WXforum.net
Weather Station Hardware => Davis Instruments Weather Stations => Topic started by: doubleohwhatever on June 07, 2021, 03:32:08 AM
-
Hi All,
Is anyone interested in an adapter that would allow an RM Young serial or 0-5v anemometer to be connected to a VP2 ISS (or standalone anemometer transmitter)? We have a need for such a device and have put one together. We're currently testing a prototype and the next revision of the PCB is on it's way from China. We'll likely do one or two more revisions before calling the design finalized. Any input on what should be included is welcome.
Prototype:
[ You are not allowed to view attachments ]
Next Revision:
[ You are not allowed to view attachments ]
This is not something we'd be selling but if people are interested we can share the design files on github. We've kept the design as simple as possible and 100% through-hole for easy assembly. The board does require 12-24v DC (mains adapter or solar) but it does allow for a single cable to connect to the ISS and provide wind direction, wind speed, 5v for ISS power, and 2.5v for fan power. No ISS modifications are required. I'll post a photo of the ISS tomorrow.
We're still dialing in the performance via firmware but we're getting there. I've attached some screenshots from weatherlink.com. Red is a RM Young 9200 connected to an ISS on channel 1 and blue is a stock VP2 anemometer on a separate transmitter on channel 2. This is our test setup:
[ You are not allowed to view attachments ]
A few notes:
1) The 9200 is configured to update once per second (same as the Davis/LCJ sonic sensor) and take 200 samples per second (likely much higher than the Davis/LCJ sonic sensor).
2) The wind direction gap at north still exists. The VP2 is built around this and there's nothing we can do about it. Some resolution is also lost due to the wireless protocol stuffing the direction into 256 values.
3) The 9200 is has a much lower starting speed and thus wind direction is much more responsive to light winds. This could be adjusted via the 9200 configuration if desired.
4) We've tested emulating wind speeds up to 200MPH. So there is no practical upper limit imposed by the adapter.
5) Tested anemometers are RM Young 9100, 9200, 86000, and 05103V. However, basically any serial output or 0-5v output anemometers will work with firmware adjustments
6) We'll likely add RS485 support in the final design
7) Pinouts for the two RJ45 ports and the ISS connections currently look like this:
ADAPTER I/O PORT (RJ45):
1) Green/White (3.3v)
2) Green (5v)
3) Orange/White (GND)
4) Blue (TX TTL)
5) Blue/White (RX TTL)
6) Orange (SDA)
7) Brown/White (SCL)
8) Brown (A0/DAC)
ADAPTER ISS PORT (RJ45):
1) Green/White (ISS +5V)
2) Green (ISS GND)
3) Orange/White (WIND V+)
4) Blue (WIND DIR)
5) Blue/White (WIND GND)
6) Orange (WIND SPEED)
7) Brown/White (FAN +2.5V)
8) Brown (FAN GND)
ISS WIND (RJ11):
1) Orange/White (WIND V+)
2) Blue (WIND DIR)
3) Blue/White (WIND GND)
4) Orange (WIND SPEED)
ISS PWR (RJ11):
1) Green (ISS GND)
2) No Connect (CABLED ISS DATA)
3) No Connect (CABLED ISS DATA)
4) Green/White (ISS +5V)
FAN PWR (Molex 2510 Female):
1/Black) Brown (FAN GND)
2/Red) Brown/White (FAN +2.5V)
-
Davis used to make one of these that interfaced with RM Young and other Anemometers, but they stopped making it.
And even though I don't have an RM Young, yes, I think you should produce this. It is going to be a low volume, niche item though.
But what you are making will save people a lot of hassle and money - and if I had one - you would be a LIFESAVER.
So if you want to sell or just release the PCB CAD file and bill of materials - DO IT.
We need more people like you!
-
This is exciting.
I don't have your talent, but always consider such things 'possible' if you have the training and time to pull something like this off.
A few years ago a smart guy down in FL, I believe, had come up with a way of taking info from more modern booms and converting it so old Heathkit 4001 and 5001s could interface, restoring them to usefulness, as they are still very functional until components start to fail.
Amazing, but I think that some problems with obtaining local info to fill in for missing sensors came about, and never heard more.
Ingenuity!
Almost all RMYoung stuff comes as raw output, 4-20 ma output and 0-1 or sometimes 0-5 v interface. Is your setup able to take any or all of those types, or just the 0-5?
-
Davis used to make one of these that interfaced with RM Young and other Anemometers, but they stopped making it.
I really wish Davis made more adapters for 3rd party hardware. I can understand why they want to stick to their plug and play hardware though.
It is going to be a low volume, niche item though.
So if you want to sell or just release the PCB CAD file and bill of materials - DO IT.
We're not looking to make a profit or sell anything. Once the design is final and tested (mainly just to make sure it's ISS safe), we're going to upload everything to github with build instructions.
Almost all RMYoung stuff comes as raw output, 4-20 ma output and 0-1 or sometimes 0-5 v interface. Is your setup able to take any or all of those types, or just the 0-5?
Most of the RM Young mechanical anemometers are offered by default as pulse output. However, almost all of those pulse output sensors can either be ordered with 0-5v instead *or* the sensor can be paired with the RM Young 05608C "pulse to 0-5v" translator. The newer anemometers RM Young has released have been mostly ultrasonic and all of those offer serial output with a few models tossing in 0-5v outputs in addition.
Note that our 0-5v interface can be scaled down via software to work with 0-1v at the same 16-bit resolution as 0-5v. No additional hardware is needed for that.
To make it clear, these are the interfaces that ARE supported:
- RS232 Serial
- RS485 Serial (currently via adapter but will be built-in with final design)
- 0-5v (0-1v would use the same interface)
These are the interfaces that ARE NOT supported:
- NMEA
- SDI-12
- 4-20 mA
- Pulse Output
I'm 99.9% sure NMEA and SDI-12 support could be added via software without any hardware changes and we make look at these at a later date. We may also look at adding support for pulse output and 4-20ma output via a daughter board (or an updated main board design) in the future as well. For now though, we're going with serial for digital input and voltage for analog input.
Overall the hardware will support these (and likely other) models:
- RM Young 86000/86004 via RS-232/RS-485 Half-Duplex *or* 0-5v
- RM Young 91000/91500/91000B/91500B via RS-232/RS-485 Half-Duplex
- RM Young 92000/92500 via RS-232/RS-485 Half-Duplex (All sensors othes than wind speed and direction ignored)
- RM Young 03002V/05305V/05103V via 0-5v
- RM Young 05106 via 0-5v provided by 05603C
- RM Young 05108/05108-45 via 0-5v provided by 05608C
- Vaisala WXT532 via RS-232/RS-485 Half-Duplex
- Vaisala WXT533/WXT536 via RS-232/RS-485 Half-Duplex (All sensors other than wind speed and direction ignored)
- Lufft WS200/Ventus/Venus-X via RS-485 Half-Duplex
- Lufft WS500/WS501/WS502/WS503/WS504/WS540/WS600/WS601/WS700/WS800 via RS-485 Half-Duplex (All sensors other than wind speed and direction ignored)
The firmware is not going to be a massive all-in-one program. We're going to create a github repository for the hardware and will have a list there of other repos containing firmware. While we won't be providing firmware for *all* of the above models, we are going to provide two firmware packages to start off. One for the RM Young serial interface models in the above list and a second for 0-5v devices that will support the RM Young models listed but also any 0-5v sensor out there.
Btw, I've attached a photo of how our board connects to the ISS. It plugs into the existing ports on the ISS and uses a female molex connector to plug into the fan connector. No splicing or anything that would void a warranty.
-
Indeed a fine project.
I know those of us above the freeze line can't depend upon ultrasonics for snow build up etc. Vaisala has the WXT5xx series which is snazzy, but a real pain when we get wet or a heavy fast snow.
Thanks for all the work, and having dug into how these things work to be able to pull together your project.
I have no real love for 4-20 ma loops, but some of the surplus Young stuff out there on eBay, etc. has this. The pulse or 3 ac cycles per revolution are a bit prone to noise in some situations, but all their 51xx can be wired to get that out if you don't mind very small wires!
None the less, I'm very interested and will keep looking for your posts on providing details when you are happy with the design.
Thx again.
-
.... I know those of us above the freeze line can't depend upon ultrasonics for snow build up etc.
Ever tried something like a LUFFT Ventus-X under ice conditions?
-
Interesting what your using. Not sure what v2 will look like...
I see your using a Featherwing, looks like a 7805, an Opamp or comparator, etc....how are you getting the 2.5v? LM317?
The Featherwing could be condensed down to a single ATMega328 and ancillary components, just provide the flash .hex file or the Arduino source file. That's what I did with my custom ISS fan monitor project. You can add an ICSP header like I did for firmware upgrades.
https://www.wxforum.net/index.php?topic=41813.msg427952#msg427952 (see Transmitter board picture)
-
One of the goals is to keep the board as simple as possible in order to make assembling it doable for almost anyone. Cost is important but ultimately secondary to this goal.
how are you getting the 2.5v? LM317?
The prototype board in the photo up above is just using a hacked on dc-dc converter. We uh, forgot to include a 2.5v regulator on that board #-o
Every board since that one uses a 2.5v and 5v regulators from Traco Power. We like them because they have built in filter caps but they aren't cheap. However, they are pin compatible with cheaper regulators if someone wants to go that route.
The Featherwing could be condensed down to a single ATMega328 and ancillary components, just provide the flash .hex file or the Arduino source file. That's what I did with my custom ISS fan monitor project. You can add an ICSP header like I did for firmware upgrades.
We went with a feather board design for a number of reasons:
- Going with a dev board like a Feather reduces the number of different components that have to be soldered onto the board. It's much easier for someone to solder on a couple of headers and plug in an entire MCU than have to solder on individual components.
- Going bare bones with a microcontroller IC and an ISCP header is also out of the question because it's not nearly as user friendly as plugging in a USB cable for firmware upgrades.
- Doing firmware upgrades in the field is going to be super easy because we can just swap out the feather board.
- Feather boards were chosen due to the footprint. It's basically a standard now and as such there are numerous board models and wings available.
- Lastly, this board is not going to be single purpose. We have plans for future projects involving the board and different Feather boards will be needed for those projects. One PCB for numerous projects is a win f
For this use of the board, we went with an M0 because we don't really work with 328/32u microcontrollers anymore. We really only work with M0/M4 and ESP8266/ESP32 microcontrollers now.
-
Quite interesting. Thanks for the reply!
-
this is a fantastic project, just what davis stations need. UU If you need a beta tester, I'm available! I have over 40 weather stations in my region.
-
I'd like to see more projects like this that can support not only Anemometers, but rain gauges and temperature/hum probes as well.
Davis and the SHT31 is starting to get long in the tooth, there are better SHT sensors and even more accurate ones from Vaisala, etc. out there.
If the OP would entertain that idea, I think he would be a MASSIVE benefit to the community. Rain gauges and temp probe conversion shouldn't be as hard as an Anemometer as there isn't as much math involved.
I mean imagine the possibilities!
-
Hi All,
I'm out of town this weekend with very slow internet access. I'll post an update on the wind project on Monday.
As for interfacing different rain buckets, that should be super simple as long as the bucket is 0.01 inch per tip. It's just a matter of figuring out the pinout for the rain RJ11 port. I'll take a look when I get home. I may have to order a spare tipping spoon assembly to take apart. I know two wires are for the reed switch but at least one of the remaining wires will be connected to something in order to let the ISS board know the rain sensor is connected.
The temperature/humidity sensor is a bit trickier. The problem is that Davis has thus far chosen to not switch the temp/hum sensor over to an I2C variant. They've stuck with the older Sensirion Bus chips because they don't want to go through the expense of changing the ISS board design. I imagine since the radio is integrated into that board they'll have to get the updated design FCC tested and approved which is costly. Unfortunately for them, the older Sensirion Bus chips are no longer manufactured. I'm sure they have a massive stock of the chips but at some point they are going to have no choice but to go with a different temp/hum chip that communicates via SPI, I2C, or voltage (analog). I2C would be my bet but any of those would make it super easy to swap out the temperature and humidity sensor. As it is now, you would have to emulate the Sensirion Bus protocol. I'm honestly not sure going down that path would be worth the time seeing as how Davis is going to have to modernize the temperature/humidity sensor in the relatively near future.
-
Exactly. My question would be could the I2C returns (i.e. query bus and it returns 83.1 as temp via i2C via the SHT library for ATMega or ESP32, can it be mapped out voltage wise to send analog returns back to the ISS ex. 4.22v for that return), or is the Sensiron bus a special protocol in itself that does not lend itself to doing that?
-
The Sensirion Bus is a digital bus similar to but not compatible with I2C. You would have to have an intermediary microcontroller that reads an I2C sensor and outputs Sensirion Bus to the ISS.
I don't think it would be impossible to do and I think I came across a project on GitHub at some point that could read from Sensirion Bus sensors using an arduino's I2C port. That project would likely have a majority of the code needed.
It'll probably take another three weeks to get the anemometer project wrapped up (mainly due to PCB shipping from china). After that I'll take a look at the temp/hum sensor. I think an adapter for that would be useful for at least a few people.
-
Interesting and again wow I'm stoked. Glad to have you here and thanks again for the replies!
-
Quick update. We're still waiting on the latest PCB to arrive. We should have it either tomorrow or the next day.
We've concluded our comparison testing between the RM Young 92000 using our adapter and the stock Davis anemometer. I've attached a chart from today. We were initially trying to get the two anemometers to agree as much as possible. However, we've come to realize that is just pointless. The 92000 performs great in light winds (<2mph) while the Davis anemometer just doesn't due to it being mechanical. It's an apples to oranges comparison no matter what and I think it's as good as it's going to get. Notes:
1) Our 92000 was calibrated just a few months ago.
2) During lab testing for wind direction, whatever value our adapter sent the ISS was reported accurately to weatherlink.com and the console within plus or minus two degrees. Wind direction is hard due to the dead zone in the stock anemometer which the ISS accounts for *and* the fact that the Davis wireless protocol squeezes 360 degrees into 256 values. Dead on accuracy just isn't going to happen.
3) During lab testing for wind speed, whatever value our adapter sent the ISS was reported accurately to weatherlink.com and the console. I hesitate to say it's perfect but we did not notice any errors. 5mph was reported as 5mph. 4.4mph was reported as 4mph. 4.5 mph was reported as 5mph.
4) The 92000 performs very well at low wind speeds (<2mph) where as the stock Davis anemometer does not due to it being a mechanical sensor.
5) When no wind pulse is sent to the ISS, it will stop reporting wind direction. This happens often with the stock anemometer due to its starting speed. It's rare for the 92000 to register no wind and thus is almost always sending wind pulses (via adapter board) to the ISS. Even if it's reporting 0.2mph, it's still sending pulses which results in the ISS still sending wind direction even if the wind speed is rounded down to 0mph. You'll see this in the attached chart as breaks in the blue line on the wind direction comparison.
6) When reading the attached chart, keep in mind that blue is the stock anemometer and red is the 9200. Also note the North to North North-West wraparound on the wind direction comparison.
7) Either the mechanical sensor is picking up gusts better than the 92000 or it's overrunning (or both).
8) All variations on the attaches chart on within tolerances for the respective anemometers.
9) Bottom line is that we know our 92000 is calibrated and we know that what the adapter sends to the ISS is reported correctly. We're going to see if we can put together some kind of demo to show the accuracy later this week.
this is a fantastic project, just what davis stations need. UU If you need a beta tester, I'm available! I have over 40 weather stations in my region.
Sorry I missed your post. I'm hoping we'll have something we can put on GitHub by the end of the month.
-
are there any problems in heavy rain and fog with the sonic anemometer?
-
are there any problems in heavy rain and fog with the sonic anemometer?
The answer depends on the model you're referring to. Currently we use Young 86000 units when heating is needed and Young 91000 (black) units when heating is not needed. We also have a number of Young 91000 units. We've had zero issues with these models. They basically use the same
We've used quite a number of sonic anemometers over the years and they basically fit into three categories. I'm sure there are proper names but this is what I call them:
1) Cross-firing: We've used RM Young 86000, 91000, etc and Vaisala WXT models. We've never had an issue with any of these in any weather situation. That said, NOAA reportedly had an issue with Vaisala WXT models on buoys reporting erroneous wind spikes. We never encountered that issue. We stopped using WXT models because the cables and the connectors are expensive.
2) Upward-firing: We've used numerous Lufft models but primarily the WS200. 99% of our stations are located in Alabama where it's warm and snow/ice is rarely a concern. However, we have done stations in slightly colder areas and we found that the transducers on the lufft stations would get blocked by even light snow a lot more frequently than the RM Young 86000/92000. We eventually switched all of these stations out to heated Young 86000 units. Lufft makes heated variants and I am 100% sure those would work great. Again, we stopped using Lufft sensors due the cost of the connectors and the premade cables.
3) Downward-firing: You tend to see these on consumer-grade and cheaper professions stations and sensors. There's a reason for that. I really don't want to discuss this too much because it'll lead to discussing the consumer-grade stations that use these sensors and at the end of the day I'm just happy to see people enjoying the hobby. If you're going to spend $1k USD on a sensor for a professional station, go with cross-firing or upward firing.
Note that we've never had a use for 3D sonic anemometers have no experience with them.
Back on topic... the PCB for the latest version of our adapter board arrived today. I'll get one built and start playing with it tomorrow. Unless we find any issues with this board, we really only need to do one more revision adding RS-485, test that revision, and then we're good to release everything.
-
Hi All,
I just wanted to post a quick update. We're still moving along on this project but it's been a bit slow going. A shipment of boards got lost in the void of customs and some parts were backordered. However, we now have the second iteration of the board built and installed on the test station. A photo of the board is attached.
The next version has a few minor issues resolved but we've run into a small problem that we're having to decide how to handle. We want to keep the board 100% through-hole in order to make it easy to assemble. However, through-hole components are larger and we're running into routing issues while trying to add rs-485 support to the board. Basically, we have two options:
1) Ditch the IO RJ45 port and add RS-485 half duplex support
2) Leave off RS-485 support and require an external adapter
Using an external RS-485 to RS-232 adapter isn't a big deal and a lot of wind sensors offer RS-232 anyway. However, having it onboard makes for a cleaner install.
For those not familiar, this is an example of an adapter:
https://www.amazon.com/Electromagnetic-RS232-RS485-Converter-Transmission/dp/B08CBX497H/
Does anyone have any thoughts on this?
-
why two rj45 ?
-
The left RJ45 port voltage outputs, I2C, TTL serial, and a 0-3v DAC output.
-
Hi All,
We've sent off the (hopefully) last board revision (v3) for manufacturing. We'll test this last revision when it arrives and then we'll upload everything to github.
This revision just contains non-critical fixes from the previous version (v2). We've had one of the v2 boards running without issue for a few days now. So I'm attaching the gerbers in case anyone wants to give it a go. JLCPCB is who we use and the attached gerbers will work there without issue.
-
the site you mentioned is also able to assemble the various components on the pcb and flash the firmware to make it work in a "plug and play way"? For non-expert users it would be essential, otherwise only few people would be able to produce working parts. UU
-
the site you mentioned is also able to assemble the various components on the pcb and flash the firmware to make it work in a "plug and play way"?
I'm afraid not. You'll be able to order bare boards from JLCPCB or pretty much any other fab house using the gerber files we provide. You'll then need to order the components in the BOM (bill of materials), solder the components onto the board, and flash the firmware (just an arduino sketch) onto the microcontroller (Feather M0 Basic Proto).
The current BOM cost is around $78 USD not including the PCB but you can leave a few components off if you don't plan on using certain features. I've attached the current BOM we've been working off of. I had to zip it in order for it to upload. It's just an excel/google sheets file.
As mentioned above, the firmware is just an arduino sketch. It's not complicated nor it is anything fancy. I've attached what we're currently running on our stations. The target board is an AdaFruit Feather M0 Basic Proto and you'll need these two libraries:
https://github.com/khoih-prog/SAMD_TimerInterrupt
https://github.com/jmalloc/arduino-mcp4xxx
Again though, once we've received/built/tested the final board revision, we'll upload everything to github. I'm just attaching what we have now in case anyone reaaaally want's to build one now.
I do apologize if all of this is more complicated than some people thought it would be. If someone wants to take what we have and turn it into a commercial product then they have permission. We needed this adapter for our own stations and are simply sharing it as-is. It's free as in beer. No license, no conditions, no warranty, no guarantees.
-
Can I ask why your not using caps on the VR? I see no space for them. If it's a 7805 it'll need 0.1 and 0.33uf caps on it to prevent oscillating. Different may need some other combo per datasheet.
Granted you can get away without using them I've done it just not best design practice.
-
Sorry for the delay. I'm out of town this week.
It's not a standard voltage regulator. It's a pin compatible drop in package that has the caps built-in. Ex for 5V:
https://www.digikey.com/en/products/detail/traco-power/TSR-1-2450/9383780
The last board revision arrived last week but I've been out of town for the last six days and haven't had a chance to assemble one yet.
-
I figured there's something special.
Thanks for the reply 😊
-
what is the input voltage for the board ?
-
Technically 9-36v. However ~12-24v is ideal.
-
Technically 9-36v. However ~12-24v is ideal.
thanks UU
-
hope everything is ok ...!
-
All good. Thanks for checking in!
At the moment I'm just backed up with projects that pay the bills. Hopefully I should have some time soon to finish this project up.
-
Regarding your image (in this post (https://www.wxforum.net/index.php?topic=42145.msg430811#msg430811)): what are the cables going into the RJ connector at the top edge and under the white cover?
-
The green pair is providing 5v to the ISS. The brown pair is providing 2.5V to the shield fan. Basically the adapter board provides all power for the Davis hardware. The adapter board has a wide input range of 9-36v (12-24v nom) which makes it suitable for the entire setup to be powered by a larger 12v or 24v solar system.
-
The green pair is providing 5v to the ISS.
I see, thanks. I forgot that the ISS can take 5V power on that socket.
-
any update ??
-
Sorry for the delay.
We've run into a firmware issue that took us while to narrow down. We noticed that every ~1.5 months the microcontroller would crash. We added a data logger and dumped info every second to pinpoint the problem. We thought it would turn out to be a memory leak but the first time the crash happened with the logger in place, the memory usage was fine. The second time around we dumped more info and discovered that it was related to millis overflow. Our code handles that just fine but apparently a timer library we're using does not. We've tossed a temporary fix in place (software reset the board once a month) and we're going to look at the library code when we have time.
Btw, this is absolutely a back burner project for us at the moment and will never be an "easily purchased and put to use" product. There are now several ultrasonic anemometers on the market that offer Davis compatibility. So if you need a solution sooner rather than later I would look into one of those. We're looking into the non-solar powered LCJ unit ourselves.
-
Hi which lcj model are you looking at?
Thanks
Frank
-
This model:
https://lcjcapteurs.com/en/girouette-anemometres-capteur-vent/sa-dvc-2/
Trying to locate a reseller in the US.
-
Thanks. Looks similar specs to the solar powered one Davis sells.
Curious why you thinking of going with the power cabled one?
Ask because the solar powered one is available for sell in the US.
Thanks
Frank
-
Several reasons:
1) The output rate is greater on the wired version. The solar powered version outputs a signal to the Davis ISS once per second. The non-solar powered version provides a continuous output like the standard Davis anemometer.
2) All of our sites are already powered via mains (with backup batteries) or larger solar systems. Dealing with a second (fragile - See #3) battery is not something we're interested in dealing with.
3) The solar powered unit uses a LiFePO4 battery. That battery chemistry CANNOT be charged below the freezing point. We use LiFePO4 batteries in our solar setups but they're BattleBorn self-heating units. The batteries use their own power to keep their internal temperature above freezing.
-
This model:
https://lcjcapteurs.com/en/girouette-anemometres-capteur-vent/sa-dvc-2/
Trying to locate a reseller in the US.
Rainwise used to have this version on their website as well as the solar powered one. I don't see either on there now, so maybe it has something to do with their acquisition by NK, but either way, may be worth reaching out to them.
-
Thanks!
-
Any updates on this project? I would like to move from the Davis cups anemometer, and have looked into a Young, or going with a sonic one...but not the Davis model...
Did you find a US source for the LCJ unit?
-
LCJ don't works well. We need a RMY !!
-
Did you find a US source for the LCJ unit?
No. However, you can get in touch with LCJ and they can ship you a wired Davis-compatible unit. Apparently the only Davis-compatible version readily available in the US is the solar powered unit. The problem is that they want a good bit more for the wired version than what you can get the solar/battery powered version for. Doesn't make sense to me.
Ultimately it doesn't matter though. We were able to play with a serial output unit and it did not perform well in heavy rain.
Any updates on this project? I would like to move from the Davis cups anemometer, and have looked into a Young, or going with a sonic one...but not the Davis model...
No real updates. We have thirteen stations using the beta boards. Currently we're aware of two problems with the firmware and one problem with the board:
1) millis overflow is not handled well - the problem is in a M0 timer library that we just haven't had time to dig into
2) a wind spike is produced upon powering on the board - we need to disable the output for a few seconds to allow the board to stabilize before sending data to the ISS
3) In a power outage event, 5V DC flows from the ISS to the board draining the ISS - just needs a diode
At the moment we're undecided on whether to create another iteration of the board or just fix the boards we already have. If we do create another iteration of the board we'd likely aim to make it more compact by ditching the analog voltage input and IO RJ45 port.
Whatever we do, it won't be until spring. What we have in place now is technically working and none of us like cold weather.
LCJ don't works well.
I have to agree. I think the appeal of them has to be with the size and ruggedness. The sensors are well shielded as opposed to something like a Vaisala WXT or Young 86000/91000. The transducers on the Young and Vaisala units are delicate.
-
Did you find a US source for the LCJ unit?
No. However, you can get in touch with LCJ and they can ship you a wired Davis-compatible unit. Apparently the only Davis-compatible version readily available in the US is the solar powered unit. The problem is that they want a good bit more for the wired version than what you can get the solar/battery powered version for. Doesn't make sense to me.
What you're seeing are the standard LCJ list prices. Davis obviously managed to strike a good deal with LCJ for the supply of the 6415 unit, presumably because of the numbers involved. But the special pricing only relates to 6415 and not to any other LCJ models.
-
Ah! I didn't think about Davis making a deal for a lower unit cost due to sales potential.
-
is there a way to use https://www.raspberrypi.com/products/raspberry-pi-pico/ raspberry pico for this project ?
-
is there a way to use https://www.raspberrypi.com/products/raspberry-pi-pico/ raspberry pico for this project ?
Almost certainly yes - the Pico has a range of inputs including interrupts, A2D, I2C etc etc and so I'd be pretty sure that it's possible in principle. But I suspect you're asking whether anyone has actually done it and can supply details. That's a different question.
-
As John explained, it's 100% doable. The feather board we used contains an Arm Cortex M0 and the Pi Pico contains an Arm Cortex-M0+. Both can run micro python and both can be used with the Arduino IDE.
That said, I'm not aware of anyone using a Pico to interface a 3rd party anemometer with a Davis ISS.
We've actually retired this project. We have two coastal stations that we'll continue to use our adapter on for now. The rest of our stations we've switched to the stock Davis anemometer. We did the math and discovered it's cheaper to just use the Davis anemometer in the long run. We plan on replacing speed cartridges yearly and the entire anemometer every 3-4 years.
-
I haven't read this thread in awhile and scanning it in response to the author answering another thread, I can only imagine the enormous amount of work and skill and technical background that went into the iterations of the board and tracking down those pesky errors.
It isn't just slapping a few things on a bread board and going merrily on the way.
And taking the time to discuss the journey for those of us who know little about the professionalism of doing such things was enlightening. Thank you Dale
-
No worries. Hopefully someone got some useful info out of it.