Holset HX50 Supermoto Jetbike

[slittlewing]

Well, I was hoping to have been able to take the bike to a farm this weekend, to do afterburner testing. Everything was looking good - weather, no social or family stuff, trailer available to borrow... but it was not meant to be!

I came home Thursday evening to fire up the jet bike with the mods (15.5 injector and plug cooling) and there was no fuel pressure. I thought maybe the high side drive had somehow blown, so Rewired the pump to the one normally used for radiator fans just for a quick fireup. Still no fuel pressure.

Took ecu out, after some head scratching found that where the mosfet legs attach to the pcb some of the copper had frazzled. Pump had quite a low resistance so assumed it was almost short to ground.
Wiring it into the fan drive had done the same damage to that one. Checked on my pcb software - I have used huge great big copper patches on the mosfet connections to transfer the current but unbeknown to me, the software recognises that and adds “thermal relief” as a default, to make it easy to solder so the copper patch doesn’t absorb all the iron heat. That makes the big copper areas completely pointless, as there are a few 0.5mm very short hairs connecting it to the mosfet leg ☹️

Anyway I bridged across the dead connections with solder and still no joy. New components on order to try and fix the existing board.

The mystery continues because after removing the pump, it works ok on the bench from a 12v battery (no load) drawing similar current to a known working one. The pump in question is an 044 clone, but a good make (sytec).

For the sake of being safe, I decided to swap the pump with one from the kart AB. The kart pumps are Bosch motorsport variant designed to constantly deliver 8bar (normal 044 designed to deliver 5). The kart AB is too free flowing to ever need 8 bar of fuel pressure so can make do without the high pressure rating.
In addition a friend suggested too low a PWM frequency (490hz) might be causing current spikes. After some Googling I agree, and have modded the ECU to use a different output pin which will provide 980hz and give the pump windings and mosfet an easier time.
image stolen from google search:


I’ll get the new components on soon and see what happens.
All good learning for future ecu design at least.

Cheers

Scott

[racket]

Hi Scott

Bloody electrics :-(

Hope you can get it sorted soon

Cheers
John

[slittlewing]

Couldn't have put it better myself John!
I was extremely annoyed on Thursday night, one of those where the Mrs knows to leave you well alone!!
Had been working on the thing tirelessly for most nights of late, and then all for nothing with weeks to wait for another afterburn session opportunity 

Anyways.. some good news. The new components arrived and I managed to fix the board which saves another ~£120 buying a full ECU worth of components and several weeks of PCB manufacture etc.
The fuel pump is swapped over to the Bosch Motorsport 8 bar, even though I now consider the sytec to be OK.

The next "challenge" is that there are only 2 pins on the microcontroller which do 980Hz PWM. I was already using one for AB fuel pump and now the other for main fuel pump. As it happens, the one I am now using for main fuel pump is also connected to one of the onboard arduino LEDs and is automatically made to flash twice on startup. Meaning when I switch the bike on, the fuel pump is pulsed twice at full voltage.. not ideal!

This functionality is defined at a much deeper level than my code, its effectively hard programmed by the "bootloader" low level software (like BIOS). After many hours googling I think I have figured out how to edit the bootloader and recompile it. Waiting on a special adaptor and lead to "burn" my edited bootloader the arduino. This is far beyond my expertise and not something I really want to get into, but I will give it a try and see if I can get around it. Dont want fuel being pulsed into the CC before every startup.

If all goes well and the lead arrives, I will try another fireup at the weekend (i.e. repeat of a week ago!!) to see if the spark plug melting is sorted. If it doesn't go well I will have killed an arduino 

Also got 20A fuses instead of 30A for the fuel pumps to try and avoid any of this happening in future.

For future ECU's, I will make the controlling "ECU" brain a separate PCB to the "power section" and mosfets. Should make it cheaper to develop or change either half, and avoid EMI issues further.

Cheers

Scott

[slittlewing]

Gave the bike a couple of fireups today. No EMCs issues back with the original plug 👍

However, the Bosch motorsport pump was too “grunty” on pressure to give a low idle (below 0.6bar). It also managed to produce more than 10bar pressure, meaning the max P2 was over 2.1bar at full throttle even with the smaller 15.5gph injector!

I have refitted the sytec pump which I believe is still OK and will try again.

In terms of spark plug, after full throttle running at 2.1bar and stopping the engine again the ground electrode had melted, but it had “balled” back to a point rather than disappearing altogether. Maybe could have done with 3mm cooling holes instead of 2mm haha.

Anyway I decided it had melted itself to a sustainable point so I cut it back so it did not protrude any further than the positive electrode and could spark “sideways”. Gave it another fireup and it survived so hopefully that’s problem solved on that front!!

Post run:



More to follow pending fireup with original pump.

Cheers

Scott

[slittlewing]

Hi gents,

Great news. The old pump is still fine, it must have been the EMC issues when trying other plugs that blew the ecu parts.

Just done a fireup and was able to have a nice low idle of 0.5bar. Still can make 2.1 bar at full throttle, 620TOT. Everything seems to be working well (fingers crossed) and the cut down plug is still good.

Afterburner testing next!!

Cheers
Scott

[racket]

Hi Scott

Progress :-)

Now for some "fun"

Cheers
John

[slittlewing]

Hi All,

A little video encompassing all of the trials and tribulations over the last few months!!




Cheers

Scott

[slittlewing]

Hi All,

I finally managed to take the Bike back to the Farm for Afterburner testing.
It went really well, found the "zone" of fuelling through which it would run. 15% PWM (per bar P2) was the minimum setting at which it ran, 17% PWM (per bar P2) was better, and 20% was too much. Those values are on top of a static amount of PWM that is needed to get past the NRV after the fuel pump.

I did a 60 second thermal stress test and everything survived minus some tiny melting on the rear panel and number plate mount (which I have now refabricated from Aluminium)

I also did "repeat ignition" tests about 10 seconds after a cold start to try and ensure the thing will ignite at the drag strip, where there is very little time to get it going and then "stage" and run!

Video below:


Next step.. racing

Cheers
Scott

[turbojet atv]

Hi Scott,

Man that thing sounds absolutely wicked!! I’m eager to see what it can do on a track.

-Cullen.

[racket]

Hi Scott

Now for some fun :-)

Sounding sweet .

All the best with throwing the leg over her .

Cheers
John

[slittlewing]

Thanks Gents Not long now before it rolls up the track, assuming it behaves

Cheers

Scott

[slittlewing]

A long overdue update.. Just wanted to close this off by letting you all know that I did make it to the track earlier in the year with the Jetbike - and had a really awesome day!! 

After arrival, fired it up in the car park to "warm it up" and prove out the afterburner settings learnt previously - it started/ran/afterburned perfectly.
Then wheeled it to the start line and (of course).. it all went wrong haha. Two failed starts, and some headscratching as to why (it was do with an instant pop/bang of ignition showing as P2 pressure, making the ECU think the engine was running).
Then a hose blew off (has probably done 50/60 starts since rebuild and never been an issue. Also one of only 2 non-JIC hoses on the whole bike!). Easily fixed luckily.
Then we got it started and I went to do a run, but the AB refused to light because of inadequate warm up (trying to AB only 10 seconds or so after startup). Then the engine cut out on throttle off (hasn't happened before, but quite different riding it and going from 100%-->0% throttle immediately vs slowing increasing/decreasing during static tests).
So plenty of annoying issues that hadn't shown in any previous testing!

Anyway... things seemed to be going badly so went back to the pit area and made a few tweaks to the ECU code with the laptop, and all was fixed. I applied more "filtering" to the raw throttle signal so the fuelling backed off slower - no cutting out even if you snap the throttle shut from 100%. Changed the startup fuel ramp settings to prevent the sudden "pops/bangs" and we were back in business with the bike running as it has before 

Subsequently managed 3 runs down the 1/4 mile, and by the last one it was even seeming fairly "polished"! I increased the afterburner fuelling a little on each subsequent run from "medium" to "absolute maximum" (according to previous static testing where it blew out if any richer)! 

The need for a slick operation was really evident during the day, with a big queue of other cars and bikes and basically getting about 15 seconds of time to "stage" and run. So the routine we ended up with was to fire it up with one person ahead in the queue, immediately warm it to >500deg TOT using full throttle, then light the afterburner and ride to the start line, stage and go!! I think this would have been impossible (or at least very difficult) without the ECU control!

The oil pressure was rock steady throughout, it was running about 2.1 bar P2 and at full afterburn, TOT's were just starting to reach get to 750 No melting or thermal issues, even though I could feel warmth on my bum through the leathers and sponge seat!! Top speed reached was 56mph which I was pretty happy with! 

I need to sort the video but a few pictures attached below. I just want to thank this forum for the huge amounts of help and advice - the bike is now finally considered "finished" and its sitting in my garage as a nice ornament   

Cheers

Scott
 

[Sweetenough]

Congratulations Scott!
Impressive speed, do you the top speed is reached after 1/4?
Or would a longer straight allow you to go faster?
Looking forward for the video!

Kind Regards
Thomas

[ausjet]

Congrats mate! Taking it down the quarter is like the grand finale of accomplishments 👌🏻

[slittlewing]

Many thanks gents!! I reckon that was probably the top speed, or at least very close to it! Possibly would do 60 with a smaller or lighter rider 😂

I’ll try and sort a video at some point 👍👍

Cheers
Scott