Bug found / Suggestion

Report problems on the Megajolt firmware and PC software

Moderators: JeffC, rdoherty, stieg, brentp

Post Reply
Darren Brightman
Posts: 5
Joined: Fri May 05, 2006 12:10 pm

Bug found / Suggestion

Post by Darren Brightman »

I recently took my Zetec powered Caterham with Megajolt to a well know and respected rolling road , he was slightly dissapointed when we tried to set up the car we found.

1) The live tunning didn't work properly , as even if we held the revs / load exactley on a site it would alter the wrong site
making setup very very hard. We eventually found that you have to hold revs and load just above the threshold for it to
alter the correct site. e.g The software doesn't alter the nearest site but the next one down.

2) He suggested more points would be helpful e.g 16x16 which he tells me is standard on some of the more expensive systems.

3) He also thought the live display didn't show the interpolated advance.

( If I can be of any furthe help please let me know , I am a software engineer by trade and very familiar with C++, if I can help )

MartinM
Posts: 433
Joined: Sat Aug 27, 2005 12:08 pm

Post by MartinM »

Hi Darren, welcome!

All the reports I've had given to me about RR experiences have been very positive and when I've been at a RR tuning various engines with MJLJs, we've had no problems at all

1. Can't comment - never used live updating. You could have changed the bin you wanted and sent the whole map to the MJLJ as that would definitely have updated the bin you changed. But of course this needs looking into and I'm sure Brent will do so

2. Maybe. Or maybe 16x16 is just a gimmick/sales aid as engines don't actually need an ignition map at that resolution given the interpolation that the MJLJ does between adjacent bins. Do the "expensive systems" have 16x16 AND interpolation? Genuine question by the way, I'm not trying to be defensive about the MJLJ...

3. It definitely shows the interpolated advance - it is reading the advance from the MJLJ in real-time (which is definitely interpolated in V3 software and accurate to the nearest degree - it's a simple sending of the internal variable in the MJLJ processor memory that holds the current advance value, across the comms link). It's definitely not doing a un-interpolated configurator-side estimate of the advance. Why did the tuner think it wasn't showing the interpolated value?

Regards
Martin
...hey...my 100th post...do I get a star by my name now Brent? :D

MartinM
Posts: 433
Joined: Sat Aug 27, 2005 12:08 pm

Post by MartinM »

...just adding my 101st post to see if I get a star for over 100 posts - doesn't look like it :(

Darren Brightman
Posts: 5
Joined: Fri May 05, 2006 12:10 pm

Post by Darren Brightman »

Starting in reverse.

3) I wasn't sure if the advance value in the live update was interpolated or not but the tuner didn't think it was but he may have been confused.

2) I also thought 16x16 was over the top but if most other systems offer this and and the Megajolt can hold 2 10x10 maps
then I am sure a 16x16 would be possible.

1) This was the main and most important point , On the rollers the operator changed the map whilst watching the BHP rise and fall and listening for pinking etc , this was very hard to do when the correct bin wasn't changing , changing each value then flashing it over works but is far too slow when the cars actually being mapped on the rollers.

We ended up building a map based mainly on experiance which works fine but wasn't the ultimate we would have got if the live mapping worked properly.

SRandle
Posts: 3
Joined: Wed Oct 27, 2004 3:40 pm
Location: Rugby England
Contact:

Zetec Map

Post by SRandle »

Would it be possible for you to share the map that was created, as not many people are coming up with zetec maps

Steve

brentp
Site Admin
Posts: 6274
Joined: Wed Jan 24, 2007 6:36 am

Post by brentp »

Hi Darren,

Think of the 3d tuning view as a 3D version of the main ignition map grid.

This is important: The number above the bouncing ball is the ignition advance for that cell in the Map for current load/RPM Bin, not the actual ignition advance at runtime.

As the load and RPM varies on the engine you are almost always between Bins. A big feature of the MJLJ is it's support for full linear interpolation in-between RPM and Load bins. For this reason you need to keep an eye on the actual ignition advance gauge (the big green digital display) as you edit the Ignition Map, regardless of whether you are editing with the grid or the 3D tuning view.

For example; if bin 1 is 1000 RPM and bin 2 is 1500 RPM, and the engine is at 1490 RPM- the runtime will show at bin 1 until it crosses the threshold for bin 2 (1500 RPM). This is by design.

The MJLJ interpolates between the current bin and the next. So, if it's close to crossing the next Bin's threshold, say 1490 RPM as in the example above, editing the next bin may have more effect on the actual advance than the current bin.

Does this explanation seem to help? Try hooking up your laptop again and make small adjustments to the Map with this knowledge in mind and let me know what you think.

I'm planning on creating an 'Operation' guide to compliment the Assembly and Installation guide and it would certainly cover these concepts. However, if this is confusing enough for people perhaps making some clarifying adjustments in the software could help as well- but... historically it hasn't been a hot topic- indeed it might be first time we discussed this.

So Martin: I see that names aren't decorated nor are the avatars turned on! I'll see what I can do, but I don't like mucking with the forums software any more than I have to- but I did just recently put in some anti-spambot software which was a big relief for me. :)

Regards,
Brent Picasso
CEO and Founder, Autosport Labs
Facebook | Twitter

brentp
Site Admin
Posts: 6274
Joined: Wed Jan 24, 2007 6:36 am

Post by brentp »

I thought about the topic a bit more.

As before, the live tuning window shows the current working Bin, determined by the controller at runtime. Perhaps it might be more intuitive to have the software select the nearest bin rather than just the active bin. This way, when bumping the cell values up and down it would have a more direct effect on the actual ignition advance.

However- it still wouldn't directly affect the actual advance since that is still an interpolated value.

Thoughts?

Darren Brightman
Posts: 5
Joined: Fri May 05, 2006 12:10 pm

Post by Darren Brightman »

Hi BrentP,

I totally understand how it works , when on the rolling road the operator was very skillfull at getting the correct rpm
and load however , often it was editting the next bin down , if it was the nearest cell that was editored then the amount of iterpolation would be next to nothing. As you say just a few rpm down and the wrong cell is edited and the advance
doesn't appear to change until you get to the next bin down.

P.S I'm not trying to be negative the system works very very well , but this small chnage woud make the live mapping easier.

Also for Steve I will add my map to the database either later or tomorrow.

brentp
Site Admin
Posts: 6274
Joined: Wed Jan 24, 2007 6:36 am

Post by brentp »

Hi Darren,

No worries! The kind of feedback you're providing is extremely valuable. I just wanted to make sure you knew how it behaved and why, even though it may be confusing under certain certain conditions.

Thanks,
Brent Picasso
CEO and Founder, Autosport Labs
Facebook | Twitter

Post Reply