Massive project for temp. and i just heard of picaxe

bullethead67

New Member
OK first. I just heard of picaxe so i know almost nothing. I have downloaded the manuals and am starting to understand some of it.

What i need to do is record temps on 56 very large shaft bearings. These bearings are on a ship (no really im serious) and are spread out about 240ft (73 meters).

27 bearings Port side
27 Bearings Starboard side
2 bearings Forward center

I would like to have them change a RGB led

RED - too hot
BLUE - too cold
GREEN - in operating range

This is a lot of pins needed. I would like to use the digital temp sensor. is there a max distance for the sensors? whitch chip should i be looking at as im still a little confused as to how many pins the sensors would take on a larger chip.

What about the RGB leds. wouldnt they take up 3 pins each. I have read a little about minimizing pins but im new and dont want to get in too something complicated for my first project.

Voltage- I have 24 volts DC and 120v AC everywhere. I assume i can use it but could someone point me in the right direction.

I imagine at some point i will be using a Xbee or Picaxe.net unit? is that right?

Im not being paid for this project. im trying to get the company to just pay for parts and tools. Pretty much im stuck here with nothing to do. Im bored
 

nick12ab

Senior Member
So you want one RGB LED per bearing do you? And one temperature sensor per bearing as well? That is a lot if I/O! Could be done with a few PICAXEs.

56x3 = 168 I/O if you don't multiplex. Multiplexing and hserial could be done at once so 168 LEDs to multiplex with 32 I/O per 40X2 but 1 I/O is used for hardware serial background receive but you could still do 240 LEDs. It will be quite dim OR flickery if you multiplex with the rather slow PICAXE however unless you're willing to overdrive the LEDs considerably but then you'd cook a row of them if the PICAXE froze so you'd have to extensively test your program for that PICAXE using no overdrive first.

The temperature sensors - you might be able to multiplex them using a 4093, 4063 or similar but again - test it! 56 could be done with 7 8-bit multiplexers so again could be done on one PICAXE but don't control the LEDs on the same PICAXE as the temperature sensors as reading all of them will take a while (if you're using a 1-wire device, it's not the fault of the PICAXE).

What is the temperature range you will be measuring...?
 

bullethead67

New Member
Temp Range

The temp range would be around 15 to 100 C

It doesnt need to be all 1 picaxe. I plan to light a panel so i could use different modules. I was thinking 1 for Port side 1 for Stbd input. Then 1 for port 1 for stbd LEDs. What about distances? the panel is 70 meters from the farthest bearing

and yes I do want 1 temp sensor per bearing and 1 led indicator per bearing.
 

eclectic

Moderator
Welcome to the Forum.

Perhaps start with one "module":

08M2 / DS18B20 / RGB LED / resistors

Get that working perfectly, then build LOTS! :)

Use one module per bearing.

70metres should be possible.
Others more expert than I will provide further details on this.

e

Added
vttom's ideas (p.5) are better than mine.
 
Last edited:

vttom

Senior Member
If it were me, I'd use 1 small PICAXE per temper sensor (so 1 PICAXE per bearing). The 08M2 would probably do nicely. Then wire them up so they all drive a common serial line (or maybe, say, 1 serial line for the port sensors, a 2nd serial line for starboard sensors, and a 3rd line for the other 2). I'd think if you size the driving transistor appropriately, and maybe use a shielded cable (like 50 or 75-Ohm coax), you should be able to drive 70meters without too much trouble (so long as you keep the data rate slow). If need be, you could daisy-chain the temp. sensor PICAXEs rather than putting them in parallel to reduce the length of cable each one drives, but you run the risk of 1 taking out the whole communications chain with a single PICAXE failure.

The light panel could be driven by a separate PICAXE.

The temp sensor PICAXEs would poll their temp sensor and store the most-recent value. Or perhaps collect a number of samples and keep some kind of rolling average. The light-panel PICAXE would poll each temp. sensor PICAXE in turn, and ask each one for the temperature it's collected, then use that information to light the light panel. I'd think the easiest way to control the light panel would be to use a bunch of outboard serial-in/parallel-out latches to drive the LEDs.

With this many components, you should also think about how to indicate a hardware failure somewhere in the system (like if a temp. sensor PICAXE stops responding when polled, or when a temp. sensor PICAXE fails to get a reading from the temp sensor).
 

hippy

Technical Support
Staff member
On the input front, the PICAXE-X2 have OWIN/OWOUT commands so 54 DS18B20's could technically be handled on a single 1-wire interface, plus it's possible to trigger multiple temperature readings, wait, then read them all back so all could be read in around a second. The biggest difficulty perhaps is going to be distance and crosstalk interference so one PICAXE per set of sensors would likely be best.

I'm presuming this is a commercial venture and budget shouldn't really be too much of a problem, so one 08M2 per bearing is the easy option. You can likely create a much more robust signalling system that way.

westaust55 is the resident 1-wire interfacing expert so if he passes by it will be well worth listening to any advice on that front.
 

Dippy

Moderator
Distance and (electrical) noise would really make one-wire interface (DS18 style) a potential nightmare.
A buffered shared serial line is very possible and far more robust.

Do you mean you will have lots of RGB LEDs on a large printed display panel??
If it were me I'd pop results on a GLCD along with the warning lights.
It would provide a double-check and reduce ambiguity.

A drawing (system schematic) of your proposed/desired system would make it a lot clearer for us lot.

I would also lean to a sensor/PICAXE/buffer per bearing. Then your serial nodes (node=your circuit at each bearing) and master (the PICAXE receiver behind your panel) could include parity/error checks etc, as I would have thought that the extra cost/effort was well justified. Penny-pinching could be a BIG mistake in a project like this so buy quality components.
Power obviously won't be a problem though good regulation/suppression practice should be incorporated.
And I assume each bearing site will require tough IP enclosures?

I would suggest that each PICAXE node used a PICAXE which could be crystal/resonator clocked. You will want maintenance-fre reliabilty and I'm not a great fan of internal resonators in 'challenging' environments.

A very interesting and feasible project at last - I'm jealous :)
 

Svejk

Senior Member
Very nice project!

As mentioned, one picaxe per bearing is the best solution. The comms between units may be made with RS485 using 4 lines only: 2 data and 2 power. Made properly it can handle hundreds of meters. Consider HMI at the receiving end. Fit an RGB LED at the unit as well, it makes fault finding a lot easier.
 

bullethead67

New Member
Drawings etc

OK so I gather 1 picaxe 08M and 1 digital sensor per bearing

and 1 big Picaxe for the panel.

I dont know anything about serial etc but i did pickup on the reliability comment.
I dont want to babysit this. It has to work when im off the ship. The others cant do this. Im going to try to get some of my engineers into the electronic side

Why i dont want LCD
Are there idiots where you work? Yeah here too. I have submitted a request for the company to discontinue the hiring of idiots. I never got a response but I obviously was denied.
So a lcd panel is out. using buttons to do anything is out. that leaves me with

blue = cold
red = hot
green = good

Bearing temps are taken by junior personnel in the engine room NOT engineers. They repeatedly have trouble remembering what number is too hot or just right.
Repeatedly reviewing this has gotten old fast. I believe some are even pencil whipping (gun decking) readings on some of the vessels. And other times your just too busy replacing a oil cooler or cardan shaft or some other big heavy metal thing that just broke and needs to be fixed right now.

I want a panel that is in the control room where all engine room personnel walk past a hundred times a day. anything not green would stand out and would be caught within minutes.

Right now all of the older boats have to
1- remove a deck plate or access cover
2- use temp gun to get reading
3- write reading in logbook (yes i know someone will bring up logging i diddnt yet because my brain is full already with what i got)
4- goto next bearing

all temps are done every hour.

That means i have 1 guy that just walks in circles all day absolutely hating his job.
and of course the bearings are holding a shaft rotating at 1500 rpms.
 

nick12ab

Senior Member
Why i dont want LCD
Are there idiots where you work? Yeah here too. I have submitted a request for the company to discontinue the hiring of idiots. I never got a response but I obviously was denied.
So a lcd panel is out. using buttons to do anything is out.
......

They repeatedly have trouble remembering what number is too hot or just right.
And not discontinue the existing idiots? By 'discontinue' I mean sack and not kill!

You could use the LCD to show heat status instead of temperature. Three dots of 2x2 pixels each could be used for each bearing, or you could use a GLCD with RGB backlight and when some get too hot or too cold you flash the backlight red or blue and at the same time show the numbers of the bearings in question. If some are too cold and some are too hot and some are normal, you simply cycle between red, green and blue showing the hot ones on red, the normal ones on green and the hot ones on blue? Sounds good? Feel free to criticize this idea. Also, if you want, I can make some previews of what the screens could look like.
 
Last edited:

manuka

Senior Member
By chance my UK born dad was a marine engineer of many decades service,much of it during WW2 on the likes of the celebrated Aug. 1942 Malta Convoy - his ship on this run was the lucky MV "Brisbane Star". As a kid I spent quite some time on NZ coastal voyages with him, and well recall the "hot bearing" issue. During WW2 it was normal practise to have lowly paid crew (usually Indians used to the hot menial conditions) doing little more than specifically monitoring key shaft bearings...

But fast forward - what sort (& age) of vessel is this? Retrofitting a network of sensors may be such a huge job that it'll be unrealistic & not cost effective. For starters,reliable thermal bonding to the bearing housing may be a serious engineering task. Given the agressive environment of an engine room,just running industrial strength wiring to the likes of numerous DS18B20 sensors alone cost $$$$$.

But you mention tedious IR gun readings- have you at least considered trials with a modern thermal imaging device? These would massively improve productivity I'd say, especially since readings can be stored in the unit, c/w data/time stamps. FWIW the FLIR i3, retailing for ~US$1100, is currently considered the best entry level IR imager. From a suitable vantage point you could thermally survey numerous bearings & hot points, almost at a glance. Locate your misplaced hot drink as well ! Stan.
 

Attachments

Last edited:

bullethead67

New Member
Floorplan

Here is a floorplan. the bearings are the squares going down each side. it only shows 26 the missing one is at the engine. The fwd ones just arent on the plan...... but they exist i swear.

Picaxe bearing temps.jpg
 

bullethead67

New Member
But fast forward - what sort (& age) of vessel is this? Retrofitting a network of sensors may be such a huge job that it'll be unrealistic & not cost effective. For starters, reliably thermally bonding the sensor to the bearing housing may be serious engineering task. Given the agressive environment of an engine room,just running industrial strength wiring to the likes of numerous DS18B20 sensors alone cost $$$$$.
Believe it or not the company is fine with giving wire. It takes forever to get a spool but theyll give one

FYI Boats 15 years old
 

Paix

Senior Member
Interesting indeed:
I'm unfamiliar with the specifics of your environment, but have the general idea having watched lots of movies :).

What constitutes a HOT bearing and why would a bearing go cold?

I do I assume that sensors would need to be mounted in tubes bonded to the bearing housings with something like a metal filled epoxy filler? Obviously big bearings, what are the mounting opportunities?

Following this with great interest. Something that can be rolled out incrementally to demonstrate it's worth and gain acceptance/funding.

Good luck with your worthy project.
 

bullethead67

New Member
And not discontinue the existing idiots? By 'discontinue' I mean sack and not kill!

You could use the LCD to show heat status instead of temperature. Three dots of 2x2 pixels each could be used for each bearing, or you could use a GLCD with RGB backlight and when some get too hot or too cold you flash the backlight red or blue and at the same time show the numbers of the bearings in question. If some are too cold and some are too hot and some are normal, you simply cycle between red, green and blue showing the hot ones on red, the normal ones on green and the hot ones on blue? Sounds good? Feel free to criticize this idea. Also, if you want, I can make some previews of what the screens could look like.
actually yes i would like to see a preview. its hard for me to display it in my head being new to this.
 

bullethead67

New Member
Hmm i seem to have lost the post i wrote?

ill redo it
bearings dont go cold but i would have them display blue when below 50c. This way if i was underway and a sensor came off it would display blue. that would get checked out.

also having a LED always lit would instantly show if one is burnt as it should always display a light. out = bad

I was hoping to mount on the surface with magnets. those little neodymium magnets are strong. And they can be removed easily when overhauling a bearing.
 

MPep

Senior Member
@Bullethead,

I suggest you look up RS485 or RS422 specs. For cabling, I would recommend NMEA0183 cabling, shielded twisted pair. Are you looking to have a seperate power supply for each sensor or supply from the 'display' station?

You might like to look at Jurjen's two-wire system as per this link. It should prove viable for your purposes. This system provides power downstream, and sends data upstream.
A BIG bonus is that the hard work is done for you!!!

Incase you want to monitor the data stream on a computer, each DS18B20 has its own unique serial number so that each sensor can be individually queried.

Hope it all works out well for you. Ask questions, buy some 08M2's and some DS18B20s and it will soon start making sense.

MPep.
 

Dippy

Moderator
Lovely project.
Will you have to cost this and then present it to the Company / Ship Operators?
And if they say "yes" then they hand you a wedge of cash for the build and installation?

I wouldn't display individual dots on a GLCD as it too small.
I'd go for an LED method (+flashing) and simply use the GLCD to give extra info upon a) Error condition or b) User request.
I'd also have a serial plug-hole at the master to allow more detailed interrogation with a Net/Notebook.

If I were doing this I'd get the Master to poll the slaves for a response and then you could get warnings of individual failures.

Look into the specs suggested by MPep or , at least, use some clever driving at each node.
I don't think I'd worry about DS's serial number as you can generate a more meaningful one with each PICAXE.

I would look into parity/error checking. It sounds tricky but is easy.

I'll leave the physical connection of sensor--bearing to you ; just remember the effects of temp cycling, vibration and shock and humidity.

You'll also have to investigate connectors, enclosures and cable types in that environment. Quality = expensive.

I'd cost it and get their cash and permission before going too far.
 

hippy

Technical Support
Staff member
I'd cost it and get their cash and permission before going too far.
I agree there. At least get enough money so you can prototype a one, ideally two, sensor system with another PICAXE as the display driver, or do that as a pre-requisite anyway. Despite being a big job all-in it's a pretty low-cost prototype.

I can't see that it won't work but best to try it before committing to a full build which then doesn't work.

Take a look at kranenborg's "two-wire data and power network" as this seems ideal to use here -

http://www.picaxeforum.co.uk/showthread.php?7694-quot-SerialPower-quot-true-two-wire-data-power-network
 

nick12ab

Senior Member
actually yes i would like to see a preview. its hard for me to display it in my head being new to this.
This is what it would look like for my first idea but as Dippy suggests the small size would make it a problem on a GLCD but if any are wrong the backlight colour could be changed. It does still look like it would be quite confusing for the 'idiots' you have described.



Previews for my second approach:



My second approach alternates between the three backlight colours of an RGB-backlit GLCD showing which bearings are hot, OK and cold. Bearings are numbered 1-F in this preview but it would make more sense to be 0-F. A number being present on the appropriate colour indicates status. It should cycle between each one every few seconds, omitting the status screens that have no bearings on them.
 

hippy

Technical Support
Staff member
It does still look like it would be quite confusing for the 'idiots' you have described.
I think it could be quite confusing for non-idiots ! The second set are much better, though in practice there should be far fewer ever too hot or too cold at any one time which opens up other options.

I'd go with the RGB LED idea per sensor though I'd have static green for ok, flashing red and blue to draw attention to a fault and that can have multiple rates of flashing to indicate severity of the error. That's useful with RGB LED's at the sensor itself as it's usually easier to see and understand something flashing than static.

Then add a simple text LCD at the master display which can scroll vertically a list of sensors too hot or too cold, shown in priority order, too hot first then too cold, those which have been hotter for longer first etc. That also allows a more meaningful description to be given to the sensors so there's no doubt where it and failure is. I'm guessing the RGB LED's will be placed on a deck outline but perhaps not.

An interesting question is what happens with sensors which go too hot or cold then normalise themselves. This could be missed but it may be worth adding a flag so the red or blue blips to show a failure had occurred at some point. That would need some 'clear' button to be added.

Most real world projects have at least two aspects to them; the actual engineering and the ergonomics of the human interface side. The thing most people hate is a perfectly engineered solution which doesn't fit the human way of things. That usually arises through the engineers deciding what the user will get rather than asking them what they want.
 

manuka

Senior Member
15 yo - but what sort of vessel size/crewing/sailing conditions/voyage duration are we talking about please. $$$$ costs are considered near incidental if serious berthing charges or delays arise...
 

John West

Senior Member
And of course, a really loud buzzer!
Nick has an excellent suggestion there. While idiots are typically attracted to brightly colored light, (I know I am) an audible alarm for any red or blue indication would get a quicker reaction to the problem. Assuming this is located in a noisy environment, there are some low cost very loud alarms available to do the job.

As for an additional LCD readout, I'd suggest using the "Big Number" sort of display style and sequentially flashing the output of each 'over/under temperature' bearing number and the temperature associated with it. That keeps the info in a simple display format. Hopefully there would be only one at a time flashing on and off, but this format would ensure that more than one 'out of range' bearing would still get displayed without cluttering up the display and confusing the folks looking at it.

This may seem like a lot of additional gear added on, but it's easy to do and would only constitute a very small part of a fairly large project and it provides useful additional info to anyone capable of reading it, so it makes sense to implement it.
 
Last edited:

hippy

Technical Support
Staff member
An interesting aside - RGB LED's are available in two type; 4-lead R+G+B+Common and 6-lead with two blue LED's R+B+Common / G+B+Common - The later may prove useful in some multiplexing / charlieplexing way to minimise pin count but I haven't worked out how or if it would work.

It would seem more useful for each separate node to use the smallest 08M2 there but I think any solution would still require 3 pins so maybe no gain there.

Thinking on how to do a node on an 08M2 with a local RGB LED, it needs 3 outputs for LED, 1 out for serial, 1 in for serial, 1 bi-directional for DS18B20. The 08M2 has 2 inputs, 1 output and 3 bi-directional. That's going to require some pin sharing.

I expect some flicker would be acceptable if Serial Out shared a LED, maybe green so it can also be used as a comms 'it's alive' heart-beat as well - the old trick of turning a limitation into a feature :)

That should work even with a multi-drop Serial Out bus but would require some thinking about so it didn't mess up the bus comms.
 

Dippy

Moderator
Hippy has nicely expanded on my own thoughts.

Brief (or accidentally ignored) excursions should obviously set a flag and time data. As would system failures or fault flags.
A good design would have a pile of memory so that things can be viewed post-event.
This harks back to rolling data-logging techniques - then you can make your Bearing Black Box.

I wouldn't try and cram too much on a single small GLCD.
Looking at Nick's first GLCD image I would have to bracket myself as an Idiot as I haven't got a clue what it means.
I would use the GLCD to display supplemental info , not the only info. LEDs are far more reliable than cheap GLCDs.
If you want to display it all real-time then a bigger/better/expensiver display and front-end is needed.
For ease a bigger version of the picture you posted printed on a large panel with LEDs would be very clear, plus a GLCD and a few buttons so that interrogation could be done by non-Idiots.


I don't quite follow John's suggestion with the bgg number display... does this mean I'd have to wait 5 minutes for a particular bearing info to come around?

I think you should keep your LED driving well away from the comms. Complication increases cockup factor... especially when the projecteer is not entirely experienced.


You'll get a lot of replies and suggestions here because this is very interesting , but firstly find out if this project can be properly funded and whether you would be allowed to do it.
If the answer from the ship operators is either "we'll give you a tenner" or "don't start putting home-made stuff in our boat" then we can all go back to sleep again :)
 
Last edited:

John West

Senior Member
I don't quite follow John's suggestion with the bgg[sic] number display... does this mean I'd have to wait 5 minutes for a particular bearing info to come around?
Nope. The only values displayed would be for bearings that were out of range. If there are ever more than one or two, there are likely much larger problems to deal with. With a four or five second display of any out-of-range bearing's number and temperature, any out of range bearing and it's temperature would be quick and easy to identify.
 

SAborn

Senior Member
Would a simple thermistor be good enough for a sensor as there is some nice encased thermistors made that would allow ease of mounting, It should also allow for some reasonable long cable runs from the sensors to, there is nothing worst than having individual circuits at each bearing location, and is far better to wire to common control points.

In reality we are only talking a 3 point sensor level here and i dont see the need for the DS sensors, also are the DS sensors not rated to only 100c max, so a faulty bearing might kill a DS sensor.

There is also good ADC chips of various types like I2C that would allow for several master picaxe chips to collect the data from the thermistors and send it via serial to a display master.

I also agree with Dippy that a LED panel would be far better than a LCD display, any led indicating other than a green light would require a instant physical check by maintenance so there would be little need for displays of data.

I would not use neo magnets for mounting as all neo's come with a temp rating and once the temp rating is exceeded they loose their magnetism and do not recover, in most GP neo's this is around 60-80c, also neo's have gone through the roof in price of late with a hugh increase in price.
 

Dippy

Moderator
My data sheet shows +125oC for DS, though I would also consider alternatives eg. thermistor or LM35.
But that's the easy part.

I can't see I2C being reliable over that distance, I would use a more robust method where the programmer has more control as discussed earlier.

And I certainly wouldn't rely on magnets for the mechanical mounting. Asking for trouble.


John; oh I see.


Another worry has crept in; with all these posts the OP will get confused :(
 

bullethead67

New Member
An interesting question is what happens with sensors which go too hot or cold then normalise themselves. This could be missed but it may be worth adding a flag so the red or blue blips to show a failure had occurred at some point. That would need some 'clear' button to be added.
Normalize? ok thats a new word for me what is normalizing?
 

vttom

Senior Member
Here's a thought on the display... What about a "strip chart" type of display? Each sensor would be represented on a plot, with time the X axis and temperature the Y axis. You could have 3 zones in the strip chart which are color Red, Green, and Blue. Under normal conditions, all the lines would be in a clump in the green zone. This would make it easy to see excursions into the Red or Blue zones.
 

SAborn

Senior Member
I can't see I2C being reliable over that distance, I would use a more robust method where the programmer has more control as discussed earlier.
I did not intend for I2C to be used over a long distance, but more so the thermistors wired to the I2C chips located near the picaxe circuit, this then utilizes a ready made bus system and only 2 picaxe pins needed for many sensors.

I have used thermistors with 10m of cable without problems but dont know what the max lenght of cable run permitted for a thermistor before problems might be encounted.

Selecting a sensor might be the easy part but it is what also dictates how the entire system is designed, so i would think it to be a important part to establish first up.
 

bullethead67

New Member
OK a few answers

I like the serial idea. is it compatable with PICAXE.net (in case i need to do plan B)
I am supposedly flying back home on the 7th. Ill have 30 days to gather what i need ( and try to enjoy a little life). any suggestions. fyi no shopping is available here. requisitions take 4 weeks minimum. if you can even get the dock.

The floorplan i posted is a file that would be edited and cleaned up for a panel.


OLED panel
. i was not able to completely understand how it worked within 5 seconds without reading the instructions. so it wont work for a oiler. mabye an engineer but this is not for him.


A Really loud buzzer
Absolutely not! One thing that is forgotten by many designers is their equipment is not the only one on a boat. Did you know everything here has a alarm.... everything. most dont have a ability to turn them down or off. No thats not a good thing either. Ever tried to back a ship into a slip with less than 1 meter on each side in heavy current. Then the damn alarms start going off. And they are loud. I cant hear the VHF. Things like "Youre coming in to fast" "Current has the bow bail out" "loss of Hydraulic pressure port side" ...... its the little things that get missed. Sometimes just displaying in a prominent location is enough. Oh and the panel would be in a sound proofed control room that is in the engine room

You'll get a lot of replies and suggestions here because this is very interesting , but firstly find out if this project can be properly funded and whether you would be allowed to do it.

I am gathering info now and will have to present something to see if i can get funding as i dont want to pay out of pocket.
If they dont go for it I have a plan B.

PLAN B = The Port main is a 16 cyl diesel that has no monitoring of exhaust temps. but somehow whenever the heads were last changed they installed new heads with the sensor installed!! its just sitting there. This project i would pay for but would keep it portable like on a PICAXE.net so i can keep it.

ASK permission? no... ill ask for parts

but what sort of vessel size/crewing/sailing conditions/voyage duration are we talking about please. $$$$ costs are considered near incidental if serious berthing charges or delays arise...

73 meters in length
14 crew
Operate within 10 meters of offshore structures when vessel is able due to weather
Stay offshore in any weather except hurricanes
Come in for supplies and repairs
Berthing is a a premium not price space available is near zero. everyone is willing to pay. everyone is fighting for a spot.
Delays are constant and incurr penalties if the vessel is unable to perform as chartered
none of this means anything to a guy who sits in a office on land and gets to approve or deny
They are not rational.
 

KingsJester89

New Member
Are all/most of the bearings visible from any particular area of the deck?

A cheaper and simpler option might be to attach a Picaxe, temperature sensor and three high powered leds (red, green and blue, colour coded as you said) somewhere on the bearing, roof or walls. Anyone checking the bearings would simply have to glance into the room and check all bearing from a standing location. Another benifit would be minimal interference to the ship as the only wiring required would be for the power supply. You could also use a colour cctv camera and monitor to view the leds from anywhere on the ship.

Although it would require more work, instead of using picaxes a cheaper solution might be to use op-amp chips as Voltage Comparators. Using a thermistor and resistor to make a voltage divider on one pin and a potentiometer on the other they could be adjusted to a range of temperatures. although may not be as reliable as Picaxes.

If you consider this idea you would probably want very bright leds (I would guess), Sparkfun electronics carry Triple Output High Power RGB LEDs:
http://www.sparkfun.com/products/8718
 

hippy

Technical Support
Staff member
Normalize? ok thats a new word for me what is normalizing?
"Returning to normal", such as going back from too hot to normally expected operating temperature. Apparently we can spell it with a z in the UK as well !
 

bullethead67

New Member
"Returning to normal", such as going back from too hot to normally expected operating temperature. Apparently we can spell it with a z in the UK as well !
yeah i really should utilize the spellchecker more. but its the begginning of the month and the 1st is a heavy paperwork period
 

nick12ab

Senior Member
As for an additional LCD readout, I'd suggest using the "Big Number" sort of display style and sequentially flashing the output of each 'over/under temperature' bearing number and the temperature associated with it.
This big?
This big?
or this big? (I wonder how much these cost?)

I think it could be quite confusing for non-idiots ! The second set are much better, though in practice there should be far fewer ever too hot or too cold at any one time which opens up other options.
Yes, the first one does look much more confusing once it's small rather than massive in the image editor. An improvement to the second one could be to use a colour LCD/OLED and show the status of them all at the same time.
 
Top