|
|
Post by wannabebuilderuk on Apr 25, 2021 13:53:32 GMT -5
Is that another project hidden under the bench there Scott? đ  |
|
|
|
Post by racket on Apr 25, 2021 20:29:35 GMT -5
Hi Scott
Thanks for the extra pics :-)
Is the rub in any one quadrant of the NGV wall or all around it ??
The wall at that point should stay relatively cool as there copious amounts of cooling air travelling up the other side on its way to the Tertiary holes as well as a blanket of cooling air travelling down the rear of the wall
If the colouring of the rub areas was caused by the rubs and the rest of the wheel "recess" is uncoloured then we might need to look at fueling as Chris suggested .
I had abberant heating of my 12/118 engines NGV due to blocked injectors , this required the remaining injectors to increase flows to the point where combusting that fuel in the Primary Zone was impossible and a hot streak was the result , the addition of the extra "high pressure" filter ( thanks Smithy) just prior to the fuel entering the engine cured that problem , the low pressure suction side filter didn't filter rubbish coming from within the pump .
The "flat" rear wall of the outer can is a concern , but Anders didn't have problems with his engine , though I prefer a concave rear wall and slipjoint to remove all axial loads from the NGV .
The problem with rubs is that they tend to "increase" , they might start with just the smallest touch but that little bit of raised metal "grows" leading to larger and larger scabs forming , you may find that when you remove those scabs the clearance will seem enough.
The front of the turb wheel will need a side grinder in the lathe to true up the blades , bit too hard for turning , probably better to grind off any "wobbles/high spots" on the wheel blades than machine more metal out of the NGV wall to accomodate those high spots , that way the turb efficiency won't be as badly compromised by having to provide more clearance than is absolutely necessary .
At least your exducer shroud wasn't affected .
More pics please :-)
Cheers
John
|
|
|
|
Post by slittlewing on Apr 26, 2021 8:56:33 GMT -5
Chris, Many thanks for the very knowledgeable reply. That all makes perfect sense, in regard to the transfer of material, and thus melting temps and possibility of cracking on the turbine. When I hit the kill switch at the end of the test due to the smoke, the TOT gauge froze at its last value which was 1000 degrees - presumably what John is referring to when he mentions the combustor becoming an afterburner! In the video you can see smoke suddenly gushing around the V-band. I think I was reasonably quick to kill the engine as it was dragged down so I hope the turbine wheel will be salvageable. If not I guess I will have to buy another. I will check the fuel injection needles as you recommend, and once I have tried to blend out any damage on the rear of the turbine I will get it UV dye crack checked as we have those facilities where I work  Good spot Ben, that's a possible "Stage 2" for this kart which John made  John, My pleasure and thanks for the info. Last year I spotted some light rubbing marks on the back face of NGV which I presumed to be historical. From what I can see there is rubbing in only one quadrant but I will confirm later, I was missing a large imperial socket to remove the turbine and I am borrowing one from work tonight. I tried to measure the existing clearance with feeler gauge pushed behind a blade. The biggest I can get is 0.35mm, but thats nearer the hub than the tip. Additionally, that's without oil pressure so the shaft has some extra room to tilt, providing a bigger gap for the feeler gauge. No doubt with forward thrust on the compressor wheel and oil pressure on the rotor the clearance would be much less. Excellent point about a small rub tending to "increase"! Thanks to all of the above advice, my general plan of attack seems to be: -Try and reface back of turbine wheel assuming there isn't major damage. I can put the shaft in my pillar drill at a high speed, and then use a dremel to very gently grind and "blend" out damage and simultaneously ensure the back of the blades are "square". -Crack check turbine wheel -Strip rest of engine for inspect, check fuel needles, replace bearings whilst I have it apart -Remove at least 0.5mm from NGV back face. I don't really want to have to repeat this procedure so if I should remove more to be on the safe side then please advise -Rebalance and reassemble In the long term I may look to retrofit a concave surface to the rear wall. It would certainly help strengthen and resist any bending loads. I take the point that Anders engine also has a flat rear wall, but his NGV also looks to be made of thicker material (10mm I think was mentioned in his build thread). More pictures to follow, Thanks so much for taking the time to help, Scott |
|
|
|
Post by andym on Apr 26, 2021 11:11:05 GMT -5
I gave you a socket for the turbine .... Still no video of failed starts with over fueled engine... dread to think how hot the turbine section got... there would of been at least twice the fuel going into the engine... with the main pump running at 20 volts...and perhaps fuel burning inside once the engine shut down...
In some of your pictures it looks like melted material on the back plate....
Personally as I have done in the past... I would of stripped the engine down to check after a few questionable starts....
John and Chris... you have both made some very good comments.... but for me... I need to know is this cause by bad design... or the engine running outside its parameters... that why I want to see the video that has not been shared
|
|
|
|
Post by slittlewing on Apr 26, 2021 11:45:17 GMT -5
That you don't see why it should be done the way I explained... that you think you no better.... there is a reason to do it the way I explained... but not going to waist my time any more.... you clearly know better than the person that designed and built the engine Also, I have been at this some time, the ngv back plate so called rub... was not the reason for the fail... would see much more damage... and there would be noise heard in the test Yes I do know better, and yes there are engine start procedures where I work. Guess what... they don't involve putting a tangential shock load on the crankshaft and rapidly spooling it to 20,000rpm with no oil pressure in the journal bearings "whilst at the same time waiting for oil pressure to build" because that would be stupid. Yep you've been at this for some time and oh look, the "so called rub which was not the reason for the fail" is proven to be the cause of the fail. There were no other starts, only spoolups with air. When the main fuel pump had 20v on it, the fuel pressure was about 2.5bar, a little less than what you have setup as the nominal pressure at full throttle. With insufficient air, it simply refused to ignite. It is extremely obvious by the video that the engine starts nicely, idles at constant speed for about 10 seconds. Only once it has warmed and started rubbing, the speed starts dropping. Likewise, the rub has gone back to being minimal with the engine at room temperature overnight. You can pretend "the engine was run outside its parameters" all you like, it doesn't surprise me, nor do I care. If the rub had magically been done by the previous failed start attempts, it simply wouldn't have spooled up on air because the friction torque would have been too high. Its common sense. The cracks on the welds of the studs is due to design error, not "repeat hot starts". When you weld in front of a thread, the weld takes all the load and the thread takes none. These tack welds will have failed during the 3bar running, the engine hasn't been stripped since. In any case the anti rotation isn't required so its not a concern. |
|
|
|
Post by andym on Apr 26, 2021 12:01:38 GMT -5
Scott.... say what you want... you clearly think you know better... just like the afterburner... you saying it does not work... you reason for making changes... when it had not been tested... yet my others have work... all for your own hubris
So it's better to spool a shift with restive load on it from 5 bar oil pressure.. not explain why it should of been started the way I explained... but clearly you know better... proving my point, that you feel you can explain away the reason you choose to ignore the reasons for the way the engine should be started.
I have pictures of the welds after 3 bar run... and engine strip down to check all is good... as I do after a test pushing the engine... and guess what no crack... Happy to share...
In the video you can clearly hear the rota spool down.
Since you got the engine you seem to have found fault with every things... the afterburner.. oil system... even the length of the kart...all I know engine worked fine in my last video...
Then share the video...
|
|
|
|
Post by slittlewing on Apr 26, 2021 14:11:06 GMT -5
So I have stripped the engine and itâs mostly good news đ Prior to stripping it, I tried again to measure the turbine to NGV wall clearance but this time using two feeler gauges on blades approximately opposite each other, and away from the âscabsâ. The maximum I could get was 0.30mm on one side and 0.25mm the other, an average of 0.275mm clearance. Having removed the turbine wheel there is a reasonable amount of scoring and material removed so I guess this 0.275 would have been less before the material transfer. The turbine wheel looks recoverable, I canât see any cracks by eye and the blade tips are not burnt away. The comp is good and hasnât touched anything. The journal bearings look ok but I will replace anyway. The thrust bearing is damaged and overheated, expected due to reacting the friction loads. Presumably the same energy that has gone into melting the turb blades has been absorbed by the thrust bearing. Likewise the steel disc that sits against the thrust bearing is a bit scored/overheated and needs replacing. The fuel injection needles are not blocked when tested with a syringe which is good. I tried to measure the height of the scabs and they are 0.25mm tall. This correlates with the clearance feeler gauge measurements and explains why there is barely any rub with the engine cold (0.025mm clear), unless you tilt the rotor. So it appears the issue is purely one of clearance and thermal expansion. In terms of the flat rear casing wall, I feel it would be an improvement to add a concave rear dome. The NGV wall thickness is 5mm on both inner and outer compared to Anders engine which I think are 10mm each. The question is how much skim to take off, ensuring some safety margin but not too much that weakens the NGV? Pictures and video below:                     Andy, Iâm sorry you feel that way but you have got the wrong end of the stick. I have previously described your engine as a work of art and complimented it thoroughly. There are numerous exquisite details on the engine, and I can both see and respect the hours and care that have gone into it. I wanted to change the afterburner fuel ring and some other details but thatâs just personal preference and opinion, not factual fault and certainly not intended as an insult. There will always be some customising involved when buying any project. đ Cheers Scott |
|
|
|
Post by racket on Apr 26, 2021 17:23:12 GMT -5
Hi Scott
Thanks for the extra pics , very informative , we learn so much more from them :-)
Is the quadrant at the "bottom" ( 6 0'clock) region ??
Could you please post a pick of the rear end of the flametube please , the rest of the FT is looking OK .
Yep , grind the wheel rather than the NGV for extra clearance to keep as much meat in the NGV as possible.
The thrust bearing wear is understandable , normally the thrust is always forward , the rearward thrust you experienced was forced onto the "bump bearing" that isn't really designed for it , its only there to provide the endfloat stop.
I'll keep checking the pics out for more clues :-)
Cheers
John
|
|
|
|
Post by madpatty on Apr 26, 2021 18:41:01 GMT -5
Hi Scott Thanks for the extra pics , very informative , we learn so much more from them :-) Is the quadrant at the "bottom" ( 6 0'clock) region ?? Could you please post a pick of the rear end of the flametube please , the rest of the FT is looking OK . Yep , grind the wheel rather than the NGV for extra clearance to keep as much meat in the NGV as possible. The thrust bearing wear is understandable , normally the thrust is always forward , the rearward thrust you experienced was forced onto the "bump bearing" that isn't really designed for it , its only there to provide the endfloat stop. I'll keep checking the pics out for more clues :-) Cheers John Hi Racket. Reading through the comment I understand everything except the part, âthe rearwards thrust you experiencedâ and the âbump bearing â you are referring to. Which is that component you are referring to? Also how did the rearwards thrust come into play? From the pictures isnât it unclear what contacted what first? is it the turbine wheel that contacted the NGV plate first or the thrust collar contacting the brass thrust bearing first? If the thrust bearing had a contact first then it can be an insufficient oil pressure issue. AND if turbine have had a contact first then the front face of the brass thrust bearing shouldâve had more rubbing marks. Just my 2 cents. Regards Patty |
|
|
|
Post by racket on Apr 26, 2021 19:33:39 GMT -5
Hi Patty
As the scabs built up , it would require the shaft to move rearwards putting load onto the bearing (bump bearing)which normally doesn't see load as our turbo shafts normally want to go forward , turbos occassionally do have rearward thrust if the turb stage is undersized and the exhaust manifold is running much higher static pressures than the boost going to the engine as often happens during hard acceleration , its a cause of thrust bearing failure .
As theres only several thou of endfloat in the shaft one would imagine the thrust bearing to have been destroyed before the turb wheel contacted the NGV which had several times more clearance when assembled .
Its more likely the wheel contacted the NGV first , the fact that the scabs are located over only a section of the NGV is "interesting" and would indicate to me that there may have been some warpage , possibly from a varying temperature profile around the NGV , hence a very good check of the flametube and fueling ring required to determine if there was any hot/cold regions, but again , the spooldown may have produced abberant conditions which would affect the parts condition ...............R & D ain't easy :-( ...................multi million dollar engines have development issues .
Its rather irrelevant what has actually caused the problem , and probably impossible to determine what happened, when , because as the rotor spooled down for whatever reason , things weren't "right" .
Parts do "relax" and/or change shapes under varying temperature gradients, thrust bearings do "run in", they are only turbo parts made at the lowest cost not aerospace specs , without extensive data collection at the time of the incident all we really can do is fix the parts , reassemble and check everything as we go to make sure nothing has warped due to any abberant temperatures.
Cheers
John
|
|
|
|
Post by slittlewing on Apr 27, 2021 4:24:46 GMT -5
Thanks again John! The rubbing was around the 12 o clock region (at least between 11 o clock and 1 o clock) I have marked the orientation on the pic below. The rear flame tube 5mm wall that leads to the NGV doesnât show much other than being sooty but there are heat marks on the back of the NGV wall.     There is a very masterful manual machinist at work who is going to give it a go taking 0.25mm off the back face of the turbine, from the high spot. Equally I plan to take 0.25mm off the NGV back wall to âsquare it upâ and remove the grooves. Maximum (cold) clearance would then theoretically be ~0.75mm but that will reduce when the comp pulls the shaft forward under load, and will also reduce when the engine warms as proven! I found a hx80/hc5a mellett rebuild kit that I will buy, I think it will fit the engine, not sure about the thrust collar and oil slinger being the same but I have no turbo number to go on. Cheers Scott |
|
|
|
Post by finiteparts on Apr 27, 2021 8:17:08 GMT -5
It feels like the thrust brg damage, the deeply scalloped turbine with the large compressor and the turbine aft rub all point to a rotor thrust problem.
That thrust brg was really worked hard.
Chris
|
|
|
|
Post by racket on Apr 27, 2021 16:31:19 GMT -5
Hi Scott
Thanks for the extra pics :-)
Interesting that it was at the "top" of the engine, and that the front of the wall is coloured whilst the rear isn't despite the frictional issues, coloured patches indicate heating and potential warpage .
Could you please explain how your fuel system was working during the incident , is there "variability" or was it simply a case of idle fuel set, and the valve opened as the engine spooled up and was left at that setting until you switched it off ??
Cheers
John
|
|
|
|
Post by slittlewing on Apr 27, 2021 16:35:24 GMT -5
No worries, it sort of feels like the coloured patches on the front have come through from the friction at the rear.
The fuel system is as built by Andy so a fixed voltage on the pump, and a mechanical bypass valve linked to the throttle pedal and retuning flow to the tank. This valve is a special reprofiled needle type valve that operates on a quarter of a turn.
The throttle pedal and pump voltage were untouched from the moment I flicked them on during the air spool, and I watched the fuel gauge with confusion as it sat stable whilst engine speed decreased.
Cheers
Scott
|
|
|
|
Post by racket on Apr 28, 2021 17:26:47 GMT -5
Hi Scott
OK , so with fuel switched on during spoolup and left untouched until you switched it off once the engine was at lowish RPM , the external flames from the jetnozzle should be exactly the same at the same RPM in either increasing or decreasing RPM , but they aren't :-(
At ~1min 55sec on the video you have a nice "afterburner" type plume from the nozzle , but at no point on the spooldown between 2min 40 to 2min 50 secs did that occur despite the RPM going through the same range as spoolup.
I'll have to stick with my original diagnosis , you "ran out of fuel" , how or why I don't know , same with the fuel pressure observations remaining stable as the engine spooled down , the fuel pressure should have decreased as the P2 "backpressure" decreased.
The scabs weren't the cause , but may have been caused by abberant heating of the NGV front wall causing buckling , the rear scabby side of the wall isn't coloured to the same extent as the front side , hence my thoughts that there were "flames" hitting the front of the NGV wall.
Before doing another spoolup attempt the complete fueling system will need a comprehensive range of testing to determine whats going on , the patchy colouration on the NGV sorta indicates "hotstreaks??" , perhaps "rubbish" has made its way to the injectors , lotsa unknowns :-(
Hopefully this will be of some help
Cheers
John
|
|
