FridgeFreezer Posted June 30, 2015 Share Posted June 30, 2015 So this is going to be a bus-based gauge rather than usable on any old Land Rover / kit-car? That's a real shame. Quote Link to comment Share on other sites More sharing options...
robertspark Posted June 30, 2015 Share Posted June 30, 2015 It sounds like it will be both as the pins can be configured for both as mentioned earlier Rob Quote Link to comment Share on other sites More sharing options...
simonr Posted June 30, 2015 Share Posted June 30, 2015 So this is going to be a bus-based gauge rather than usable on any old Land Rover / kit-car? That's a real shame. Tut Tut - Mr Utterage, you have to pay attention in class! It's going to be an ANALOGUE gauge with a WiFi interface for configuration and to talk to an ODBC WiFi dongle plugged in to your OBD Port (if you happen to have one). The advantage of reading the OBD port is that you don't have to replicate the sensors already fitted (lowering the cost). However, and by connecting it by Wifi you leave more Analog inputs for other sensors - and you can have both if you want! The OBD / WiFi dongle will simply be an add-on if you want it. It will NOT be a CAN Bus device. [Though, since it will be open source, if anyone wanted to add a CAN interface, it is perfectly possible] Si Quote Link to comment Share on other sites More sharing options...
FridgeFreezer Posted July 1, 2015 Share Posted July 1, 2015 I told you I was a luddite OK my bad for not paying attention, I kinda tuned out for all the TLA's. As I said, I've been hankering after something relatively cheap and simple, the power being in programmability, all this wifi/bluetooth etc seems to add a hell of a lot of work. You could do away with buttons on the front by just using an IO pin on the back, if the user wants to wire a button to it they can, or they can configure it once and then have a clean front. You could even use a resistor-chain (a-la many steering wheel controls - bling!) or keypad matrix to save IO pins. Steering wheel mounted controls for your customisable gauges would be rather groovy There is one land-rover specific option that works nicely for an LCD, as demonstrated by TSD many years ago: Quote Link to comment Share on other sites More sharing options...
simonr Posted July 2, 2015 Share Posted July 2, 2015 I don't know why it's so hard to take photos of this display - camera refuses to focus on it properly! This was taken with the flash and the display on max brightness - and it's still readable This was taken with the display set on minimum brightness in the dark It's just showing random data. The display uses a progressive update to reduce flicker (though the refresh rate is still 100Hz) which explains some of the artifacts you can see in the images - where some bits of the screen have been updated & others not quite. To the naked eye (with persistence of vision) you cannot see them. These are my ideas for different screen layouts - if you have any other ideas / suggestions I'm all ears! Si Quote Link to comment Share on other sites More sharing options...
CwazyWabbit Posted July 2, 2015 Share Posted July 2, 2015 Try manual focus and a longer exposure Si Quote Link to comment Share on other sites More sharing options...
Mo Murphy Posted July 2, 2015 Share Posted July 2, 2015 Nice idea Si, this is long overdue and you suggested read outs look good. Mo Quote Link to comment Share on other sites More sharing options...
FridgeFreezer Posted July 2, 2015 Share Posted July 2, 2015 With a graphical display I'd have to stick some easter eggs in, like "Terror alert level in muppets" or the HaaH improved / extended pain scale On a more sensible note, something that looks close to a standard analogue needle gauge would be groovy. Also perhaps a "minimal" display that shows almost nothing while the reading is within limits and only expands when there's something of interest. Other ideas could be nicked from Megaview / Megatune / Megatunix / TunerStudio, for instance the rolling graphs are a good way to spot transient events and trends. If the gauge has more than one input, an x/y display (and any number of variations on that) could be rather useful. Also on the dual input thing: sticking a coolant sensor on the inlet & outlet of your rad could be an interesting exercise, especially if the gauge has an output or two that could control a cooling fan or EWP. Quote Link to comment Share on other sites More sharing options...
Anderzander Posted July 2, 2015 Author Share Posted July 2, 2015 Of your displays - the top left one would be good if it was configurable to pick up different sensors. Quote Link to comment Share on other sites More sharing options...
David Sparkes Posted July 2, 2015 Share Posted July 2, 2015 If I've understood correctly, it may well turn out there are two 'styles' of meter inputs. The primary one, that I understand Simon is working to, is based mainly around the outputs from the OBD socket. These are preset by the manufacturer. In this case it's slightly misleading to talk about 'different sensors', because although physically separate sensors are sometimes involved, other OBD outputs are fed directly from the ECU concerned. I do recognise there will be provision for a limited number (1?) of inputs outside the ODB pathway. The secondary style will involve the user installing additional sensors (which is why I'm uneasy using the term sensor with the primary style). Secondary style sensors could involve a K type thermocouple for EGT. Thermocouples could be used again for intercooler inlet and outlet temperature measurements, although resistive sensors might be adequate. The same comments apply to coolant or gearbox temperature measurements. There is a notable additional cost, fabrication, and time investment when you start adding additional sensors. I confess I have some history when it comes to monitoring engines, but back in 1995 the information via the standard diesel OBD was very sparse. Rovacom did the best they could, but I built a multipath interface, coupling inputs directly into the engine ECU, with the outputs being fed via Picoscope to my on-board laptop. The EGR thermocouple was an exception, the interface output went directly to a multimeter, strapped to the dashboard. I did learn that it was impossible to drive the car (sensibly) AND monitor the outputs at all closely. The fact that I had to use a laptop as the display interface for both the Rovacom and the Picoscope meant I could record data for later analysis. I think I'd only be interested in a 'multi-super-meter' if I could record the data it was monitoring. I'm not sure if directly onto a memory card would be adequate, having the option of direct input to a PC / tablet hard drive might be better. I'd regard USB as simpler, but fashion might dictate a wireless link. :-) Regards. Quote Link to comment Share on other sites More sharing options...
GW8IZR Posted July 2, 2015 Share Posted July 2, 2015 I don't know why it's so hard to take photos of this display - camera refuses to focus on it properly! P1000668.jpg This was taken with the flash and the display on max brightness - and it's still readable P1000669.jpg This was taken with the display set on minimum brightness in the dark It's just showing random data. The display uses a progressive update to reduce flicker (though the refresh rate is still 100Hz) which explains some of the artifacts you can see in the images - where some bits of the screen have been updated & others not quite. To the naked eye (with persistence of vision) you cannot see them. wp_20150702_01_11_45_pro.jpg These are my ideas for different screen layouts - if you have any other ideas / suggestions I'm all ears! Si Its sunny here in Oslo so it made me think about how does the display look when looking through polarised sunglasses? For none essential information if you cant see it its not such a disaster but if it were a coolant loss or similar it might be? Oled displays on mobiles are a bit poor when using some sunglasses Its not usually a problem home in Anglesey as its generally not sunny :-) Quote Link to comment Share on other sites More sharing options...
robertspark Posted July 2, 2015 Share Posted July 2, 2015 You could maybe log directly to SD card using SPI interface... If the board has it, at the cost of a pin as the mosi, miso, and sck are not normally used for anything else, which would also be useful if you were to use SPI thermocouple chips (MAX31855) Quote Link to comment Share on other sites More sharing options...
simonr Posted July 2, 2015 Share Posted July 2, 2015 OLED Displays do not use polarisers - so you will be able to see it with your sunny's on. David - you are pretty much spot on. There are (currently) 4 analogue inputs, 2 digital plus serial & I2C brought out on the connector on the back in addition to 2 digital outputs. I see where you are coming from that the OBD inputs are not strictly speaking 'sensors' - though in the context of a gauge, you can treat them as if they were wired directly in to the meter. It would be fairly easy to connect an SD card or a PC to the interface on the back for logging - but I think that would be a later 'add-on'. I agree that you cannot usefully scan all the readings while you are driving. My intention would be to leave the meter on one reading - RPM say and set high and / or low triggers on the other inputs I'm interested in. If one of the readings goes out of range, the display will change to that (with a flashing border). If more than one is out of range, cycle through those. It incorporates two capacitive touch sensors below the display which I'm going to use as a swipe interface to scroll through the screens. Although this is not a tactile interface, it does not need your fingers to locate buttons, just be in vaguely the right place. Thus, it should be useable even when you're moving (though I might suggest not). I've 3D printed an enclosure. There is a collar which screws on to the body of the gauge to secure it in to a dashboard. 15 Way D connector + USB port on rear. Quote Link to comment Share on other sites More sharing options...
robertspark Posted July 2, 2015 Share Posted July 2, 2015 Is it all 3d printed, nut + threads too (new at 3d never gave it much thought before)? How long did it take to print? Looks good Rob Quote Link to comment Share on other sites More sharing options...
Mo Murphy Posted July 2, 2015 Share Posted July 2, 2015 Early days I know but could the display be green to match Land Rover back lighting and would it dim when the lights are turned on for night time driving ? Mo Quote Link to comment Share on other sites More sharing options...
simonr Posted July 2, 2015 Share Posted July 2, 2015 Is it all 3d printed, nut + threads too (new at 3d never gave it much thought before)? How long did it take to print? Looks good Rob The print took 4 hours or so. Si Quote Link to comment Share on other sites More sharing options...
discomikey Posted July 3, 2015 Share Posted July 3, 2015 Early days I know but could the display be green to match Land Rover back lighting and would it dim when the lights are turned on for night time driving ? Mo the dimming could be easily programmed in. Si, what about adding an MPG gauge onto it, instant and average? Quote Link to comment Share on other sites More sharing options...
CwazyWabbit Posted July 3, 2015 Share Posted July 3, 2015 the dimming could be easily programmed in. Si, what about adding an MPG gauge onto it, instant and average? You can add anything, providing you also add the correct sensors! So for MPG you need a way to meter the fuel Quote Link to comment Share on other sites More sharing options...
simonr Posted July 3, 2015 Share Posted July 3, 2015 I'm not sure you want to know your MPG! Trust me - it's frightening! You need two flow meters - one for the return line too then the difference is the amount the engine has used. Other than that - it's easy. Si Quote Link to comment Share on other sites More sharing options...
robertspark Posted July 3, 2015 Share Posted July 3, 2015 The print took 4 hours or so. Si Si sorry OT, but is that normal and a limitation of the deposition rate required or could you use a better extruder and faster drives? Rob Quote Link to comment Share on other sites More sharing options...
CwazyWabbit Posted July 3, 2015 Share Posted July 3, 2015 That sounds quite fast to me Rob based on what others with 3D printers have told me. Quote Link to comment Share on other sites More sharing options...
simonr Posted July 3, 2015 Share Posted July 3, 2015 Si sorry OT, but is that normal and a limitation of the deposition rate required or could you use a better extruder and faster drives? Rob Mine (Ultimaker 2) is one of the fastest filament printers on the market. But the faster you print, the worse the quality - so I tend not to run full speed. This was on about 50% - but for really high quality, I'll run 10 or 20% on 0.02mm layers but then this would have taken about a week! Si Quote Link to comment Share on other sites More sharing options...
robertspark Posted July 3, 2015 Share Posted July 3, 2015 but what is the limitation, the amount / speed of plastic you can melt or the speed of the drives or the time it takes to cool the plastic? If I was to build an attachment for the plasma, would I be looking at a £300 all metal extruder being really fast and good or are they all limited much the same? (drive speed I have, so there is not a problem in that respect). z-axis travel would be a limiting factor at ~ 75mm but I could always change that and lower the bed a little and have a longer z drive for plastic. (sorry hijack). Back on topic (I can type on a laptop, PIA on a phone....), you could use a rough algorithm to work out fuel consumption, it all depends on how accurate you want the answer.... given Si is using ODB data via the k-line speed is sometimes available (car dependant I guess) as is injector trim, and if you program in the injector flow rate you could probably get fairly close, and then just time average the readings on a rolling average.... without k-line / OBD2, if you know (or could estimate) the injector flowrate, and you can tap into a speed sensor (use the range rover inline pulse module for the LT230) and you know or could log the MAP you could then using the stoichiometric fuel equation work out the approximate MPG. (I have a spreadsheet I did many years ago with all of the formulas in that may be of help if someone wants to explore it, suggest this is probably best an after market customisation development though) Rob edit-> extract When it comes to speed and accuracy, the Ultimaker 2 appears to be in a class of its own among the fused-filament fabrication printers. Make magazine called the Ultimaker Original the "most accurate and fastest 3D printer" in November 2012, and the Ultimaker 2 is capable of the same performance with a slightly larger build volume. It can print from 30-300 mm/s at a layer resolution of just 20 microns (0.02 mm) within a 225 x 225 x 205 mm build volume. By comparison, the Makerbot Replicator 2 has a minimum layer resolution of 100 microns (0.1 mm). so, a 50mm dia can shape, ~ 75mm long, with a wall thickness of say ~3mm would have a volume of ~42,363mm³ / 0.02 (layer resolution) / 1mm (bead width.... [presume??] / ~150mm/sec (~50% speed) / 60 seconds / 60 mins = ~3.9 hours.... or at ~60mm/sec (~20% speed) = ~10 hours.... (or ~20 hours at ~10% speed), wow... Quote Link to comment Share on other sites More sharing options...
simonr Posted July 3, 2015 Share Posted July 3, 2015 The problem printing fast is the same as milling. If you want to precisely control the position of the extruder you need a very rigid frame. Although, unlike a mill, it doesn't take any force to push the extruder around, it does have mass and thus momentum & inertia. The reason the Ultimaker is so good is a combination of a rigid frame and one of the lightest weight extruders on the market - so it can accelerate fast. There are also issues with irregularities in the thickness of the filament. If you push it out slowly, the differences get evened out in the 'pool' of molten plastic in the heater. As the speed increases, the pool gets smaller and provides less damping. You can increase the temperature, but that has other undesirable effects - worse than the variations in flow. Like most things, it is a set of compromises. The balance of those in an Ultimaker in most peoples opinion is as good as it gets currently. Although 4, 10 or 20h sounds a long time, because it prints completely unattended you can just leave it working over night or while you are at work. Other than the desire to get out of bed to see how it's going - that works for me! Si Quote Link to comment Share on other sites More sharing options...
CwazyWabbit Posted July 3, 2015 Share Posted July 3, 2015 Is that really the pH of a lemon? Quote Link to comment Share on other sites More sharing options...
Recommended Posts
Join the conversation
You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.