SLC AFR display direct from IOx

[Rod S]

Jean,

That looks good.

Co-incidentally mine arrived yesterday,



Without the benefit of a comparison before I ordered I just went for red as that's usually the brightest and the red ones on my TechEdge displays are more than adequate.

Not so spectacular as they're not assembled yet let alone powered up....

Two questions,

Bus extenders ? As I know nothing about I2C (apart from how to spell it) and there was another thread on your forum very recently where someone was using them, I googled it last week (in case I needed to order any bits) and all the places I saw said 1m plus easily unless the speed was really high. But if it is neccisary, which chips do you recommend ? (as I'll need to include for them on the board my displays will be mounted on and also a revised carrier board for my IOx-OEM).

Brightness ? I knew from the datasheets that it could be controlled but hadn't realised you were intending to do so. What would be the control input ? My best guess would be a biased light sensitive resistor (or transistor) into a spare IOx ADC or were you just thinking of a manual input, or just fixing it at one of the 16 levels ?

EDIT - and re. the 5V, I was intending to give them their own supply as I assumed the current would be quite high and I'm also going to include 10 LED "bar" displays (from the SLC analogue output) as per Don's suggestion so my power consumption will be even higher.

Rod.

[jbelanger]

Rod,

I agree that the red is the safest bet and what I would have used. But having seen the other ones, I must say that I like the white (which looks better than the pictures show) and the yellow is also not bad. The blue is also nice (and stands out more in the picture than it does in person) but blue LEDs are too "bling" for me and my eyes can't quite get used to them for some reason.

The bus extender may not be needed but I like it for a couple of reasons. First, it isolates the CPU so anything that happens past the extender chip won't affect it. And the I2C bus pull ups on the CPU side won't have a (slightly) different pull up voltage from the other supply. Also, since each display has 10k pull up resistors that will result in a much lower combined pull up value and I would prefer to keep it to a higher value on the CPU side.

Most of these are quite minor, and even insignificant, reasons so you may choose to not use an extender since they are not that cheap. What I use is the P82B715 which is available in an SOIC-8 or DIP-8 package so they are quite easy to solder by hand.

As for brightness, the initial version will at least have the possibility to configure one value in the settings. The other obvious way is to use an ADC input and I was thinking that the most flexible way would be to have a user-defined set of threshold values and hysteresis for the 16 available brightness levels. This way it would be possible to use with a non-linear sensor (or a non-linear user preference for brightness) without having to calibrate it.

The current draw should be about 15mA per lit LED plus a few mA for the chips and pull ups. So my little test was way over the rated current for the MS regulator and it's not surprising that the voltage was going down; it was probably near a protection shutdown. The next tests will use their own supply.

Jean

[Rod S]

Jean,

The other thing I forgot to mention about using red LEDs is, as we all know, anything red on an A series adds a few BHP......

The white ones weren't listed in the UK and the blues ones were more expensive (although I do have a fondness for blue).

Re the bus extenders, yes, I forgot the golden rule about not taking any CPU pins outside the boxes. I guess it was an RS232 or CAN mindset where they are all isolated from the CPU whereas the I2C is direct to the CPU as well as the peripherals.
The chips you suggest are only £3.50 over here so well worth it for the protection.
And I hadn't realised the displays already had pull-ups (one thing I did read was pull-up values become important on long runs because of the capacitance of the connection).

I'll start sorting out a carrier board with it's own power and the bus extender chip and work out a way to fit one on my revised IOX-OEM carrier board (the IOx-OEM and SLCs will be in the same case anyway).

Rod.

[Rod S]

Jean,

One more question before I order the buffer chips, using figure 9 (page 10) of the datasheet as an example (but without the common 5V supply) do you think the optional ESD protection suggested is worthwhile ?

I'll use a screened cable as a matter of course, and it will only run just over 1m and all inside the cabin so nowhere near HT leads or anything else I can think of that might cause high voltages to be induced.

Alan seems quite keen on ESD protection for the SLCs themselves but the LSU cables do get close to HT in the engine bay so I can see the logic there. But it seems a bit OTT for a screened cable inside the cabin.

Rod.

[jbelanger]

Rod,

It can't hurt to have them but I agree with you that it is seems OTT in your setup.

Jean

[jbelanger]

I did a quick check yesterday in direct sunlight. The green display is totally washed out and unreadable. The most readable are the yellow and blue with the red and white being acceptable. The main difference in readability seems to come from the fact that background for the yellow and blue being light grey while the red and white have a darker background which gives less contrast.

You can see this in my pictures and the second one (lowest brightness) comes close to what you would see in direct sunlight with the highest brightness.

Jean

[Rod S]

Jean,

I know this is a long shot but I thought worth asking.....

Following on from a recent discussion on the TM forum where you explained to TD that the sensor temperature feature on Alan's latest SLC version (not the SLC-OEM, something completely different) is to monitor health of the LSU itself, would it be possible to include this with the SLC-OEM / IOx combination.

Nothing fancy like the adafruit LED displays but just to have it in the IOx log or possibly displayed in TS.

I see from the SLC-OEM technical manual, the number isn't calculated by the SLC but requires an external calculation from two data addresses on the SLC's I2C bus.

So is it at least possible to get the IOx to log those two values so I could calculate "after the event" in a spreadsheet ???

Obviously you know I have a fairly unique reason for being interested, hence why I said "a long shot".

Rod.

[jbelanger]

Rod,

I had been thinking about adding it since Alan does think it is a valuable thing to monitor (and I can see why). And after my post on the TM forum, I have been thinking how I would do this and integrate this with the AFR reading and display writing on the I2C bus. This is something I want to add and it will likely be logged as the delta between the 2 values since this is what is needed to compute the temperature.

Jean