not in Iran). Regulators will typically have thermal shutdown at 155C so probably roll your own.
I've got pucker thermal chambers for -65C to +125C (-85F to 257F), and a small Memmert oven that goes to over 200C (or so it says on the knob) - just fired it up to see what it'll get to - will see what happens to the following parts at temperature:
PIC micro controller (probably an 18F something or other) running an on-board ADC, spitting out serial.
AA or AAA cells (bog standard)
LiIon 18650 - standard COTS
Will film this for YT and see what happens (got a fire extinguisher...), watch this space :scared:
-
:popcorn:
-
Small-ish thermos bottle filled with water, boiling keeps the temp at 100C
Thats basically a pressure cooker. It wont work, the pressure will just keep rising and water boils at higher temps at higher pressures. I guess you could put a valve so you keep pressure constant but you'll run out of water eventually.
Anyway, maybe you could use thermoelectric cooling down there?
-
Batteries are available toat least 270C (but probably not in Iran). Regulators will typically have thermal shutdown at 155C so probably roll your own.
I've got pucker thermal chambers for -65C to +125C (-85F to 257F), and a small Memmert oven that goes to over 200C (or so it says on the knob) - just fired it up to see what it'll get to - will see what happens to the following parts at temperature:
PIC micro controller (probably an 18F something or other) running an on-board ADC, spitting out serial.
AA or AAA cells (bog standard)
LiIon 18650 - standard COTS
Will film this for YT and see what happens (got a fire extinguisher...), watch this space
Sound a very good Idea and experiment :-+
-
Small-ish thermos bottle filled with water, boiling keeps the temp at 100C
Thats basically a pressure cooker. It wont work, the pressure will just keep rising and water boils at higher temps at higher pressures. I guess you could put a valve so you keep pressure constant but you'll run out of water eventually.
Anyway, maybe you could use thermoelectric cooling down there?
Nothing last forever. But IF you read OPs post he needed 72 hours of operation.
Anywayz, as later found out he has no place to went the steam or ambient pressure is too high in the borehole.