🔥 Win a T-Shirt or Mug by leaving a written review that can include a technical question for Andre or a future guest like Matt to answer - hpcdmy.co/reviewpodcast Reviews also really help the Tuned In podcast gain more traction, ensuring we can keep bringing you more episodes like this one 🧐 - Taz
I love the nerdy side to modern ECU tech. The reason Haltech are so successful is they have targeted the right audience by making their tuning software super intuitive and easy to use. Great podcast Andre!
Thanks for digging into the HPT vs. Turbosmart turbo details! Until listening to this, the internet had given me the impression the TS turbos were just re-labeled HPT. Glad to hear they are unequivocally proprietary/bespoke to Turbosmart. I've been in the market for a 58-62mm turbo for a while and I'll be giving the TS 58/62 and 62/62 a very close look! Turbosmart quality is beyond reproach.
I was lucky enough to do the Haltech elite training with Matt when it toured in Melbourne Such a great experience thank you Matt I initially felt he was a bit of a traitor leaving haltech so it was great corrected by the context from him I really do appreciate the vision he was was part of at haltech and I was gutted by the RWB acquisition of that iconic Australian brand I'm now really so happy for him and his new frontier. I also really enjoyed the discussion about pressure ratio load axis as a guy doing VGT's on petrol engines pressure ratio tuning is almost mandatory but I've really had trouble forgetting the old Gauge PSI and inHG numbers . Although i understand the science and the math is sound it's still hard for me to get used to . Great work as always Andri and again i will post my tuned in requests Gale banks and Ray hall
I recall drooling over the "new" E6K when it came out. It had a very impressive spec sheet. In 2017, I bought an already-turboed IS300 with an E6K. It ran pretty good but needed some tweaks. After a huge effort to get the correct (beta) software on my laptop (thanks Haltech), I got to enjoy the glorious experience of working with and tuning an E6K. For starters - it doesn't run in Windows. It runs in DOS. Modern windows doesn't have a DOS prompt so you need to use a DOS emulator. The whole thing feels like Atari. The unit doesn't have 3D maps. I suppose the processing power required for that was just too much for the ECU. Instead, your "fuel map" is comprised of a bunch of tables of injector PW vs MAP (in PSIG only). There's one for 500rpm, another for 1000rpm, then 1500rpm and so on... Same situation for your spark "map". It makes finding the correct area to tune/tweak very cumbersome when dealing with any kind of transient. Don't forget there's no mouse/touchpad functionality in DOS - so navigating within the tuning environment is keypad only. Hit the first letter in the name of a menu to open that menu - then arrow down to your choice. Datalog was a .csv file list of numbers. Very limited on what all channels can be logged. No means of characterizing injector dead times, fuel flow, fuel pressure. It's ALL baked into the injector PW in your fuel tables. On paper, the unit was capable of closed loop fueling based on a wideband - but nobody was ever able to get that to actually work...so the unit was effectively open loop only. That said - I was able to get it tuned in pretty darn decent. I could maintain between stoich lambda ± about 5% all the time (except idle) and a consistent 0.78 to 0.82 under boost. I could never get it to consistently idle at stoich. The injectors weren't that big (unknown size, but probably about 750cc) but they needed an idle PW of about 0.9ms. Sometimes I could get it to idle nice at stoich with that PW, but other times it would lean out and die. I needed to turn the PW up to about 1.3-1.5ms to get it to consistently maintain a nice steady idle - but that meant idling at 0.85-0.88 lambda. Dunno if that was instability in the injector drivers or just inconsistent dead times and/or inconsistent non-linear zone of injector operation. Aside from lacking modern features, the biggest issue with the E6K was the "throttle pump" (acceleration enrichment) feature. It didn't work. I guess this was a known issue - but it was a big bummer. Really hurt throttle response when engine was warm, and made it so the engine wouldn't take throttle hardly at all until ECT was at least 120ºF. It's funny to compare the advertised "spec sheet" vs. the real-world functionality... Gotta start somewhere, I suppose!
Things have come a long way, but we're yet to find an ECU/software that is perfect from any manufacturer to be honest. That's not a mark against any of them or the people working hard on such projects, rather just the reality of it all noting even the major OEMs with infinitely more resources have bugs and issues too. On the other hand working out such things is part of the attraction to the work for many on that side of the fence 😎 Glad you got a reasonably decent result in the end with that project, sound like a fair amount of work and possibly seems like a waste of time, but I'm sure it also taught you a lot of helpful things you wouldn't know without that project - Taz.
I think Chris hit the nail on the head with his comment on modelling. Fluid dynamics requires huge computing power that just wasn’t available to smaller companies 10, even 5 years ago. It will be very interesting how much further AI will take us. Work has already begun in this space and the results are both exciting and a little scary.
VE/modeled fueling is great. But honestly the old school simple pulsewidth method is pretty OK. No need to characterize dead times, fuel flow @ various pressure, etc. Just more or less PW to get the desired effect. Don't get me wrong - I don't use the PW method anymore - but it did the job pretty well in a very simple way.
@@GroovesAndLands in this respect I was referring to the design of the turbocharger. It’s a fluid dynamics problem and requires some hefty computing power to get any data worth using. I spoke briefly with Willem Toet about the relatively simple aero packages on Fomula SAE cars and he said data on anything less than a 16 core processor is worthless.
@@woznaldo Gotcha. My response is "don't let perfect be the enemy of good". My FSAE days were over 15 yeas ago now; and I've not really followed it since. I assume the aero has improved quite a lot. Anyway, back then it was abundantly clear that almost any aero was a real advantage. I didn't (and don't) know the first thing about designing aero stuff for cars - so I was interested to ask the other teams about their aero. What I learned is most of them hardly knew anything about it, either - and had done rather primitive CFD and analysis in designing theirs... Yet they obtained excellent results that were clearly visible even in very low-speed skidpad event. Watching, it was obvious that just a diffuser generated more grip than nothing at all, and a rear wing + diffuser outperformed just a diffuser, etc. All this to say, coming from an automotive/race engineer that's been at it for 20 years - don't let perfect be the enemy of good.
@@GroovesAndLands all good. The FSAE aero issue was still present when I did my last comp 8 years ago. Too many teams put effort into aero but, their cars weren’t reliable and couldn’t finish the endurance event. Time would be far better spent investing in the team (people) and building a reliable car. We digress. Your point is taken.
@@woznaldo Heard! My school was small and FSAE did not count for credits. As such we had a small team. There were 15-20 total members but only 5-6 "core" members we could count on to get anything real, done. Thus, we had to focus on the core aspects of the car. We certainly didn't have the bandwidth to get into aero. FSAE was an amazing experience but looking back - I/we took the wrong things seriously. We focused on designing/building an awesome machine when we should have focused on the logistics and management of excellent teamwork. Achieve that and the machine will come.
Great podcast but it's a shame that you didn't get to discuss cold side boost control. I only saw this for the 1st time in a Steve Morris Engines video the other day although he says that somebody else (1 of his customers) has been running this system for the past year and it uses a Turbosmart stepper motor controlled butterfly valve system fitted to the cold side with nothing (controls, valves, etc.) on the hot side. Interesting concept. EDIT: Just watched Steve's latest video and it seems there are reasons why the above isn't viable and still needs a wastegate on the hot side.
I think maybe its to keep the compressor and turbine spinning in its optimum range (waste gate), whilst maximising boost intake (dbw bleeder). Especially common in smaller motors, where if you cram so much boost into the motor, its gonna some drive Intake air temps high and reduce efficiency. You see it in rally cars, but I would love to hear whether or not this is a valid answer in a part 2 :)
My favorite thing about Haltech is that you can edit axis parameters to make custom tables . I feel like other ecus dumb it down too much and don’t include this .
For the guy running a shop, the customers coming in the door have such a variety of requirements, from the guy who has a VW 1800 on bike throttle bodies to the guy wanting to update the fuel system on a Sierra cosworth that is chasing 500hp. Having a ECU where the manufacturer sends 2 software update a week is a wrong fit as is a software that only offer 25 break points on the rpm scale when your tuning a 12000rpm rev range and from 1200-3500 rpm it needs mapping every 50 rpm. Difficult
Awwww matt has left haltech I've seen it to often people come in and take over for purely profit it never ends well investors are happy but the consumer never is
Yes it is a loss for them, but a win for Turbosmart. To be fair though RWB generally manages to keep companies running well enough under their umbrella, although for sure there is a trade off for every and any business structure. If you look them up you might be surprised at what they own - Taz. www.racewinningbrands.com/our-brands/
![066: Why Domed Pistons are a BAD Idea from an Ilmor Engineer. [#PODCAST]](http://i.ytimg.com/vi/zhOUPj_DhSI/mqdefault.jpg)
![066: Why Domed Pistons are a BAD Idea from an Ilmor Engineer. [#PODCAST]](/img/tr.png)
![105: Is Honda’s K-Series Really the Greatest 4-Cylinder Ever Made? [#PODCAST]](http://i.ytimg.com/vi/w5oXPYzymFc/mqdefault.jpg)
![Twenty One Pilots - The Line (from Arcane Season 2) [Official Audio]](http://i.ytimg.com/vi/XCz7WUotXs8/mqdefault.jpg)
![054: Unlock Your Car’s Full Handling Potential with Mike Kojima [#PODCAST]](/img/1.gif)
🔥 Win a T-Shirt or Mug by leaving a written review that can include a technical question for Andre or a future guest like Matt to answer - hpcdmy.co/reviewpodcast
Reviews also really help the Tuned In podcast gain more traction, ensuring we can keep bringing you more episodes like this one 🧐 - Taz
This was an incredible episode. I could listen to Matt for hours on end. Thanks for sharing this.
I’ve met Matt, great guy, very technical and a great educator…
Man this was one of the best yet. Matts awesome and having those 2 big players under his belt gives him a huge advantage.
Matt is the man! Super humble and very knowledgeable!!!
One of your best episodes ever
Cheers Paul! We think so too, glad you enjoyed this one as well - Taz.
I’ve met/spoken to Matt a couple of times when he was with Haltech, really helpful and an absolute gentleman!
Great podcast guys 💪🏻
I love the nerdy side to modern ECU tech. The reason Haltech are so successful is they have targeted the right audience by making their tuning software super intuitive and easy to use. Great podcast Andre!
Thanks for digging into the HPT vs. Turbosmart turbo details! Until listening to this, the internet had given me the impression the TS turbos were just re-labeled HPT. Glad to hear they are unequivocally proprietary/bespoke to Turbosmart. I've been in the market for a 58-62mm turbo for a while and I'll be giving the TS 58/62 and 62/62 a very close look! Turbosmart quality is beyond reproach.
Had a great chat with Matt at WTA last month
Huge fan of your channel guys im gonna get all your courses
I was lucky enough to do the Haltech elite training with Matt when it toured in Melbourne Such a great experience thank you Matt I initially felt he was a bit of a traitor leaving haltech so it was great corrected by the context from him I really do appreciate the vision he was was part of at haltech and I was gutted by the RWB acquisition of that iconic Australian brand I'm now really so happy for him and his new frontier. I also really enjoyed the discussion about pressure ratio load axis as a guy doing VGT's on petrol engines pressure ratio tuning is almost mandatory but I've really had trouble forgetting the old Gauge PSI and inHG numbers . Although i understand the science and the math is sound it's still hard for me to get used to . Great work as always Andri and again i will post my tuned in requests Gale banks and Ray hall
brings back memories.. i started with the haltech f3..
I recall drooling over the "new" E6K when it came out. It had a very impressive spec sheet.
In 2017, I bought an already-turboed IS300 with an E6K. It ran pretty good but needed some tweaks. After a huge effort to get the correct (beta) software on my laptop (thanks Haltech), I got to enjoy the glorious experience of working with and tuning an E6K.
For starters - it doesn't run in Windows. It runs in DOS. Modern windows doesn't have a DOS prompt so you need to use a DOS emulator. The whole thing feels like Atari. The unit doesn't have 3D maps. I suppose the processing power required for that was just too much for the ECU. Instead, your "fuel map" is comprised of a bunch of tables of injector PW vs MAP (in PSIG only). There's one for 500rpm, another for 1000rpm, then 1500rpm and so on... Same situation for your spark "map". It makes finding the correct area to tune/tweak very cumbersome when dealing with any kind of transient. Don't forget there's no mouse/touchpad functionality in DOS - so navigating within the tuning environment is keypad only. Hit the first letter in the name of a menu to open that menu - then arrow down to your choice.
Datalog was a .csv file list of numbers. Very limited on what all channels can be logged. No means of characterizing injector dead times, fuel flow, fuel pressure. It's ALL baked into the injector PW in your fuel tables.
On paper, the unit was capable of closed loop fueling based on a wideband - but nobody was ever able to get that to actually work...so the unit was effectively open loop only. That said - I was able to get it tuned in pretty darn decent. I could maintain between stoich lambda ± about 5% all the time (except idle) and a consistent 0.78 to 0.82 under boost.
I could never get it to consistently idle at stoich. The injectors weren't that big (unknown size, but probably about 750cc) but they needed an idle PW of about 0.9ms. Sometimes I could get it to idle nice at stoich with that PW, but other times it would lean out and die. I needed to turn the PW up to about 1.3-1.5ms to get it to consistently maintain a nice steady idle - but that meant idling at 0.85-0.88 lambda. Dunno if that was instability in the injector drivers or just inconsistent dead times and/or inconsistent non-linear zone of injector operation.
Aside from lacking modern features, the biggest issue with the E6K was the "throttle pump" (acceleration enrichment) feature. It didn't work. I guess this was a known issue - but it was a big bummer. Really hurt throttle response when engine was warm, and made it so the engine wouldn't take throttle hardly at all until ECT was at least 120ºF.
It's funny to compare the advertised "spec sheet" vs. the real-world functionality... Gotta start somewhere, I suppose!
Things have come a long way, but we're yet to find an ECU/software that is perfect from any manufacturer to be honest. That's not a mark against any of them or the people working hard on such projects, rather just the reality of it all noting even the major OEMs with infinitely more resources have bugs and issues too.
On the other hand working out such things is part of the attraction to the work for many on that side of the fence 😎
Glad you got a reasonably decent result in the end with that project, sound like a fair amount of work and possibly seems like a waste of time, but I'm sure it also taught you a lot of helpful things you wouldn't know without that project - Taz.
I think Chris hit the nail on the head with his comment on modelling. Fluid dynamics requires huge computing power that just wasn’t available to smaller companies 10, even 5 years ago. It will be very interesting how much further AI will take us. Work has already begun in this space and the results are both exciting and a little scary.
VE/modeled fueling is great. But honestly the old school simple pulsewidth method is pretty OK. No need to characterize dead times, fuel flow @ various pressure, etc. Just more or less PW to get the desired effect. Don't get me wrong - I don't use the PW method anymore - but it did the job pretty well in a very simple way.
@@GroovesAndLands in this respect I was referring to the design of the turbocharger. It’s a fluid dynamics problem and requires some hefty computing power to get any data worth using. I spoke briefly with Willem Toet about the relatively simple aero packages on Fomula SAE cars and he said data on anything less than a 16 core processor is worthless.
@@woznaldo Gotcha. My response is "don't let perfect be the enemy of good". My FSAE days were over 15 yeas ago now; and I've not really followed it since. I assume the aero has improved quite a lot. Anyway, back then it was abundantly clear that almost any aero was a real advantage.
I didn't (and don't) know the first thing about designing aero stuff for cars - so I was interested to ask the other teams about their aero. What I learned is most of them hardly knew anything about it, either - and had done rather primitive CFD and analysis in designing theirs... Yet they obtained excellent results that were clearly visible even in very low-speed skidpad event.
Watching, it was obvious that just a diffuser generated more grip than nothing at all, and a rear wing + diffuser outperformed just a diffuser, etc.
All this to say, coming from an automotive/race engineer that's been at it for 20 years - don't let perfect be the enemy of good.
@@GroovesAndLands all good. The FSAE aero issue was still present when I did my last comp 8 years ago. Too many teams put effort into aero but, their cars weren’t reliable and couldn’t finish the endurance event. Time would be far better spent investing in the team (people) and building a reliable car. We digress. Your point is taken.
@@woznaldo Heard! My school was small and FSAE did not count for credits. As such we had a small team. There were 15-20 total members but only 5-6 "core" members we could count on to get anything real, done. Thus, we had to focus on the core aspects of the car. We certainly didn't have the bandwidth to get into aero.
FSAE was an amazing experience but looking back - I/we took the wrong things seriously. We focused on designing/building an awesome machine when we should have focused on the logistics and management of excellent teamwork. Achieve that and the machine will come.
That's how I think about wiring individual sensors at a time 2 or 3 or 6 wires at a time etc
Straight gate was used in aircraft years ago.
Yes i have experienced the copy versions about few years and they leak you end up buying twice just get what you need and not worry about it
Great podcast but it's a shame that you didn't get to discuss cold side boost control. I only saw this for the 1st time in a Steve Morris Engines video the other day although he says that somebody else (1 of his customers) has been running this system for the past year and it uses a Turbosmart stepper motor controlled butterfly valve system fitted to the cold side with nothing (controls, valves, etc.) on the hot side. Interesting concept. EDIT: Just watched Steve's latest video and it seems there are reasons why the above isn't viable and still needs a wastegate on the hot side.
@@phillipmcintosh6489 even more reason for a part #2!
I think maybe its to keep the compressor and turbine spinning in its optimum range (waste gate), whilst maximising boost intake (dbw bleeder). Especially common in smaller motors, where if you cram so much boost into the motor, its gonna some drive Intake air temps high and reduce efficiency. You see it in rally cars, but I would love to hear whether or not this is a valid answer in a part 2 :)
Such a good and informative chat by a couple of legends 👏
Started out with f9a. Remember ordering my e6k from Fausto of interco trading.
My favorite thing about Haltech is that you can edit axis parameters to make custom tables . I feel like other ecus dumb it down too much and don’t include this .
Awwwww Scotty and Matt were friends I would love to meet them 2 and Andre I've learned so much from you guys
For the guy running a shop, the customers coming in the door have such a variety of requirements, from the guy who has a VW 1800 on bike throttle bodies to the guy wanting to update the fuel system on a Sierra cosworth that is chasing 500hp.
Having a ECU where the manufacturer sends 2 software update a week is a wrong fit as is a software that only offer 25 break points on the rpm scale when your tuning a 12000rpm rev range and from 1200-3500 rpm it needs mapping every 50 rpm.
Difficult
👍👍👍👍
Awwww matt has left haltech I've seen it to often people come in and take over for purely profit it never ends well investors are happy but the consumer never is
Boeing, anyone?
😂lol
Yes it is a loss for them, but a win for Turbosmart. To be fair though RWB generally manages to keep companies running well enough under their umbrella, although for sure there is a trade off for every and any business structure. If you look them up you might be surprised at what they own - Taz.
www.racewinningbrands.com/our-brands/