5 Vref

http://jbperf.com/io_extender/index.html
http://jbperf.com/io_extender/tinyIOx.html for the TinyIOx

5 Vref

Postby Yves » Tue Feb 13, 2018 3:59 am

I learned that MS has a current limit on the 5 Vref. Can you make an additional 5 Vref circuit to supply the 5V to additional sensors ? Any problems with incompatibility when you do this ?
Yves
 
Posts: 93
Joined: Sat Feb 28, 2015 9:44 am

Re: 5 Vref

Postby dontz125 » Tue Feb 13, 2018 1:43 pm

Do you mean the MegaSquirt or the MicroSquirt? The big MS has a voltage regulator rated to 500mA and some pretty serious heat sinking. The uS has a regulator rated for a full amp, but has almost nothing in the way of a heat sink; B&G have recommended that no more than 25mA be used for VRef.

If you need to use an external power supply to add a few extra buckets of 5v milliamps to a uS or uSM, there are two ways to do it. You can use a following regulator that self-matches the output at the VRef header pin (and I *cannot* find an example of this online), or you could remove L1, and wire your regulator output to the L1 pad closest to the board edge. This allows the secondary regulator output to be read by the VDDA pin on the processor, to keep your sensor reading accurate. Don't just try to run it through the VRef pin on the header, though; the slight resistance - the *changing* resistance - of the polyfuse will confuse the VRef reading at the chip.

If you really are talking about an aluminum-boxed MS, you can do the same trick as with the uS, just remove L2 instead. With the bigger MS, there is also the option to simply uprate the regulator; there are 5v TO220 LDO regulators that go as high as 5 amps! If you go this route, you'll probably need to uprate both polyfuses, or else they'll simply block all your shiny new current.
dontz125
 
Posts: 199
Joined: Mon Feb 22, 2010 10:27 pm

Re: 5 Vref

Postby Yves » Tue Feb 13, 2018 3:07 pm

Ok, this is not as simple as I hoped apparently.

I use a V3.57 board, so it's a no-go as far as taking off parts and mounting others.

Any ideas what stuff like a hall sensor, map, iat, clt, iox etc use in terms of current ? My guess is that with everything I'm currently running I might be close to .5 A
25 mA looks like a very low number, considering that the on board map alone uses something like 7 mA.
Yves
 
Posts: 93
Joined: Sat Feb 28, 2015 9:44 am

Re: 5 Vref

Postby dontz125 » Tue Feb 13, 2018 4:04 pm

The V3.57 uses the same through-hole style voltage regulator as the V3.0; it's the left-most TO220 on the heat sink bar, beside the DB9. L2 is just above the boot header, and comes away easily with a soldering iron, some fresh solder and lots of flux.

The IOx processor is the big eater in your list, and I would say around 75-100mA, but the IOx has its own power supply. The rest ... I'd be astonished if it added up to 50mA. The normal sensors are 2-5mA, the MAP /Baro is 7-10mA. The Hall sensor depends on the model, but again, it shouldn't be more than a handful; the Littelfuse 55505 is rated for 20mA max. You should be nowhere *near* the 500mA limits of a V3.57 board. I don't know what the consumption is of the MS3 card itself, but the MS2 chip was IIRC close to the 100mA of the IOx.

Just to clarify - 25mA is the VRef limit of a MICROSQUIRT, which doesn't have an on-board MAP sensor. Most of the power it consumes goes to operating the controller itself, and doesn't have the heat rejection capability to process more current.
dontz125
 
Posts: 199
Joined: Mon Feb 22, 2010 10:27 pm

Re: 5 Vref

Postby Yves » Tue Feb 13, 2018 4:38 pm

To clarify I'm currently reworking my manifold and was planning on adding 8 map sensors (one for each runner) with the goal of easily balancing individual cyls and seeing what goes on when driving in terms of pulsations in the runners

8 map sensors would total 8 X 7 mA = .056 A

Currently I run the following :
Iox
iat
clt
2 X maf (frequency)
8 WB ego's
crank hall sensor
cam hall sensor
wheel hall sensor
tps
coolant pressure sensor
fuel pressure sensor
fuel pump speed regulator
Oil pressure sensor
Oil temp sensor
acceleration sensor
Yves
 
Posts: 93
Joined: Sat Feb 28, 2015 9:44 am

Re: 5 Vref

Postby dontz125 » Tue Feb 13, 2018 4:53 pm

The WBO2 sensors are going to be hogs, but aren't they and their controllers 12v? If these are 14point7 SLCs, they're either self-powered, or you've got so many they should be in a box with their own power supply! Same with the IOx; it should have its own on-board regulator.

The hall sensors will probably be 10-20mA each, the temperature sensors should be 1-2mA each, and the TPS 5mA. I have no clue what the pressure sensors draw; that's something to check no the data sheet.
dontz125
 
Posts: 199
Joined: Mon Feb 22, 2010 10:27 pm

Re: 5 Vref

Postby Yves » Tue Feb 13, 2018 5:19 pm

The WB's do have their own ps. They're 14point7 Unless mistaken the iox doesn't have an on board 5V regulator.

And Don the pressure sensors came from you remember...
Yves
 
Posts: 93
Joined: Sat Feb 28, 2015 9:44 am

Re: 5 Vref

Postby jbelanger » Tue Feb 13, 2018 5:26 pm

I think you have the full size I/O Extender and not the TinyIOx.

As Don says, the IOx has its own power supply. I also assume you've got some of those sensors connected to the IOx ADC inputs so you should be using the Vref (and ground return) from the IOx to power them. The IOx has the same type of voltage regulator as the Microsquirt but it has more heat sink capacity so I would not worry about adding 100mA of current from sensors.

I should also add that the IOx has been designed with the possibility of using an addition power supply for Vref. If that's something you want to consider, it is simply a question of removing a solder jumper and connecting the additional power supply to the correct pin on JP5.

Jean
Image
jbelanger
 
Posts: 3621
Joined: Sat Oct 03, 2009 12:24 pm
Location: Quebec, Canada

Re: 5 Vref

Postby Yves » Tue Feb 13, 2018 6:20 pm

jbelanger wrote:I think you have the full size I/O Extender and not the TinyIOx.

As Don says, the IOx has its own power supply. I also assume you've got some of those sensors connected to the IOx ADC inputs so you should be using the Vref (and ground return) from the IOx to power them. The IOx has the same type of voltage regulator as the Microsquirt but it has more heat sink capacity so I would not worry about adding 100mA of current from sensors.

I should also add that the IOx has been designed with the possibility of using an addition power supply for Vref. If that's something you want to consider, it is simply a question of removing a solder jumper and connecting the additional power supply to the correct pin on JP5.

Jean


Yes, I have the full size IOX. So if I'm right, I can put the additional maps on the spare adc's without worry.
I assume pin 3 is the one to use on the DB37 for power to the map sensors ?
Can I assume I can use the standard conditioning circuits as per your website explanation ?
Yves
 
Posts: 93
Joined: Sat Feb 28, 2015 9:44 am

Re: 5 Vref

Postby jbelanger » Tue Feb 13, 2018 6:31 pm

Correct on all points.

Jean
Image
jbelanger
 
Posts: 3621
Joined: Sat Oct 03, 2009 12:24 pm
Location: Quebec, Canada

Re: 5 Vref

Postby Yves » Thu Mar 22, 2018 3:23 pm

Jean, I have an additional question. I have a custom ini that you once wrote for me to eventually use the 8 map readings. Is there some way I can average all those values and use that as map input in my MS ?
Yves
 
Posts: 93
Joined: Sat Feb 28, 2015 9:44 am

Re: 5 Vref

Postby jbelanger » Thu Mar 22, 2018 4:17 pm

The custom.ini file only tells TunerStudio how to interpret, display and log the data on the PC. It does not send anything back to the MS or IOx so you cannot use it for this purpose. And there is nothing at the moment that would allow you to use an average of many analog inputs as data for the MS.

And while this sounds like a useful thing to add to the firmware, I'm not sure if it would actually give something reasonable without also having the possibility of sampling the data at fixed positions in the engine cycle. And that is a lot more complicated than computing an average.

If you could log and post something that would show what the simple average would look like that might convince me that even the simple average is useful enough.

Jean
Image
jbelanger
 
Posts: 3621
Joined: Sat Oct 03, 2009 12:24 pm
Location: Quebec, Canada

Re: 5 Vref

Postby Yves » Thu Mar 22, 2018 5:00 pm

jbelanger wrote:The custom.ini file only tells TunerStudio how to interpret, display and log the data on the PC. It does not send anything back to the MS or IOx so you cannot use it for this purpose. And there is nothing at the moment that would allow you to use an average of many analog inputs as data for the MS.

And while this sounds like a useful thing to add to the firmware, I'm not sure if it would actually give something reasonable without also having the possibility of sampling the data at fixed positions in the engine cycle. And that is a lot more complicated than computing an average.

If you could log and post something that would show what the simple average would look like that might convince me that even the simple average is useful enough.

Jean


I have to revise a couple of things on my manifold. As this is an ITB manifold and previously I used a vacuum chamber to which map was connected along with IAC, PCV and brakes, I get crosstalk between the cylinders. While I can solve that for the IAC, PCV and brakes with one way valves, I cannot do the same with the map lines. For that reason multiple maps connected to the runners would solve this in so far that I would have some way of averaging them or logging the minimum map. I understand this may be difficult/not within the scope of ms or iox.
Yves
 
Posts: 93
Joined: Sat Feb 28, 2015 9:44 am

Re: 5 Vref

Postby Rod S » Fri Mar 23, 2018 7:44 am

Why not just use a passive averaging circuit , ie, do it in hardware, just a bunch of resistors ?
I'd assume the IOx input impedance is high enough, worst case you might have to do a custom calibration or, second worst case, use an Op Amp averaging circuit (resistors plus Op Amp) which would "ignore" the IOx input impedance.

I looked at this a couple of years ago when I was thinking of average temperatures for another project and found plenty of circuits on the web. I never pursued it, so can't claim it works, but what I read seemed mathematically sound.
Rod S
 
Posts: 310
Joined: Mon Sep 06, 2010 4:03 am
Location: Rural Suffolk, England, UK

Re: 5 Vref

Postby Yves » Fri Mar 23, 2018 3:03 pm

I have just received a couple of pcb layouts for a circuit that would allow for 8 maps where the output would be a function of the lowest value. Still need to study it though.
Yves
 
Posts: 93
Joined: Sat Feb 28, 2015 9:44 am

Re: 5 Vref

Postby dontz125 » Fri Mar 23, 2018 3:26 pm

This is the basis of what I (once did and shall again) market as a QuadraMAP, with a more robust power supply and filtering. Adding more sensors really is just a matter of ... adding more sensors, each with its diode.

index.jpg
index.jpg (5.32 KiB) Viewed 1492 times
dontz125
 
Posts: 199
Joined: Mon Feb 22, 2010 10:27 pm

Re: 5 Vref

Postby Yves » Sat Mar 24, 2018 4:50 am

I somehow doubt this will do it's job of supplying the ecu with a good value to base it's fuel calculations upon. The ecu should base it's calculations upon an average of all cylinders which are measured with individual maps. Preferably this would be the average lowest value. In an ITB application the runners is only 1/4 of the time on vacuum. The rest of the time it's at atmospheric pressure. The circuit above pulls down the map value to that of the lowest value it finds in the various cylinders. The question is if this is correct fueling wise.

I found out that when interconnecting cylinders with a common line (whether that is for reasons of map, FPR or other) creates cylinder cross talk. If one runner is on vacuum and another at atmospheric pressure, it's going to draw air from the runner with the atmospheric pressure. In effect a sort of vacuum leak. It carries some serious side-effects and at the same time prevents the major advantage ITB's have and that is to isolate one runner from the other. Hence the reason I was looking into this. However MS is aimed at the majority of EFI users and those are not ITB applications.
Yves
 
Posts: 93
Joined: Sat Feb 28, 2015 9:44 am


Return to I/O Extender

Who is online

Users browsing this forum: No registered users and 1 guest

cron