TSD

Oh I agree there is *some* vacuum, (by definition, otherwise no air would flow in!) but I was only thinking out loud that it if it were enough to cause smoke only on the overrun because of poor seals, then the rest of the time you'd have 15psi+ blowing the other way into the rocker box, which would put a lot of oil mist into the separator and then into the air intake pipework, and a noticeably 'oily' engine. Even with the big turbo on the TGV engine, I never noticed any vacuum indicated on the boost gauge - one of the VDO ones for petrol engines, scaled in PSI on the boost side, and mbar on the vacuum side. Actually it does now indicate a little vacuum all the time, but that's because the boost spikes bent the needle smacking the end stop at 25psi I'm not sure what other mechanisms there are for overrun oil smoke (rings? turbo oil seals?), so I'm interested to see how much difference it makes. Hopefully Mikes right, especially as my shed is full of Ibexes at the moment, would have to do a bit of shuffling to get another one in there!

I always sort of assumed that wouldn't be the case on a diesel, since there's no intake vacuum to draw oil down the guides. But no practical experience either way, so

I think I know, for two reasons. One, I think you might have mentioned it recently, and two, I have a very similar looking tool in my workshop. Except I didn't need to replace mine, I just forgot to fit them at build

I don't know either! I can think of a few possibles, but my guess would be it's an overtemperature shutdown. As you know, overheat a permanent magnet motor once and it's game over. D3 (and I imagine other carswith similar compressors for air suspension) takes note of compressor run and idle times to keep the motor temperature under control, presumably for the same reason. Might be even more important when you've shut it in a fairly sealed box (as the D3 does also of course), and fitted fat power leads so there's no element of current limiting. I did quite like the other mods, more or less. My compressor in 1Bex has a similar-ish setup. It's plumbed to a small (1L?) pressure tank and manifold, with pressure switch and blowoff valve. It's remote to the compressor though, under one of the storage lockers, with airline fitting on the rear panel above the number plate. Pretty sure the compressor itself (Some name I've forgotten, from Oz) has a pad air filter as per the video, from the factory. I've never tinkered with the insides though, despite all the powdercoat falling off it still works fine after 16 years, best left alone at this point, in my opinion

I have no problem with removing the pcb, if you have reason to think it's going to improve reliability. But how can you have an opinion if you don't know what the PCB does? The problem is that even very smart people sometimes do epically stupid things, but dumb people almost never do unexpectedly smart things, so the game is rigged from the start

I watched as far as removing the pcb to improve reliability, after admitting he didn't know why it was there or what it did. Sometimes I wonder if Darwin got it wrong and it's all just a lottery.

That sounds how I picture a disco shroud would be. I can't remember the exact details, but I played with both shrouds when I made the engine mounts for my Ibex build. I think the fan sits higher in the disco shroud. Check some online pics, but from (my entirely unreliable) memory the defender shroud has the fan central (vertically), the disco one is offet upwards, and has a removeable section, where the defender one is all one piece.

This might be a daft question, but you are using a Defender fan shroud, not a Discovery one? The position of the fan in each is different, but I didn't see any mention in which direction it 'won't fit'.

I did see that, and I normally avoid anything listed as centre feed, but as said, you can just use it as a 'normal' roll. Some other makes I've had in the past the centre roll is either very thin, or not present at all, so the roll quickly collapses and becomes a useless mess. I mentioned this subject this morning to the guy who supplied my last pack from costco, and we've had 3 or 4 packs between us, and all have been useable as a normal roll. At one company, our supplier used to send a free Torq centre feed holder evey time we ordered rolls, but everyone just preferred a loose roll at the end of the bench, and the holders were usually thrown away unopened (or possibly taken home and thrown away after checking the price of rolls!!)

Tork is the brand name I've been trying to remember all day, and what I always used when I could expense the cost back to customers... https://www.costco.co.uk/Business-Office-Supplies/Cleaning-Care/Paper-Products-Tissues-Wipes/Tork-Plus-Centre-Feed-in-Blue-6-x-1575m/p/87840

The last 'proper' blue roll I had was sourced at Costco. Full sized rolls, hard paper, stable dyes. I was working in a friends lab and commented on his. Next time I visited he had sourced a pack for me. I think we split a pack of 12 rolls between us, but it's still pricey. I bought a half dozen rolls from Jawel Paints (since I was already paying shipping on a lot of cans of paint!) but the quality is pretty poor TBH. The rolls aren't as wide as the 'proper' size, they break up too easily, and worst of all the dye isn't stable with many solvents.

If you wanted to keep the counterbore small, then Reduced head / Countersunk rivnut would be less than 8mm head diameter (for M5), and nearly flush with the steel. You'll want the 'correct' drill size (rather than 'near enough'), and if you can't remove the melamine, you might have to get clever with the tooling as you probably can't get access with the rivnut tool head.

According to todays junk emails, it's World Metrology Day (how could I have forgotten?), so seems like a good time to post info on the third gauge version I've been fiddling with this week. This is a battery monitoring gauge, using one of the smart sensors fitted to a lot of modern cars to prevent embarassing failures caused by the Stop-Start systems. There's not really anything new here, others have reverse engineered the sensors before, and Hella will even sell you a box for your motorhome for about £150. But they don't seem to be widely known about, and I was curious to know what they can really do, and how well they perform. The intelligence of the system is all contained in a single housing, that replaces the ground lead terminal at the battery, and the additional wiring is very simple. It can directly measure battery voltage, current, and temperature. Current measurement range +/-250A continuously, and quoted as safe out to 1200A, so it should survive even heavy winching (though it might not be an ideal device in that situation because of the voltage drop - more testing to follow). In addition, the unit is a 'coulomb counter', it keeps a running total of energy into and out of the battery. From that information it predicts state of charge of the battery, and comparing measured performance against predicted, the State of Health of the battery. Because the real 'intelligence' is all contained in the sensor, all the gauge has to do is (other than initial configuration) poll the sensor for the info, and draw pretty pictures on the screen So here is what I have so far :- Couple of things to note - the 'analogue gauge' bits of the display are faked, just to reserve space of the screen, and both sets of digital readouts have the same info on them - that's because I only have one sensor connected at the moment. The main readouts are current and voltage, of course. The small value at the left is the 'State of Health' and on the right is the 'State of Charge'. I didn't bother to display battery temperature, as it's probably not very useful, and there's not much spare screen space! First tests have been very encouraging, the voltage and current readouts are surprisingly accurate, the current reading was well within the 0.5% spec for my Fluke 189, which is the best ammeter I had to hand. Over limited bench testing this week, it seems to track battery stored capacity well, though I'm limited to about 30A for both charge and discharge on the bench. How well it tracks in the real world is another thing of course, but since they are fitted to loads of modern cars I'd hope it could be reasonably useful. My design as it stands can support at least four sensors (so four batteries) but displaying the info would be fiddly. My use is for two batteries, so that is what I have aimed at. Now I've proved it can work, I will buy a second sensor and try to get it installed in 1Bex over the next few weeks. A few pics of the guts for those interested...

If I can scrounge up enough parts to build extras, then I will, no problem. But sourcing electronic components at the moment is madness, so no promises! You are right, I was only really thinking of configuration, where NFC works nicely (or did in the one project where I used it). Datalogging only in the sense of 'Max temp seen since last reset' kind of thing. Maybe some long term stats? I don't see much use for anything more than that, on a standalone box measuring only 1 or 2 things. Bluetooth would (for me at least) be overkill and not a good fit for the task (of configuration), though it tends to give the marketing department a moist moment. If it was designed as a datalogger, with loads of simultaneous input channels, then live values datalogging would be useful. Come to think of it, I did a design like that years ago, and it was even less interesting than the giant prototype sewage pump it was attached to Thankfully, that had a 3G modem instead of Bluetooth, so no-one had to go anywhere near it But anyway, this isn't a commercial product, it's just a bit of fun.

Like many (ok, almost all) of my homer projects, this one got designed back to front. When I found the smart oil pressure sensors, I wanted to play with one, and see if it was any good. The round displays have been sat in a drawer since before xmas waiting for me to even open the box. I really hate messing around with dev boards and loads of bodged wiring (at least for digits - it's the only proper way to do RF work!) and I already had a couple of PCB designs waiting to go out for manufacture... So by the time the ebay fairy brought the sensors, I already had a PCB design. But it annoyed me, because there was space left on it, so feature creep set in, and I started adding things until it began to look busy. All sketched on the back of an old envelope, only a few sums done, and very little detailed reading of a datasheet. I didn't write a single line of code until after I'd sent the PCB out. Not doing too badly for a prototype, only two cut/straps on the board so far, and only one glaring unfixable ****up with the pin assignments. Easily fixable in version 2, which will never happen. I have the battery monitor version up and running now, but no graphics written yet. Hopefully will have something to show in a few days. I've been clumsy, one of the displays has cracked so only half of it works. Hopefully the replacements arrive before I get bored and wander off to play with something else, leaving yet another unfinished project

Yes, but the typos were all my own work

The following is all from google and scribbling on a metaphorical fag packet, so a grown-up should check my maths. I used to run an LT230 with series speedo, but the wobble was so bad I have no idea how close the speedo was. I think the Series speedo exists in several different versions, according to build. There should be an identification number around the outside of the face, which ends with a number indicating the calibration. According to the internet, it indicates directly the revolutions per mile. I found values of 1376, 1408, 1504, 1536 mentioned, there may be others. That value will obviously affect your ideal gear. 33.1" tyres, so I make that about 610 revolutions per mile. 4.7 Diff, so the propshaft does 2867 revs/mile Speedo cable is driven by the output shaft, so LT230 ratio is unimportant. The output gear is FRC3162, which is (I think, from online photos) a 9T gear. Speedo drive gears are available as: FRC3310 / Blue / 20T / 90/110 with 235/85 or 750 tyres FRC3311 / Green / 21T FRC3312 / Yellow / 22T / 90 with 205 tyres FRC9939 / Black / 23T / V8 Middle East market vehicles FRC3313 / Red / 24T / LT77 Using the blue gear, 20T, the speedo drive is 2867 * 9 / 20 = 1290 revs/mile Using the Red gear, 24T, the speedo drive is 2867 * 9 / 24 = 1075 revs/mile So I think your best choice, without additional complication, would be the Blue gear, but even with the best speedo option from the list above, it would under-read by around 10% It is possible to get the speedo recalibrated of course, or fit a Td5 electronic speedo, or VDO universal programmable speedo (but the latter options lose the proper look, of course). EDIT : I'm wondering where the FRC9939 info in my notes came from, Google seems to deny it exists. FRC3313 has 23T according to the parts list linked in the original post. It's a typo when copied from :

Not my kind of pricing, but split cable glands are available e.g. https://www.automation24.co.uk/splittable-cable-gland-pflitsch-uni-split-gland-usg-22554-pc Overkill for something as leaky as a Land Rover I think. I'd aim for a nice snug fitting grommet, cut and glued, and either self amalgalmating tape or butyl body tape if I wanted to beef up the splashproofing.

That's quite funny - I've never seen a panel pilot before, but Lascar head office is in the next village

Agreed, in some ways I liked the Citroen approach in the 80s of replacing gauges with warning lights, and adding a 'master' warning light. (The first I was really aware of, though maybe they weren't the first). But I do like the 'trend' data learned from viewing gauges for weeks on end. It's useful to know if the temperature has suddenly shot up, or been climbing slowly all day, or if it's been marginal on hot days for months. Or, in the case of a tdi, there's no warning light at all because the gauge dropped when all the coolant fell out With most of our vehicles of course, it's 50/50 if the instrument is more or less reliable than the thing being monitored...

I like the piezo idea. Since they are inherently narrowband, I imagine it does most of the job of filtering rattly tdi dashboard noise before you even read the signal from it My 'tap the screen' sensor is just an ambient light plus IR sensor behind the glass - the same kind of thing that turns off a smartphone touchscreen when you hold it up to your face. It's simple to detect transient changes in IR level with a finger on the glass, and also measure visible light level to dim the gauge at night. Equally, the same sensor could detect a pattern on the back of a rotating ring control through the glass. If I was developing this as a commercial product, I'd put an NFC interface onboard to give configuration and datalogging via a smartphone. I'm as vulnerable to feature creep as anyone, but at its core I think the task is to provide the user with the required information, as clearly and legibly as possible. Ideally, I should be able to glance across at the instrument cluster and straight away know if all the pointers are in roughly the right place, without thinking about it. Freely admit I've broken this principle with multiple gauges per housing, and unnecessary digital readouts, and it will probably (certainly) get worse with the battery monitor gauge. I haven't even begun to think about how to handle all that. I will get the hardware up and running first and see where that takes me...

Aargh! Now I'm stuck - I tried to make it as simple as possible, no moving parts, so it could survive a few years in a wet landy. On the other hand, I really like that idea, and straight away I start schemeing ways it could be done As it stands now, it does have the option for a couple of external button inputs, but I haven't made any use of them so far. There is also a single 'tap the screen here' input option, but I've not even populated the circuit board for that.

In case I ever get to finishing the Ibex build, I've been thinking how to get all the instrumentation I want without loads of extra instruments. To keep things fairly simple I want to stay with only the common 3 extra gauges on the dash, but I always seem to have 4 or more things I want to measure. So as an experiment, I've tried making my own, just to see how it works out. The starting point was finding in rapid succession, some interesting sensors, and a source of reasonably priced round displays. With the current state of global electronics business (and my wallet) the design is driven more by what parts I have on hand, or can easily buy, than what I might have chosen for a 'clean sheet' design. There is a common 'core' circuit board design, with various optional components depending on the intended function. The one I'm playing with at the moment is combined EGT and boost pressure gauge. This uses a standard EGT thermocouple, and a boost pressure sensor from a Freelander 2, as it's cheapish, and it's nice to use LR parts where possible. The display module is bonded to the front glass (no annoying misted over gauges!) and looks a fairly uniform black face when powered off. I messed up the resin when assembling this one, so there's a visible smear on the lower left quarter which I didn't spot until it had cured overnight . Also, every electronic item must have a boot screen it seems, by some unwritten law. The first one I got working, though not yet built up into a gauge, is a combined oil pressure / oil temperature gauge. This is the one that sparked my interest originally, when I found a BMW smart sensor that provides very accurate values for pressure and temperature, from a single unit about the same size as the 300tdi oil pressure switch. This avoids all the plumbing adapters normally needed to fit both sensors, which usually make me nervous. The sensor replaces the oil pressure switch, and the gauge has an alarm output which provides the oil pressure switch function. The oil filter housing has to be redrilled and tapped, but as that's a bolt on housing, it's easy to do and nicer than using adapters. I've tried a few different graphics designs for the gauges, but the one above seems to be the most legible. The third gauge is intended to be a battery voltage/current monitor, and provide an output to operate a split charge relay. While it could (should?) be the simplest of the three, I've been waiting for parts, and have managed to bundle in a load of extra complexity, so while I've built the hardware, I haven't made a start on the software for this one yet. There are also options on the board for reading from most automotive analogue sensors, and canbus messages, though I haven't tested them and don't intend to as I don't have a use for them at present. The display seems easily readable in bright sunlight, and can be dim enough to not be annoying at night. Certainly less annoying than the nasty chinese radio/satnav. The viewing angle seems more than good enough too. The housing is 3d printed at the moment (and coloured according to whatever happened to be in the printer!), but is intended to be cnc'd aluminium in the final units. Probably something like Frigefreezers battery gauge housing

I don't think you really need to rerun that test, clearly the data you have can be represented pretty well by the polynomial shown. Any error looks to be small enough that neither driver nor ecu will notice. Comparing that curve to the one elbekko linked, they seem to have very similar gain (slope of the curve) at high flow, but the GEMS sensor seems much more sensitive around 1V where I'd expect idle to be. Maybe that's why they are problematic, not because the sensor drift itself is bad, but the high gain means any drift in output voltage represents a larger drift in indicated air mass. It might be interesting to test a bunch of 'failed but working' sensors to see what the typical changes in the curves are. I'm wondering if the change in the failed sensors is largely the offset voltage, so your curve above moves left/right, then that could be corrected by a very simple circuit. Unlikely, but I don't know what the real failure mechanism of the MAF sensor is.

I've not much experience with MAF sensors, but if you know the care and feeding of both known and target sensors, e.g. supply voltage, output type (voltage/ frequency/whatever), then maybe you could plumb the engine intake with both sensors in series, then just hoon around the neighbourhood datalogging both sensors at the same time? If they are analogue sensors, then I think you'd get enough of a curve to be able to fill the rest by eye. If both curves have similar shapes and ranges, then it could probably be made to work with either an analogue or digital 'converter' circuit. The arduino could be programmed as an analogue to digital converter, dumping readings to a laptop via usb. It's not the best resolution ADC in the world, but it should be sufficient to show if the idea has legs, before spending any cash on producing a perfect solution. I can think of plenty of reasons this wouldn't work, but not many reasons not to try it