Holset HX50 Supermoto Jetbike

[racket]

Hi Scott

You're getting the start up nicely sorted and the afterburner is making the right sounds :-)

But I have to ask ................who is FLOKEY ??

Development progressing well, but that combustor fuel burn rate has me concerned .............LOL, one concilation though is that any unburnt combustor fuel will have a second chance by contributing to the A/B fueling .

Cheers
John

[racket]

Hi Scott

Further to my last email.................I've been checking some of your earlier build thread pages and noticed your turb exducer size as 88 mm ~3.5" , thats a HUGE exducer for a 2.5" comp inducer , some 90% greater in area .

I also noticed that your turb scroll originally had some sort of "wastegate??" incorperated into it .............how did you "modify" that , I noticed a blanking plate on the outside , but what happens inside ??

Your turb scroll also appears to be "FAT" , which along with the large exducer area and most likely inducer area could be allowing a very large increase in mass flow of air , well into the choke region of the map which would account for the increased fuel flow demands above what would normally be required for a 2.5" inducer .

If you are flowing well into the choke region, be VERY CAREFUL about attempting to go to higher P2s as the tip speeds/RPM could be getting excessive , even to the point of overspeed www.cummins.com/sites/default/files/files/brochures/turbos/Holset%20HX50%20Comparison%20Test_lowres.pdf the comp map speed lines become "vertical" at choke , meaning a 1 Bar P2 could be produced at the same RPM as a 3 Bar P2 with reduced mass flow .

Can you please check your RPM and/or your T2 to get an idea of comp efficiency , you should be compressing at better than 70% , but if at the choke it could be at 50% , I once measured a 55% effic with my TV84 when it was equiped with an oversized turb stage .

Cheers
John

[slittlewing]

Hi John,

Thankyou for your informative post and nice job finding that holset compressor map!!
First and foremost though, FLOKEY is our beloved cat 😂



I have to be honest, at the point of buying the turbo 5 years ago I had no idea of the ideal turb requirements but I knew it should be from a diesel. It is indeed a wastegated twin scroll housing and it seems to be a quite rare 25cm2 FAT housing because I have tried to since search for a replacement /new spare and never found one the same since. For the wastegate I removed the external rod, put a closing plate and there is an m8 bolt that pushes on the back of the lever and poppet valve and keeps it shut. The lever is still in place and goes through the housing.

I have been under the impression that fatter turbine scrolls are better for our engines due to having less restriction, is this not the case? Maybe I am lucky for not going above 2 bar P2, I will get a tachometer and T2 probe to use in one test run 👍👍

I don’t think there is overfuelling that is then combusting in the AB tube because it runs cool and quietly and without smoke during normal operation (AB off)

Many thanks for all your advice, we are very lucky to have you on this forum!

Scott

[racket]

Hi Scott

LOL....it looks like Flokey is well pampered by the "staff" ;-)

25cm2 is a large throat area supplying a relatively large turbine wheel , ideally we need a turb exducer area ~20-30% greater than comp inducer area to produce a "balanced" match between flows, at >90% greater its going to cause problems.

Because we don't have a "swept volume piston engine" to control flow from the compressor stage we rely on the turb stage sizing to do that job , its "Corrected Flow " , your turbo would have been setup for low boost and minimal backpressure on the diesel engine to optimise engine fuel efficiency .

Generally speaking a fat scroll isn't a problem if the turb wheel is at the correct flow size , as the turb wheel exducer can choke and control compressor flow before it gets too great, its a matter of getting the right balance.

I'll be surprised if your TOTs remain so low with a sheathed thermocouple installed and the A/B fired up , the high flow rate from the comp at poor efficiency will require a greater amount of power from the turb wheel leaving less pressure in the A/B to produce velocity through the A/B jet nozzle ................be very careful about turbine temperatures once the A/B is ignited , you may need a larger nozzle diameter to keep temps down .

Some RPM readings will soon indicate where we are on the map ...................do you have a comp exducer diameter ??

Another HX55 comp map I have gives a RPM of ~90,000 for a 3:1 PR - 2 Bar P2 , but I'm not sure about that particular wheel's dimensions , you should be seeing a T2 of ~160 C at 2 Bar P2 , if closer to 195 C we're getting into the choke region .

Cheers
John

[racket]

Hi Scott

Further to the discussion on matching flows .

The Garrett Gt47 turb stage runs a 93/84 mm wheel with an exducer ~10% smaller than your 88 mm area , the turb map for the GT47 turb stage www.garrettmotion.com/wp-content/uploads/2018/05/Turbine-Flow-GT47.jpg indicates a max flow of ~55 lbs/min CORRECTED FLOW with the 1.39 A/R housing , if we add on 10% for your bigger exducer and housing throat , say ~60 lbs/min Corrected.

Now Corrected Flow = Actual Flow times sq root of absolute gas temp divided by standard temp , all divided by the Pressure Ratio going into the stage .

So 60 lbs/min = act flow X sq root of 1173K/288 K divided by say 3:1 PR

60 = act flow X 2.018 div by 3

60 X 3 div by 2.018 = actual flow

Actual Flow = 89 lbs/min

Thats interesting , thats ~50% more mass flow than the ~60 lbs/min that a 2.5" inducer normally flows in the best efficiency island on the map www.garrettmotion.com/wp-content/uploads/2019/06/G30-900-Comp-Map.jpg 62/76 mm G30 comp map., but roughly what the bigger Gt47 flows at a 3:1 PR www.garrettmotion.com/wp-content/uploads/2018/06/GTX4709-5009-80mm-comp-map.jpg

This would sorta account for your large fuel burn rate.

Yep, don't try going to a higher Pressure Ratio until you check your rpm , with a used turbo of unknown heritage best not to go past a 1450 ft/sec tip speed , a modern new high quality turbo can be pushed harder .

Hope this helps explain things :-)

Cheers
John

[slittlewing]

Thankyou John for putting that analysis and calculations together, im very grateful and I am going to go over them a few times to properly learn the maths behind it. I will order a tach!

I know the engine is burning all 13GPH as there is no smoke under normal operation and no pooling in the chamber/leakage from drain. As I see it, I assume the engine is therefore working well. One thing of note, I believe I got the inducer measurement wrong when i designed the engine, oversize by maybe 3mm meaning I have a small percentage extra air in the combustor. Not sure if this accounts for anything!

In terms of measuring T2, would a TC stuck to the outside of the combustor delivery tube work or does it need to be a probe in the airstream?

The turbo does have damage to some of the turbine tips as previously photographed but it did have a rebuild tag on it and the bearings are silky smooth without play.

Thankyou for your continued support,

Scott

[racket]

Hi Scott

If you need more explanation , only too happy to help out :-)

If your actual mass flow is over in the choke region then your flametube hole areas are now quite undersized , we normally size things assuming flow is in the best efficiency island of the comp map as this will generally produce the best power outcomes, which for the GT30 www.garrettmotion.com/wp-content/uploads/2019/06/G30-900-Comp-Map.jpg would mean a flow of say 55 lbs/m at a 2.5 PR at 115,000 rpm , but with and oversized turbine stage that flow could go out to 82 lbs/min at a 2.5PR but with rpm of 145,000 rpm ......................same PR of 2.5 but mass flow increase of ~50% and a RPM increase of 26% .

The GT30 has a 62 mm - 2.44" inducer of 4.68 sq ins , so assuming 60 lbs/min at a 3:1 PR as optimal then 60/4.68 = 12.82 lbs/sq in of inducer which is roughly the sorta flow we expect when sizing the flametube holes .

But if that flow was pushed out ~50% at choke , then the lbs/sq inch goes up to ~19 way in excess of normal design .

Your current flametube hole areas could be grossly undersized, as well as your flametube cross sectional area , it really needs to be proportioned for a 3" inducered comp , the "undersized" holes will be producing greater penetration/turbulence within the flametube and aiding combustion which appears to be OK , but at a greater pressure drop across the flametube wall ..............no free lunches :-(

A tach and T2 reading will soon point us to whether or not we're currently going in this direction.

The T2 needs to be taken in an area of modest air velocity so that any "dynamic" temperature is low , high speed airflows can produce unreliable numbers even with an internal flow position, though for our current situation we aren't too worried about the last couple of degrees of accuracy , a well insulated surface TC should give us a reasonable pointer to whats happening especially if several readings at different P2s produce a trend ...........an airspeed of a few hundred feet per second in the delivery tube only represents several degrees C of dynamic temperature , we might be looking at 30 degrees higher than optimal .

Looking forward to your "numbers" ................LOL, I like numbers :-)

Cheers
John

[slittlewing]

Hi All!

Firstly John - Thankyou for the informations and workings out. I have been through them and understand what you are saying about the choke operating point on the right hand side of the map appearing to be where I am if the mass flow is that high. The tachometer I ordered has arrived 👍
One question... in addition to possible flame tube hole rework, what advantage does fitting a billet compressor wheel provide (was thinking it might rebalance things against the big exducer?)

A few other minor developments - I swapped out the compression fitting on the AB (brass and copper olives) for a proper hydraulic steel one after the small fire Last time. I have a stainless tube and high temp push in tee on the way for the drain, which should cure the melted plastic bits.

I ordered a thermowell from amazon that fits my 1/4 bsp boss in tailpipe. It said on the listing it was 5.1mm hole, I ordered a 5mm new TC. It turned up and it’s 6mm, so I have put my old TOT probe in there which appears to be working again (but I wonder for how long!)





After posting the afterburner video on Facebook it got pretty popular and I have started a branded page. A french guy has since been reposting pictures and video of my Jetbike as his own work, so I have been forced to blanket watermark my previous pictures of the bike on this forum, sorry about that! (I have a sticker coming for the bike to promote the forum and will be talking about all the support I have had in future “how it’s made” type video)

I have taken a punt on the seat and panels not melting and they are now fitted and covered in insulation.
Hopefully this weekend will provide the opportunity for another test with more instrumentation!

Cheers,

Scott

[turboron]

Scott, it sucks that some creep would post your work as his own. What is the world coming to?

Thanks, Ron

[slittlewing]

Thanks Ron, I guess there’s a lot of people that don’t care about reloading videos and photos, it takes them a few seconds to imitate years of work eh!!
Still, I’m the one with a Jetbike to play with 😁 also thanks to all of you!
Cheers
Scott

[racket]

Hi Scott

Don't worry about changing comp wheels, even if the new wheel is a "hi flow" one with higher inducer angles , the turb stage will still be too large ................better to work with what you have , or change the entire turbo for one with a better match between comp and turb ..............I'd go for staying with your present setup and use it for a learning tool for a bigger and better project down the track :-)

Cheers
John

[slittlewing]

May 7, 2020 4:33:18 GMT -5 racket said:

Hi Scott

LOL....it looks like Flokey is well pampered by the "staff" ;-)

25cm2 is a large throat area supplying a relatively large turbine wheel , ideally we need a turb exducer area ~20-30% greater than comp inducer area to produce a "balanced" match between flows, at >90% greater its going to cause problems.

Because we don't have a "swept volume piston engine" to control flow from the compressor stage we rely on the turb stage sizing to do that job , its "Corrected Flow " , your turbo would have been setup for low boost and minimal backpressure on the diesel engine to optimise engine fuel efficiency .

Generally speaking a fat scroll isn't a problem if the turb wheel is at the correct flow size , as the turb wheel exducer can choke and control compressor flow before it gets too great, its a matter of getting the right balance.

I'll be surprised if your TOTs remain so low with a sheathed thermocouple installed and the A/B fired up , the high flow rate from the comp at poor efficiency will require a greater amount of power from the turb wheel leaving less pressure in the A/B to produce velocity through the A/B jet nozzle ................be very careful about turbine temperatures once the A/B is ignited , you may need a larger nozzle diameter to keep temps down .

Some RPM readings will soon indicate where we are on the map ...................do you have a comp exducer diameter ??

Another HX55 comp map I have gives a RPM of ~90,000 for a 3:1 PR - 2 Bar P2 , but I'm not sure about that particular wheel's dimensions , you should be seeing a T2 of ~160 C at 2 Bar P2 , if closer to 195 C we're getting into the choke region .

Cheers 
John

I finally got around to another fire up last weekend (been heavily distracted by ripping out and refitting new bathroom which has been painful!!). The seat and rear plastic panels were fitted (With insulation) and the aim was to see if anything melted and record temps, including T2. I bonded a thermocouple "copper pad" probe onto the outside of the combustor delivery tube, just after my silicone flexi joint. The delivery tube is 1.6mm stainless so there will likely be some lag in the reading!

The engine fired up easily once again althought i spilt alot of diesel over the bike whilst "filling" it, causing some smoke and fire anxiety after starting. I throttled it around whilst nervously checking nothing was melting. I held it at full throttle (2bar P2 or 3:1PR) for about 20 seconds and the T2 spiked as high as ~165 degC but judging by the gradient it was nowhere near stabilised! Picture below:



According to my turbo number, my comp is 63mm Inducer and 94mm Exducer.


I then tried a second fireup and attempted to light the AB, again to check seat plastics. I had increased the fuelling since my last run as an experiment.. (AB Fuel Delivery Pressure = P2 Pressure * Gain, previous gain 2.5, new gain 3.0). It turned out to be too much fuel, with the AB lighting with a foot long cone flame and then going out and relighting a few times. A giant plume of white smoke was made and at least one person in the neighbourhood came and complained haha.
I
So I need to revert to my old AB Fuel Gain and probably repeat the T2 measurement at 3:1PR for at least 60 seconds to try and get a steady state reading do you think?

My TOT probe died again despite being in a "thermowell", which isn't too suprising given I had re-used it and it had already failed at high temp previously when unsheathed.

Cheers

Scott

[racket]

Hi Scott

165 C for T2 isn't too bad at 3:1 PR if it was a "hot" english day with "high" ambient temps going in , assume a 20deg C day then a 165-20=145 rise =74.5% effic not too bad , at 25 deg C - 298K =78.5% very good but not realistic, I think you were right about the temps still climbing .

Cheers
John

[slittlewing]

Ok Cheers John, it was certainly quite hot (about 25deg) but with 165 not being a "topped out" I guess the efficiency numbers will drop a fair way..

I intend to repeat the test this weekend and will go for 60 seconds at full throttle. I ordered another new TOT K-Type last night, fingers crossed it comes by the weekend too, as "flying blind" isn't very confidence inspiring!!

Cheers

Scott

[racket]

Hi Scott

If you can get a reading that produces an efficiency of 72-75% you'll be OK .

Yep , gotta have reliable TOTs

All the best with the next testrun :-)

Cheers
John