It is currently Fri Jul 20, 2018 2:16 pm



Forum locked This topic is locked, you cannot edit posts or make further replies.  [ 37 posts ]  Go to page 1, 2  Next
Author Message
 Post subject: Practical and basic prototype ECU decisions...
PostPosted: Tue Jun 07, 2016 3:45 pm 
Offline
User avatar

Joined: Thu Mar 26, 2015 11:55 am
Posts: 112
Location: Massachusetts, USA
From the FJ20 thread, Mark says:
Quote:
Currently the FW uses the eTPU of PWM, so you can have any frequency you please.....and will work with the driver its attached to.


OK, that's interesting, I didn't know that. So some number of eTPU signals can be configured as PWM? I currently use 8 for coil and 8 for injectors. I currently had a couple of GPIOs going to a dual MOSFET on the CPU board, but I'm probably removing that in the current spin. I don't want power parts on the CPU board because of all the analog stuff.

So the tradeoff is saving pins on the CPU package by using the coil drivers for PWM or using CPU pins and also pins on the driver board connectors...

I have to check and see if the driver chips PWM is on the coil or injector pins...


Top
 Profile  
 
 Post subject: Re: Practical and basic prototype ECU decisions...
PostPosted: Wed Jun 08, 2016 8:46 am 
Offline
User avatar

Joined: Sat May 11, 2013 9:45 am
Posts: 729
Location: PA, USA
Horizenjob wrote:
From the FJ20 thread, Mark says:
Quote:
Currently the FW uses the eTPU of PWM, so you can have any frequency you please.....and will work with the driver its attached to.


OK, that's interesting, I didn't know that. So some number of eTPU signals can be configured as PWM?


Really all 32 eTPU pins can be used for PWM if that is what the user wants. From a practical stand point pins that aren't connected to a driver can't do anything and the driver latency is what sets the highest frequency available.

IIRC I was setting up 8 etpu pins as generic optional outputs that could do GPIO or PWM in response to up to 4 other channels (rpm, MAP, temp, etc) and planned to add a generic PID function that could be used with them but hadn't gotten that far.

My goal was to keep everything as simple as possible/practical. Simple, generic drivers that can be software configured to drive whatever (which meant external ignition module). So 24 outputs means....4 fuel, 4 spark, 16 optional or 12 fuel, 12 spark no optional or 24 optional to control an HVAC system or whatever the user wants.

The generic output are what originally drove the conversion to floats.....TS can't deal with displaying graphs and or saving/load data of varying types in the same memory location and floats provided a solution to that hurdle and without generic optional outputs the code becomes stupid long and complex. This is all the stuff I was working on when I decide to focus on the mechanics of the engine first and planned to just use a known good ECU for startup (I deviaded from that a bit and sold my motec and bought an enginelab) then come back to o5e


Top
 Profile  
 
 Post subject: Re: Practical and basic prototype ECU decisions...
PostPosted: Wed Jun 08, 2016 11:33 am 
Offline
User avatar

Joined: Thu Mar 26, 2015 11:55 am
Posts: 112
Location: Massachusetts, USA
OK. At the moment there are 4-5 GPIO and 2 analog pins between the CPU board and the driver board. There are also 5 chip selects for the SPI, so there might be a GPIO there too if we don't need 5 chip selects.

I am using 2 26 pin headers to interface between the boards, they could expand a bit. It's a common size and using the same size for both cuts down on the parts count.

Some more overview stuff for Rob.

These boards were originally designed at approximately 2.75"x5". Just before I submitted my first board iteration to fab I changed to this enclosure:
http://www.yonggu-enclosure.com/p-131.html

This made the boards get a half inch wider so now they are 85mm ( 3.35" ) x 133mm ( 5.24" ). The enclosure is an extrusion so the length of the boards is flexible and could also be different.

At the moment the ECU is designed so that all the power and driver wiring comes out one end and the analog uses the other end. I am not sure this is a win, but it was an attempt to get good quality analog sensing. Obviously this affects connector selection.

My original intention was to make the unit with pigtails and avoid the first layer of connectors. Maybe it should be offered both ways - pigtails or a variety of connectors. I did not realize when I started that the ECU business might actually just be a wiring harness business - this still surprises me. How can a chip with 25 million transistors cost less than one side of a connector? :shock:


Top
 Profile  
 
 Post subject: Re: Practical and basic prototype ECU decisions...
PostPosted: Thu Jun 09, 2016 3:52 pm 
Offline
User avatar

Joined: Sat May 11, 2013 9:52 am
Posts: 304
Location: Over here, doing 'over here' things.
Not meaning this to be disparaging or otherwise, but having some things attached at one end and other things attached at the other end increases the space the ECU requires for use.

_________________
/me goes off to the corner feeling like Jerry Springer with a mullet.

My O5E candidate: 1982 Honda CX500TC motorcycle.


Top
 Profile  
 
 Post subject: Re: Practical and basic prototype ECU decisions...
PostPosted: Tue Jun 14, 2016 4:29 pm 
Offline
User avatar

Joined: Thu Apr 21, 2016 9:25 pm
Posts: 51
Well talking about spec 55 wiring, is common used on motorsport industry for sometime already and can be single or double shielded and the core can be coated with different materials.
This is the supplier that i was talking about:
https://www.emhmotorsports.com/shop/ray ... c-55-wire/

and this is a good article of how good it is compare to GXL,TXL or SXL

https://www.rbracing-rsr.com/wiring_ecu.html

As connectors, there is some advise, Mouser have some good connectors that have over 100+ pins with hi and low current pins.


Top
 Profile  
 
 Post subject: Re: Practical and basic prototype ECU decisions...
PostPosted: Thu Jun 16, 2016 1:29 pm 
Offline
User avatar

Joined: Thu Mar 26, 2015 11:55 am
Posts: 112
Location: Massachusetts, USA
Quote:
Not meaning this to be disparaging or otherwise, but having some things attached at one end and other things attached at the other end increases the space the ECU requires for use.


Well it might or might not, I don't think that is a clear issue. I would like to make it easy to keep the analog and power circuits away from each other. We could look for extrusions that the boards would go sideways into, something with a width of 5"-6" instead of 3"-4". Maybe the same boards would work in either.

I was originally using twisted pair ribbon cable to get out of the box for the analog wiring. If we decide to use twisted pair with shield instead, how do people connect the shield at the ECU end? Outside of using something like a phono jack...


Top
 Profile  
 
 Post subject: Re: Practical and basic prototype ECU decisions...
PostPosted: Thu Jun 16, 2016 3:36 pm 
Offline
User avatar

Joined: Sat May 11, 2013 9:52 am
Posts: 304
Location: Over here, doing 'over here' things.
Horizenjob wrote:
Quote:
Not meaning this to be disparaging or otherwise, but having some things attached at one end and other things attached at the other end increases the space the ECU requires for use.


Well it might or might not, I don't think that is a clear issue. I would like to make it easy to keep the analog and power circuits away from each other. We could look for extrusions that the boards would go sideways into, something with a width of 5"-6" instead of 3"-4". Maybe the same boards would work in either.

My application is a motorcycle, so space is at a premium. This is why I made the comment.

I'd be more worried about separating analog and the injectors / ignition than I would be about the power. But even then, OEM's seem to have a way to run them all together along the loom without issue. Maybe they sample analogs when the digitals aren't doing anything important?

Quote:
I was originally using twisted pair ribbon cable to get out of the box for the analog wiring. If we decide to use twisted pair with shield instead, how do people connect the shield at the ECU end? Outside of using something like a phono jack...

As for grounding shields, I'd suggest at one end or the other. At the ECU makes the most sense, however, a bunch of shields twisted together into an eyelet or something makes no sense. If they go into the ECU, they need to connect to a ground that exits the ECU over as short a distance as possible.

_________________
/me goes off to the corner feeling like Jerry Springer with a mullet.

My O5E candidate: 1982 Honda CX500TC motorcycle.


Top
 Profile  
 
 Post subject: Re: Practical and basic prototype ECU decisions...
PostPosted: Thu Jun 16, 2016 5:39 pm 
Offline
User avatar

Joined: Thu Mar 26, 2015 11:55 am
Posts: 112
Location: Massachusetts, USA
OK. I consider the injector and coils to be power, so I lumped them in. They are the big concern because of the voltage spikes, especially the coils.

Quote:
As for grounding shields


I'm asking about how that works with connectors. Electrically I would connect them to the chassis ground, which would include the box. I'm planning on the CPU grounding to the battery and the coil drivers to the cylinder heads.

Some shielded wires have a drain wire which makes them easier to ground, others have a braid which is where it starts to get more difficult. Maybe I need to provide a chassis ground post on my board and then the braid could be bolted to that.


Top
 Profile  
 
 Post subject: Re: Practical and basic prototype ECU decisions...
PostPosted: Sat Jun 18, 2016 11:27 am 
Offline
User avatar

Joined: Thu Apr 21, 2016 9:25 pm
Posts: 51
Well i keep reading all that you say and trying to make my mind in all of this. I been working with a lot of different ECU's and for my experience you need twisted and shielded cables only for the triggers and the knock sensors. if you are planing on drivers for the coils and injectors you probably will need dedicated grounds for those drivers that can be on the same connector but isolated from ecu grounds (easy).
if you ask many people about proper grounding to the car you will have multiple different opinions some will say the chassis other will say the engine. my opinion is that any will work if you have the engine properly grounded and the battery to the car is properly grounded too, many ECU use different approaches. my opinion is that drivers need to be grounded to the chassis of the car and sensors to the engine and ecu to the battery. Power cables for the coils and injectors relay isolated to 12v directly from battery, ( I will love to design a power distribution block in the future).
Using the ECU enclosure as a ground can be useful but not the best approach, there some people who like to mount the ECU on panels mostly on racing cars and those panels most of the time the grounding not good. for the grounding for the sensor shielded cables no matter if is braided you can still solder it to one of the ecu connector pins without any problems and let it loose on the side of the sensor.
Provably i want to see a line diagram of your approach.

This is just my opinions and sorry for my English

Rob


Top
 Profile  
 
 Post subject: Re: Practical and basic prototype ECU decisions...
PostPosted: Mon Jun 20, 2016 10:42 am 
Offline
User avatar

Joined: Thu Mar 26, 2015 11:55 am
Posts: 112
Location: Massachusetts, USA
Quote:
This is just my opinions and sorry for my English


Your first couple of posts were a little hard to understand but recently all have been fine. You have good English.

Quote:
need twisted and shielded cables only for the triggers and the knock sensors.


Agree, maybe also O2 sensors perhaps?

Quote:
planing on drivers for the coils and injectors you probably will need dedicated grounds for those drivers that can be on the same connector but isolated from ecu grounds (easy).


Yes, you understand. The coils are easy because they use discrete transistors (IGBT). The fuel injectors are more difficult because they are internal to the smart driver chip. So they land up sharing a ground with the SPI and also the inputs that tell when to fire the coils and injectors. This is the downside of these smart chips for this purpose. I'm not sure if other brands of these chips also share this ground, but it is likely. Hard to have different grounds on one chip.

So there would be a digital ground line from the driver board to the same place the CPU ground goes, just to avoid extra current on the ground connection between the boards. Actually I may have to check I have also considered getting the CPU ground from the driver board because the CPU is actually a small load.

Quote:
Power cables for the coils and injectors relay isolated to 12v directly from battery, ( I will love to design a power distribution block in the future).


Yes, I think this too. At this time transistor relays may be better then electro-mechanical. Since my welder uses semiconductors, it seems they would be up to the job these days.

Quote:
Using the ECU enclosure as a ground can be useful but not the best approach,


Just for shielding etc.


Top
 Profile  
 
 Post subject: Re: Practical and basic prototype ECU decisions...
PostPosted: Mon Jun 20, 2016 2:25 pm 
Offline
User avatar

Joined: Thu Apr 21, 2016 9:25 pm
Posts: 51
Quote:
At this time transistor relays may be better then electro-mechanical

That is correct this days but they are more expensive by far 40 amps can cost 50$,
but yes is the best you dont going to have flyback current on those solid state relays.

I study some other ECU's that i have here that customers burn them and believe me companies do some crazy stuff inside and they work so i don't think we going to have problems with grounding stuff inside the ECU, isolating as much as possible and maintaining hi output away from analog stuff will be enough.


Top
 Profile  
 
 Post subject: Re: Practical and basic prototype ECU decisions...
PostPosted: Tue Jun 21, 2016 12:09 pm 
Offline
User avatar

Joined: Thu Mar 26, 2015 11:55 am
Posts: 112
Location: Massachusetts, USA
Quote:
I study some other ECU's that i have here that customers burn them and believe me companies do some crazy stuff inside


OK, this is good, I think having some blown up units to look at will be excellent experience.

Is there a general place they fail? The reason I'm doing over my board is that I tired to provide extra protection on the analog inputs and didn't do as good a job as I should have. I am also hoping to upgrade the power input protection. These things don't cost much money, but take effort to get right.


Top
 Profile  
 
 Post subject: Re: Practical and basic prototype ECU decisions...
PostPosted: Tue Jun 21, 2016 6:58 pm 
Offline
User avatar

Joined: Thu Apr 21, 2016 9:25 pm
Posts: 51
It's all about protection, filtering noise and proper power distribution.

What i see on different ECU's burn out are first: the IGBT's, improper setup of duell burn most of the driver transistor, second the power supply, noise on power
supply cousins reset's is the most common, and belive me ECU's usually don't blow up easily, behind every burn ECU's is a bad practice of the owner, after a injector drive burn it is a bad injector, people tend to jump the fuse. most of the Power supply issues are relented to jump starters over current in uncharged batteries.
something that can kill power supply is, incorrect spark plugs, none resistor spark plugs without the proper caps put a tremendous amount of nice on the power supply, i see this my self, lot os ECu resets and PS burns.


Top
 Profile  
 
 Post subject: Re: Practical and basic prototype ECU decisions...
PostPosted: Wed Jun 22, 2016 12:23 am 
Offline
User avatar

Joined: Thu Mar 26, 2015 11:55 am
Posts: 112
Location: Massachusetts, USA
Users are the worst! :-) I see what you are saying. I hear these days many people jump start with 2 batteries. So this gives me some ideas. We can look at the power input protection some more.

This unit can monitor the coil current so we should be able to protect against over current. I think this is true of the injectors too, but will have to check. Of course the software is what needs to do the checking, so that will take work too.


Top
 Profile  
 
 Post subject: Re: Practical and basic prototype ECU decisions...
PostPosted: Wed Jun 22, 2016 1:00 am 
Offline
User avatar

Joined: Thu Apr 21, 2016 9:25 pm
Posts: 51
Yeah there is some PS that can hold more than 32 volts, what i know those inexpensive battery chargers what they do on crank mode is to raise the voltage good ones raise the current.

On the side of the IGBT's, more than the dwell to high is the miss firing, if you start a engine and some how the car is miss firing meaning that you don't know what you doing the IGBT'S get hot very fast caused by noise or bad timing, is very common when you are using distributor CAS. Also i think that if you plan to put IGBT's inside the box i think that the best way is to limit on the software the dwell that you can select lets say 3.5 top and 4 at cranking, never see the reason to use more than that.
this numbers are standard on my installations i dont use more than 3, usually i use 2.7 anyway they not survive more than 3.5.
is all about proper fuse intalation on the side of the injectors I divide Power to injectors in 2 banks and never need to use more than 7.5amps every 3 injectors.
usually bad injectors dosn't blow your driver, happen to me many times, they been stored for long time and they are jammed but they dont damage nothing only one time i was starting a new engine and the engine flooded full to the cylinder heads and the driver fried but just one time.


Top
 Profile  
 
 Post subject: Re: Practical and basic prototype ECU decisions...
PostPosted: Wed Jun 22, 2016 10:17 am 
Offline
User avatar

Joined: Sat May 11, 2013 9:45 am
Posts: 729
Location: PA, USA
semi-related, I got a care package from Reanimotion who's decided to go a different direction and send me all his o5e type stuff including a bunch of dev board for various chips. I'll look tonight to see exactly what I have and post the list, I'm guess a few are things you can use or may want to play with.


Top
 Profile  
 
 Post subject: Re: Practical and basic prototype ECU decisions...
PostPosted: Tue Jun 28, 2016 12:11 pm 
Offline
User avatar

Joined: Thu Mar 26, 2015 11:55 am
Posts: 112
Location: Massachusetts, USA
Voltage clamps...

Diodes have been giving me a headache for a week or two now. Digikey lists over 45,000 part numbers for diodes. Most parts come in full reels, Digireels and cut tape so I guess there are more like 15,000 parts and many are multi sourced, but it's still a lot of parts.

I'm looking for suitable devices to clamp the analog inputs. Most of the analog inputs should be clamped to the rails and the VR inputs are differential and should be clamped to limit their swing.

The Freescale chip provides protection on the inputs, but at an input current of 3 mA an error of 2 bits is caused on adjacent pins. I think that's how to read the datasheet.

An after market ECU used in a race car or hobby project will likely be exposed to more handling of it's input sensors and also perhaps a certain degree of ineptitude in wiring etc. So it seemed reasonable to provide greater protection of the inputs. I am not looking to make this more expensive, but diodes and passive components cost very little.

The ESD protection part of this has been pretty easy, I think. Then again I started double checking the leakage currents and maybe not so much, ROFL. Originally I bought some "ESDA2" devices from Bussman, but now may go with devices from AVS because their datasheets are just better, particularly in describing their leakage current. These style devices come in 0603 packages, react in much less than 1 nanosecond and can take many 100A pulses without degrading.

You can get the datasheets from these links.
http://www.digikey.com/product-detail/en/avx-corporation/VC060314A300DP/478-2499-1-ND/684906

http://www.digikey.com/product-detail/en/avx-corporation/V2F105C150Y1FDP/478-2482-6-ND/1717349

http://www.digikey.com/product-detail/en/eaton/0603ESDA2-TR2/283-4156-6-ND/3681417

This post has become long so I'll follow up with another post about the voltage clamp diodes....


Top
 Profile  
 
 Post subject: Re: Practical and basic prototype ECU decisions...
PostPosted: Fri Jul 01, 2016 5:44 pm 
Offline
User avatar

Joined: Sat May 11, 2013 9:45 am
Posts: 729
Location: PA, USA
Horizenjob wrote:
.... also perhaps a certain degree of ineptitude in wiring etc. So it seemed reasonable to provide greater protection of the inputs.....


Describes my normal behavior perfectly! :oops:


Top
 Profile  
 
 Post subject: Re: Practical and basic prototype ECU decisions...
PostPosted: Wed Jul 06, 2016 2:56 pm 
Offline
User avatar

Joined: Thu Mar 26, 2015 11:55 am
Posts: 112
Location: Massachusetts, USA
Sorry for being late on updates here. Rob, also we need to discuss getting you the KiCad files. My most recent backup is about 250 Mib, can you take an email attachment that size or do you have a Google account and I could try to give you access to my Gdrive?

So a lot of the last couple of weeks has been spent on leakage currents etc. I think we'll be OK with a good choice from the AVS line of TransGuard pieces. The data sheets are pretty good, but they are somewhat mysterious devices. These units go before any resistors on the analog inputs, so their leakage current is one step less critical. I would like leakage to be 1 uA or less.

We also need protection for the analog inputs against shorts to battery voltage and worse also the clamp voltage of the AVS ESD devices. Here again leakage current has been making choices difficult. My first choice that was used on the first boards was a BAS70XY. That is a small package with 2 pairs ( 4 total diodes ) of series connected Schottky diodes. These diodes have a low voltage drop, but also potentially a large leakage current, especially at higher temps. So it appears Schottky diodes as a class are out.

I had thought with many thousands of parts listed at Digikey, I could get what I wanted, but really the remaining choices fall in basically a couple of classes. There are fast signal diodes and low leakage diodes. The low leakage diodes have crazy specs for their leakage, their typical spec is several orders of magnitude better than their max. I usually design to the max spec, which for these diodes is still quite good. Makes me wonder if the bulk of the faster diodes would also be like this but the datasheets don't say.

The low leakage diodes spec between 1 and 3 uSec. for their switching time. The high speed diodes are a thousand or more times faster. The fastest signals I think we need to deal with are the crank trigger and maybe things involving measuring spark durations and latencies. The crank trigger wheel at 10000 RPM and 60 teeth is still around 10 KHz., so I expect the low leakage diodes will work fine here.

For examples of these two types of diodes, look at the BAV199 and BAV99.

After all this looking at diodes I started to wonder about op-amps on the inputs instead of passive filters. The OPA365 is a part from TI made to be an input buffer for ADC systems. It has it's own internal diodes for protection that are rated for 10 times the current as the CPU chip, but also 10 times less than the diodes we're looking at. They come in about the same package as the diodes and cost about $0.75 in quantity, the diodes are more like $0.20.

So the op-amps would provide more protection. They could also provide improved filtering by using active filtering. Basically that allows a steeper rolloff after the cutoff frequency. It would take just a little more board space because of a couple more resistors and caps.

Maybe the right way to look at this is to put both of these approaches onto an analog channel on the board and do the layout and compare them. I'd like to get off the analysis and make some progress on the board again...


Top
 Profile  
 
 Post subject: Re: Practical and basic prototype ECU decisions...
PostPosted: Thu Jul 07, 2016 12:16 am 
Offline
User avatar

Joined: Thu Apr 21, 2016 9:25 pm
Posts: 51
Ok

Quote:
My most recent backup is about 250 Mib, can you take an email attachment that size or do you have a Google account and I could try to give you access to my Gdrive


Yes I have a Goolge account, you can send access to aircreations@gmail.com

I'm studding this all analog thing before cos i'm trying to do something similar at this right moment, all i can say is that i never burn a processor playing with analogs but i know that happens sometimes, i erase the firmware on a hit sometimes.
I was planing to use simple 1n5818 schottkyand i think that op amp is the best idea, but in terms of space, i know that there are some packages with 4 channels around.
I have some time on Friday to study this, i let you know.


Top
 Profile  
 
 Post subject: Re: Practical and basic prototype ECU decisions...
PostPosted: Thu Jul 07, 2016 1:40 am 
Offline
User avatar

Joined: Thu Mar 26, 2015 11:55 am
Posts: 112
Location: Massachusetts, USA
Quote:
i erase the firmware on a hit sometimes.


That's impressive. So it doesn't seem a bad idea to include those AVS ESD devices.

The 1n5818 is a good unit, but you can get in trouble using that in the analog filters and protection. Some of the data sheets for that part say 1 mA leakage at 25C and 10 mA at 100C. The throttle sensor on my Ford 302 has about 1 mA total current thru it. So a big percentage of your signal could disappear. The 1 mA leakage is at full reverse voltage but still an issue.

With these Schottky diodes the leakage is very temperature dependent, so my worry is that the temp of the ECU will really change the calibration of the sensors.

I'm not sure the package with more units in them save you space. It seems like a no-brainer but when it came to do the layout it was difficult. For boards with more layers maybe it works. The OPA356 comes in a dual package also, but it's twice as big. Still microscopic though, LOL.

On Thursday I will work to send you a link to the files...


Top
 Profile  
 
 Post subject: Re: Practical and basic prototype ECU decisions...
PostPosted: Fri Jul 08, 2016 8:09 pm 
Offline
User avatar

Joined: Sat May 11, 2013 9:52 am
Posts: 304
Location: Over here, doing 'over here' things.
... and people look at me funny when I suggest op-amp voltage followers / buffers to handle analog inputs....

You can clamp the input if you want, which might have a high current component, or clamp the output from the buffer which will likely have a very low current component

Your choice. ;)

_________________
/me goes off to the corner feeling like Jerry Springer with a mullet.

My O5E candidate: 1982 Honda CX500TC motorcycle.


Top
 Profile  
 
 Post subject: Re: Practical and basic prototype ECU decisions...
PostPosted: Fri Jul 08, 2016 9:26 pm 
Offline
User avatar

Joined: Thu Apr 21, 2016 9:25 pm
Posts: 51
My brother join the forum, he is a senior electronics engineer, he can give us a example of why we need some Op-amps on our analog designs.


Top
 Profile  
 
 Post subject: Re: Practical and basic prototype ECU decisions...
PostPosted: Sat Jul 09, 2016 2:49 pm 
Offline
User avatar

Joined: Thu Mar 26, 2015 11:55 am
Posts: 112
Location: Massachusetts, USA
Quote:
You can clamp the input if you want, which might have a high current component, or clamp the output from the buffer which will likely have a very low current component


Well you always have to check if clamping is required, there are a lot of ways to get in trouble..

We would like op amps that provide stronger protection on their inputs and prevent wiping of the firmware. We can also improve the anti-aliasing function of the input filter by using an active component. Active filters are a really big subject though. I'm using the TI web-design tool for active filters and reading up in my new copy of "Art of Electronics".

We can use a Bessel filter which provides very linear phase relationships so we can display our signals like an oscilloscope. That would be nice for coil and injector currents and maybe others ( reading combustion chamber plasma ). We could use a Chebychev or other filter with a much more pronounced roll-off.

So how you filter and sample are tied together. You need to sample fast enough to see everything that gets thru your filter or you run the risk of aliasing, which is a high frequency being interpreted as a lower frequency that isn't there.

The abilities of the opamp I mention above is way beyond what we need, but they are not expensive and it might be nice to get experience with such a device. There are also slower devices that continually trim themselves for extremely low errors ( choppers ) that might also make sense.

An opamp might help on the driver board too, because I had some issue picking a current sense resistor for the coil driver, now maybe I can get more choices and make it easier to drive the signal to an analog input on the CPU in addition to the coil driver chip.


Top
 Profile  
 
 Post subject: Re: Practical and basic prototype ECU decisions...
PostPosted: Sat Jul 09, 2016 2:50 pm 
Offline
User avatar

Joined: Thu Mar 26, 2015 11:55 am
Posts: 112
Location: Massachusetts, USA
Quote:
My brother join the forum, he is a senior electronics engineer, he can give us a example of why we need some Op-amps on our analog designs.


This sounds good!


Top
 Profile  
 
Display posts from previous:  Sort by  
Forum locked This topic is locked, you cannot edit posts or make further replies.  [ 37 posts ]  Go to page 1, 2  Next

Who is online

Users browsing this forum: No registered users and 0 guests


You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot post attachments in this forum

Search for:
Jump to:  
cron
Theme designed by stylerbb.net © 2008
Powered by phpBB © 2000, 2002, 2005, 2007 phpBB Group
All times are UTC - 5 hours [ DST ]