|
Post by racket on Mar 14, 2021 18:59:53 GMT -5
Hi Scott
One other thing , I noticed you had a TOT of 633 C even with the "oversized" A/B nozzle, 633 C is a reasonably warm temperature at 3.1 :1 PR when fitted with a "dry" sized nozzle , at 4:1 PR you'd probably be closer to 750C, your A/B sized nozzle is producing "dry" sized nozzle backpressure .
You may find that you'll need to increase your A/B nozzle size when running "wet" otherwise the backpressure could end up being too great for the gas producer if you trying running max temps in the A/B.
Cheers John
|
|
slittlewing
Senior Member
Joined: November 2017
Posts: 458
|
Post by slittlewing on Mar 15, 2021 5:42:33 GMT -5
Thanks for all the feedback guys!! Ausjet I do indeed have PWM control over the A/B Fuelling. The next job will be to drill out the restrictor for the AB fuel line to get more flow for the new turbo (currently 1.2mm for old turbo, and I was using about 90% PWM). Then I will be experimenting with how much fuel to give it. Previously I had to program a ratio in the ecu which was a pain, but now I can alter it "on the fly" since adding pots on the handlebar. Thomas cheers mate, the top cap is 5mm stainless, although I sort of now wish I had gone thicker (or even better, used a dome cap with an airgap). The speed of the spool shocked me a bit John thankyou for your wisdom, I hadn't considered the AB nozzle might now be sized more towards "dry thrust" with the increased flow. My fuel pump is still providing 10 bar at max throttle (both for the old 13.5 and newer 17.5GPH injectors) so I can be confident the engine is burning 30% more fuel than before and presumably flowing that much more air. Is there any rule of thumb or calculations on A/B nozzle size I could apply for maximum "wet" thrust , if I expect to have to cut the outlet cone back to a larger diameter? Currently AB Outlet nozzle is 71mm Dia. Old turbo: 63mm Comp Inducer (AB Outlet = 127% Area) 72mm Turb Exducer (AB Outlet = 97% Area) New turbo: 69mm Comp Inducer (AB Outlet = 105% Area) 77mm Turb Exducer (AB Outlet = 85% Area) I always had a feeling the outlet nozzle was too big for the old turbo (because it had almost no thrust at full throttle, without afterburning), so maybe it'll be OK for the new one as it is? However I think the A/B will certainly be "on the limit" design wise, due to now being slightly undersized (in terms of overall diameter/length and dump zone) and I expect it to run hotter than before with the increased fuel... Hope the seat can take it without melting haha. Excited to do more running Cheers Scott
|
|
|
Post by racket on Mar 15, 2021 16:28:24 GMT -5
Hi Scott
At 71mm for the A/B nozzle and a 69mm comp inducer you should be pretty close to right, I think I'd try it as is but with very careful attention to the TOT as you ramp up the A/B fuel.
Ideally we need to "tune" the engine with a "dry" nozzle , once we have the optimum nozzle area , we double it to account for the doubling of absolute temperature exiting it with an A/B , but because gas velocity will only increase by 40% ( square root of 2) we divide by 1.4,.... eg, your dry nozzle would be say 60 mm dia , 2827 sq mms , double that for 5654, then divide by 1.4 = 4039sq mms = 71.7mm Wet ...............duh , I think thats right , just came back from early morning fishing trip .
A very rough guide for A/B nozzle is to have it the same diameter as comp inducer
Looking forward to the next vid :-)
Cheers John
|
|
slittlewing
Senior Member
Joined: November 2017
Posts: 458
|
Post by slittlewing on Mar 15, 2021 17:59:58 GMT -5
Many thanks for the explanation John - that's much appreciated. I notice the Kart AB nozzle is sized pretty much same as its comp inducer so that rule of thumb also makes good sense. Early morning fishing trip... awesome!! Glad to hear you are living the good life Hope you are barbequing your catch for lunch. I drilled out my AB restrictor tonight (36% more fuel flow than before) and finished writing the ECU code for the new afterburner fuel PWM (still based on P2 Pressure but open loop, gain set via handlebar pot). I also incorporated some compensation based on battery voltage, after I noticed previously that the A/B seemed to have less power after 10 minutes or so running when the lead acid was running low . I will make sure to keep TOT below 750. I am very tempted to try it again this weekend, but the A/B volume isn't exactly neighbourhood friendly!! Still, I am moving house in a month so maybe that's tough luck haha. Cheers Scott
|
|
|
Post by racket on Mar 15, 2021 22:35:23 GMT -5
Hi Scott
Yep , Andy did a great job of calibrating the A/B nozzle .
Fishing was good this morning , I was only there a short while and ran out of bait , but had 3 nice fish in the bucket , its the start of the Tailor season , so early mornings over winter for me .
Sounds like another A/B trial is in order ;-)
With your rather hot combustor cap , actually excessively hot endcap :-(..... I've been trying to think of a solution , so this morning while having a chat with Andy about it (he has more experience with that design of combustor than me) we concluded that some short stub tubes inserted through those 4 or so holes at the very top of your flametube ( between the Primary holes and endcap) and angled at the endcap would solve the problem , they only need to be 12- 15mm long and welded to the FT wall.
The current holes are a tad too far axially from the endcap for their outflows to cover the cap in a cooling blanket of air
All the best with the next testrun :-)
Cheers John
|
|
slittlewing
Senior Member
Joined: November 2017
Posts: 458
|
Post by slittlewing on Mar 17, 2021 11:41:20 GMT -5
Awesome... Imagining weekday fishing in the sun is making me very jealous đ Thank you very much John for considering how to solve my hot topcap issue, along with your friend Andy. I think your suggestion is excellent so I have implemented it. I needed to remove the topcap anyway before the AB test because there was some minor air leakage/fuel weeping after failed start across the 1mm copper gasket I am using. Previously I put a layer of flange sealant on each side which worked well to seal, but appeared to be burnt on removal, hence trying it âdryâ for the last fireup. As I donât intend on removing it again any time soon I thought best to put the flange sealant on again! The 4 holes you mention are 8mm dia, I opened them up to 8.5 and tried to angle some straight tubing stubs towards the cap, but they were pointing towards the kero injector at maximum angle, rather than the cap surface. So I got my kart âbending jigâ back in action using a sand fill inside the tubes again and blowtorch, bent some tight angles and cut the tubing to wash over the topcap appropriately: Hopefully this will improve things a lot! As the holes reduced from 8mm to 6mm dia (accounting for tube wall thickness) I opened up the other 4 tiny holes on the same pcd to compensate. Also found a place in the countryside that is willing to let me run the afterburner on Saturday, fingers crossed things work properly!! Many thanks, Scott
|
|
|
Post by racket on Mar 17, 2021 16:13:42 GMT -5
Hi Scott
That should definately help the situation , I'll be interested to see the results :-)
Good news about the test site becoming available ..................video please
Cheers John
|
|
slittlewing
Senior Member
Joined: November 2017
Posts: 458
|
Post by slittlewing on Mar 18, 2021 7:34:16 GMT -5
Fingers crossed! I am considering whether to take an angle grinder with me so I could make changes to the outlet nozzle if required on the day, but that's probably a bad idea.
I guess it could become a bit of an endless optimisation loop - As its already 633TOT without AB at 3.1:1 PR, if the AB fuelling had to be kept quite low to keep temps within 750, it would suggest making the outlet nozzle larger (lets say 75mm, +10% area). But then, presumably more AB fuelling would be required just to get to the same thrust level as before, due to the lower gas velocity!
Cheers
Scott
|
|
|
Post by racket on Mar 18, 2021 16:07:46 GMT -5
Hi Scott
But , you'll be limiting potential thrust if the A/B isn't burning enough fuel to get to max temp in the A/B , thrust increases as the square root of temp increase , you need the A/B to be as hot as possible.
Cheers John
|
|
slittlewing
Senior Member
Joined: November 2017
Posts: 458
|
Post by slittlewing on Mar 20, 2021 12:33:40 GMT -5
So a report after todayâs testing... more trials and tribulations and learning âbit by bitâ haha! Firstly the bike wouldnât fire up with the same settings as last weekend. I kept adding more fuel and it eventually became apparent that the bike would not idle below 1.2 to 1.3bar P2 (33% Fuel pump PWM). Winding the idle Fuelling down slowly below this the engine would not sustain and immediately cut out. Stabilised TOTs after 30sec or so at each operating point were 485 at 1.3bar P2, 630 at 1.7bar, 700+ at 2.1bar (aborted once they got that high as I have a 6mm thermocouple inside a thermowell so it takes a while to even register the real temp which would have been higher than 700) By constrast, last weekend I had the idle set happily at 0.6bar P2 (18% PWM) and the engine still sustained down to 0.3bar when the fuel tank emptied and fuelling died off slowly. So it seems the air from the top cap fuelling tubes is somehow disturbing the fuel mist from the kero nozzle at low fuel pressures. The physical change from last week was that the 4x 8mm holes near topcap were changed to tubes of 8mm OD (6mm Id) and the 4x 2.5mm dia holes on same pcd were increased to 5.5mm to compensate for the loss in area due to the tube wall thicknesses. I removed the top cap and drilled out the new cooling tubes (only drill to hand/borrowed was 9.5mm). Thus now leaving 4x 9.5mm holes and 4x5.5mm holes on the flame tube wall. This improved the situation marginally, was able to idle down to 1.0bar P2 (25% Fuel PWM) but no lower. This was too high a flow to ignite the afterburner despite many attempts. The afterburner did light once as the engine was spooling down whilst switching off, proving it will go once the flow is reduced Throughout the day the topcap ran cool with no glowing. But I need to get back to a low idle. What do you think about sealing up all the topcap cooling holes and instead using maybe 10x 2.5mm holes as close to the top cap as possible, which would have less penetration of airflow into the CC and hopefully therefore not affect the idle flame front/fuel spray? Other info - I opened up the AB nozzle from 71 to 75mm ID. This brought stable TOTs to: 385 at 1.2bar, 480 at 1.7, 620 at 2.0bar. Going above 2bar seems to require a lot more fuel pressure and generate a lot more TOT, maybe choke region or the fact that the FT holes are still based on the old 63mm inducer. (7bar fuel pressure makes 2.0bar P2, 10bar fuel pres makes 2.2bar P2) In terms of good points, the bike was started quickly and successfully many times once fuelling was increased to the sustain level, there are no leaks anymore, and the oil pressure is now PID controlled successfully and providing stable 4bar automatically through a range of oil operating temperatures so lots of progress!! Here is a video of the first successful start where it was on the cusp of self sustain and then fuelling was increased and it idled at 1.2bar p2: youtu.be/ndbz0jBb3gcthanks for all the continued support and guidance Cheers Scott
|
|
|
Post by racket on Mar 20, 2021 18:09:14 GMT -5
Hi Scott
It might be a case of no glowing endcap causing the problem rather than a small quantity of air blowing across it .
Another indicator is your TOTs ..............385C at 1.2 bar P2 which is nice and cool and what you'd expect with 75mm jet nozzle , but your temp of 600C at only 2 bar , a mere 0.8 bar increase , is indicating something is wrong, especially as your fuel pressure increase from 7 bar to 10 bar only gives few more psi of P2 , .................delayed combustion .
You may need to go back to your redhot cap settings to "compensate" for other flametube shortcomings ..............an easy solution to test the theory :-)
I'll have a check through your other earlier data .
Cheers John
|
|
|
Post by racket on Mar 20, 2021 20:26:42 GMT -5
Hi Scott
Maybe check your fuel nozzle is spraying correctly.
I really can't see those tubes causing a problem , most commercial engines have air delivery through the end cap of the FT , the air is drawn into, and combines with the fuel spray as it spreads .
Cheers John
|
|
|
Post by finiteparts on Mar 20, 2021 22:32:51 GMT -5
Hi Scott,
My first thought on this is that the modifications to the PZ air split has destabilized the flame anchoring and thus you are experiencing harder light-offs and low power instability/blow-out. I would suggest that the glowing head-end was a sign of a significant flame anchoring at the inner surface and thus the lack of the glowing surface suggests strongly that the flame anchoring has moved farther down the flame tube, which in my opinion points to "weaker" flame anchoring. How would the tubes cause this? My thought is that the tubes essentially disturbed and segmented the recirculation toroid that would have formed between the fuel nozzle injection and the outer liner.
I do have to admit that these style combustors are my least favorite. The only suggestion that I would have is to close off the combustion liner inside the casing and thus build in an airgap between the endcap and the liner dome. It would be relatively easy to weld bolts to the liner dome and use jam nuts to "mount" it to the endcap.
As for the holes, I would suggest welding a few washers on to return to you previous hole size. The trick with these style combustors is that it is hard to determine how much of the air entering the the downstream holes is being drawn into the headend region due to the momentum of the fuel injected in the center. Since the centerline flow is traveling axially down the combustor due to the momentum of the fuel, the air entering the liner holes ends up moving towards the headend to form a recirculation region that can be marginally stable. I suspect the air tubes injecting the air closer to the centerline, essentially reduce the recirculation zone size, at least locally. Larger primary holes have greater penetration and thus will act to reduce the size of the recirculation zone also.
Good luck!
Chris
|
|
|
Post by racket on Mar 20, 2021 23:20:42 GMT -5
Hi Scott
Are you running the Monarch 80 degree PLP nozzle ??
If you are, it might pay to change to the Monarch PL version .
The PLP is a solid pattern whereas the PL is hollow cone and should be a better proposition, providing recirculation within the "hollow cone" to keep flame alive rather than an "external??" recirculation which could be the cause of your hot lid .
Its an easy test to do and hollow cone patterns are more of a "gas turbine" type pattern .
Cheers John
|
|
slittlewing
Senior Member
Joined: November 2017
Posts: 458
|
Post by slittlewing on Mar 21, 2021 5:45:25 GMT -5
Thankyou for the interesting replies and information, it seems like going back to the original holes is on the cards. Chris, I think you have articulated what I had in my mind in terms of the tubes splitting/interrupting the flow in that area. I have to say that 4 years after designing this flame tube, I would now do it differently and use a inner skin with full air gap all the way around. In terms of "retro fitting" a plate or something similar to provide that airgap, it is tempting to get one of these: www.kilnerjar.co.uk/0025411-stainless-steel-strainer-funnel/and see if its a good match for my FT. Unfortunately I think it would be tricky getting a good fit to the FT wall (which is no longer perfectly round after heat distortion) and also difficult welding through the skin of the FT to fix it in place. Maybe this could be a longer term option if I re-made the top cap and FT assy from scratch. Attempted to show the idea below:
Edit: I think I could probably make this work with the existing FT if I had an air gap between the dome and FT walls, and suspended it off the topcap with "stilts" (yellow) welded through to the outside? : John, I am using the 80deg PLP nozzle you mention - thanks for suggesting the PL type. I can get one of these but they only come in 60 degree spray angle at 17.5GPH. At the moment, my main primary holes are situated where the 80deg spray would theoretically hit the FT wall. Going to a 60deg would mean they no longer do this, see pic below. Do you think this would still be OK? In terms of the possible delayed combustion above 2 bar P2, one other thing to mention is that my FT hole areas are still based on 63mm comp rather than the new 69mm. Not sure whether it would be worth opening out the last set of small (3mm) primary holes which would also roughly line up with where the 60deg nozzle spray hits the wall? pic below: One other question.. I realise that commercial can combustors use swirlers at the injector position, which helps mix the fuel and air in a spiral fashion and provides faster combustion - enabling a smaller CC to be used. If I had added a swirler in my design with its already "large" jetspecs style dimensions, would it have given me lower TOT's? Or is it the case that for a given fuel flow and thus thermal energy, provided the burn is "complete" the TOTs would be the same? Cheers Scott
|
|