Page 1 of 1

Fred's FreeEMS prototyping hardware comments thread

Posted: Thu Feb 07, 2008 1:21 am
by Fred
Hi, Please place your comments and discussion on this in here.

Thanks,

Admin.

Re: Fred's FreeEMS prototyping hardware comments thread

Posted: Mon Mar 10, 2008 2:07 am
by jbelanger
No wonder you're having problems with the regulator: it's rated at 100mA. Use an LM2940 and you'll have 1A to play with.

Jean

Re: Fred's FreeEMS prototyping hardware comments thread

Posted: Mon Mar 10, 2008 8:03 am
by Fred
LOL, Thanks for checking, although I noted down the part number, I hadn't stopped to look it up. :-) I have another un-sinked 4 terminal scavenged reg on the other board, I just haven't run the wires yet.

I was intending that for real use the on board reg would feed the CPU and a separate reg would feed the TPS etc. The voltage offsets won't matter in reality (even if it bugs me to think about it). However, the CPU can draw more than 100mA when working hard anyway, and having that to220 just hanging there like that is a bit rough. I wonder if it's acceptable to back feed the 5v supply up through the two 50 pin IDCs. If so, that would remove the requirement to change the on board tantalum(s) to higher voltage units too. Just snip off the to220 leads and call it a day :-)

I also didn't try to measure how low the power pack output was going during this, it is just an old panasonic phone unit, so it might be going too low itself...

Thanks for pointing that out :-)

Admin.

Re: Fred's FreeEMS prototyping hardware comments thread

Posted: Mon Mar 10, 2008 9:32 am
by Fred
I checked the raw supply at 13V when it's resetting constantly. The board is being fed 11V at that (non LDO 12v reg) and it's still not coping. I've emailed Carl at Tech Art about possible solutions, we'll see what they say :-)

Admin.

Re: Fred's FreeEMS prototyping hardware comments thread

Posted: Wed Nov 11, 2009 5:35 am
by jbelanger
Wow, haven't put anything in here for a while! :-)
I've had to dig this comment thread from the bottom also :)
Yesterday afternoon I drove my FreeEMS powered Volvo up to DIYAutoTune and purchased an upgraded JimStim chip for my JimStim. While I was up there I gave Jerry a quick ride in the car - just down the road and back - understandably he didn't seem too happy about it, but never the less, it is what it is.
Really? I would think he would see the opportunity in this. Since he does a lot of PNP and other custom setups, that would give him another base for it. Unless he's not interested in anything not B&G because that might piss them off (which is most likely what it is).
Jean, my JimStim chip has 2.0.3 on it, tell me about what I bought :-)

Fred.
To really take advantage of the new stuff you need the serial cable. It's either the FTDI cable shown on my site (http://jbperf.com/JimStim/JimStim_v2_0_ ... tml#serial) or the equivalent cheaper Sparkfun board (also linked on my site). Or you could use a 3.3V RS232 chip and a DB9 connector.

Without the cable, what you get are more wheel modes and higher stable RPM for all wheel modes (the chip runs at 10MHz instead of the previous 8MHz so you get 25% higher max RPM).

With the cable you can control the tach signal from the PC with TunerStudio. You can select the wheel mode, invert the crank and cam signal and set the RPM (and you're not limited to ~16500 but to what the chip can actually do, e.g., you can do 25000RPM with the 36-1 mode). Also, you can upload new wheel patterns. I have provided the source code to communicate with the JimStim to do the upload and the wheel pattern file for the standard modes. Since the available memory is almost full, you'll likely need to remove one pattern to add another one unless your pattern is really small but since you can change them whenever you want that should not be an issue.

There's also the possibility of sending a sequence of RPM values to the JimStim. I have provided the source code for a small example to has the RPM go from 1000RPM up to 12000RPM and back down to 1000RPM. The source shows you how to interface with the JimStim. It's a very simplistic protocol without any error checking since this is intended for bench testing and should not be critical.

I show the different new features here: http://jbperf.com/JimStim/JimStim_v2_0_ ... 0_features.

Let me know if you start to play with the different features and if you want any pointers.

Jean

Re: Fred's FreeEMS prototyping hardware comments thread

Posted: Thu Sep 16, 2010 3:49 am
by hamster
I might recommend going with a SMPS instead of a linear regulator. The SMPS (switch mode power supply) doesn't need to drop the voltage into heat to function, and while they cost a little more, they are well worth it.

Here's an example: http://www.national.com/pf/LM/LM2675.html#Overview

So, it's a handful of components, but not super spendy, but they run cool and can handle rich, chunky amps of output. I've got a simple one drawn up in a schematic I could post if you are interested.

Re: Fred's FreeEMS prototyping hardware comments thread

Posted: Fri Sep 17, 2010 3:42 am
by Fred
Post it, but I think the aim for the first one is KISS, however the beauty of doing FOSS firmware is that there can be N hardware solutions out there that work and are better and worse in various ways, so it's no big deal if one uses linear and you dont like that.

Re: Fred's FreeEMS prototyping hardware comments thread

Posted: Mon Sep 20, 2010 4:06 pm
by davecramer
I just talked to technological arts. They now ship this board with a LM2937 which is rated to 800mA. I'm sure he would make a run with an LM2940 if we asked him. Is 800mA enough?

Re: Fred's FreeEMS prototyping hardware comments thread

Posted: Mon Sep 20, 2010 10:10 pm
by Fred
Fantastic!! Yes, it's enough, just heatsink it and you're good! :-)