Author Topic: Merlin and RD-180 engines  (Read 5852 times)

Offline Peter B

  • Saturn
  • ****
  • Posts: 1268
Merlin and RD-180 engines
« on: April 28, 2022, 10:14:51 AM »
I got into a little discussion on YT about the Axiom 1 launch.

One commenter said that Merlin engines were just re-branded NK-33 engines. I asked what the similarities were, and his blustery response boiled down to "high efficiency", although he also repeated the claim the design was copied. When I called him on this, in his next reply he switched over to comparing the Merlin to the RD-180.

He made five claims, which I think I know the answers to, but I thought I'd run them past the experts here for correction where necessary:

1. In 2019 Musk said the RD-180 was the finest rocket engine ever made. Sort of true - he said "excellent engineering"
2. Both engines use RP-1 and LOX. True but irrelevant - many other engines do too
3. Both engines have a duel nozzle design. False - the Merlin is a single nozzle
4. The RD-180 has a dual combustion chamber design. True but irrelevant - the Merlin doesn't
5. Reusable first stages are "showmanship without benefits", according to Russians. False - the reusability of the Falcon 9 first stage and the payload shrouds has reduced launch costs considerably, which is a benefit. (I couldn't find any reference to the "showmanship without benefits" quote but didn't look far.)

I have a suspicion he's now comparing the RD-180 to the Raptor engine, but I'm not sure, and I'm also unsure how similar they are if he is (of course, if he is, it's irrelevant to the Falcon 9...).

Any help would be appreciated. Thank you!

Offline molesworth

  • Mars
  • ***
  • Posts: 349
  • the curse of st custards
Re: Merlin and RD-180 engines
« Reply #1 on: April 28, 2022, 06:27:13 PM »
I was about to go into how the Raptor is nothing like the RD-180, so there's not really any basis for comparison, when by coincidence a new video from Tim Dodd (Everyday Astronaut) popped up on my YouTube notifications.  I've not finished watching yet, but it looks like it's a good comparison of the different engine types, flows, fuels etc.  Hopefully it'll answer all/most of your correspondent's questions or claims.



Days spent at sea are not deducted from one's allotted span - Phoenician proverb

Offline smartcooky

  • Uranus
  • ****
  • Posts: 1959
Re: Merlin and RD-180 engines
« Reply #2 on: April 28, 2022, 08:18:24 PM »
I got into a little discussion on YT about the Axiom 1 launch.

One commenter said that Merlin engines were just re-branded NK-33 engines. I asked what the similarities were, and his blustery response boiled down to "high efficiency", although he also repeated the claim the design was copied. When I called him on this, in his next reply he switched over to comparing the Merlin to the RD-180.

He made five claims, which I think I know the answers to, but I thought I'd run them past the experts here for correction where necessary:

1. In 2019 Musk said the RD-180 was the finest rocket engine ever made. Sort of true - he said "excellent engineering"
2. Both engines use RP-1 and LOX. True but irrelevant - many other engines do too
3. Both engines have a duel nozzle design. False - the Merlin is a single nozzle
4. The RD-180 has a dual combustion chamber design. True but irrelevant - the Merlin doesn't
5. Reusable first stages are "showmanship without benefits", according to Russians. False - the reusability of the Falcon 9 first stage and the payload shrouds has reduced launch costs considerably, which is a benefit. (I couldn't find any reference to the "showmanship without benefits" quote but didn't look far.)

I have a suspicion he's now comparing the RD-180 to the Raptor engine, but I'm not sure, and I'm also unsure how similar they are if he is (of course, if he is, it's irrelevant to the Falcon 9...).

Any help would be appreciated. Thank you!

RD-180
Fuel: Kerolox
Config: Dual combustion chamber
Cycle type: Closed Cycle (Staged combustion)
Length: 3.56 m
Diameter: 3.15 m
Dry weight: 5,480 kg

Merlin
Fuel: Kerolox
Config:Single combustion chamber
Cycle type: Open cycle (Gas generator)
Length: 2.92 m
Diameter: Sea level 0.92, Vacuum 3.3 m
Dry weight   630 kg

Raptor
Fuel: Methalox
Config: Single combustion chamber
Cycle: Closed cycle (Full flow staged combustion)
Length: 3.1 m
Diameter: 1.3 m
Dry weight: 1,500 kg

Identical in every way ::)

Your Russian troll correspondent is an idiot
If you're not a scientist but you think you've destroyed the foundation of a vast scientific edifice with 10 minutes of Googling, you might want to consider the possibility that you're wrong.

Offline Peter B

  • Saturn
  • ****
  • Posts: 1268
Re: Merlin and RD-180 engines
« Reply #3 on: April 29, 2022, 04:42:30 AM »
...Identical in every way ::)

Your Russian troll correspondent is an idiot

LOL! Thought so.

Offline smartcooky

  • Uranus
  • ****
  • Posts: 1959
Re: Merlin and RD-180 engines
« Reply #4 on: April 29, 2022, 09:16:08 AM »
...Identical in every way ::)

Your Russian troll correspondent is an idiot

LOL! Thought so.


Yup, I also take issue with his characterization that "Reusable first stages are "showmanship without benefits". This is just demonstrably rubbish. As you correctly point out, the reuse of the first stage and the payload fairings has dramatically reduced launch costs. While SpaceX have not directly said what the difference is between building a first stage and  reusing one, there are some things we do know, and Musk has made some hints.

1. The full fuel load is about $200K for both stages (about $550K for Falcon Heavy) so it is not really a significant part of the cost.

2. SpaceX charges $62M for a launch with a new booster, and $50M with a reused one - that is 20% cheaper, a significant cost reduction for customers.

3. In the past, Musk has dropped a few hints about his costs. Firstly, he has indicated that manufacturing the booster is 60% of the total cost of a new rocket, while the second stage (including payload fairings) is about 20%, and he has stated that a second stage costs around $10M to make, so we can easily work out that a booster should cost around $30 million to manufacture. Secondly, Musk has said the cost of refurbishing the recovered booster is only $250,000. It appears this does not require much manpower or much in the way of hardware replacements or complex inspections. I am unsure if those costs include recovery costs such as operating the three ASDS landing platforms.

4. One drawback is that reusability reduces the maximum payload because it needs to reserve fuel for boostback (sometimes), entry and landing. There is also the additional mass of the landing legs, grid fins etc. However, Musk reckons that only lowers the maximum payload capacity by less than 40% (so far only 26 of their 94 Block 5 launches have carried a max payload). After two launches, the total cost and combined payload capacity are roughly comparable to one launch of an expendable rocket, so if the booster flies at least three times they're on a win.

5. Finally, after originally insisting that all their missions were to be on new boosters, NASA are now opting for refurbished boosters, not just for the cost, but because of their "flight proven" status....

CRS-21 was launched on B1058.4 (the ".4" means it was its 4th flight)
CRS-23 was on B1061.4
DAT was on B1063.3
IXPE was on B1061.5 (the same booster that launched CRS-23)

This has even extended to NASA crewed flights...
Crew-2 was on B1061.2 (the same booster that launched Crew-1, and which went on to launch CRS-23 and IXPE)
Crew-3 was on B1067.2
Crew-4 was on the same booster as Crew-3 (now B1067.4 because in between those, it launched Turksat 5B)

And further extended to  the initially skeptical USSF...
GPS111-05 was on B1062.2
NROL-85 was on B1071.2

I thing the most telling stat though is this one.... They have 13 Block 5 boosters in their current fleet, two have flown 12 times, one flown 11 times and one flown 10 times, so those four boosters have launched 45 of their 94 Block 5 launches, almost half of them!! 
If you're not a scientist but you think you've destroyed the foundation of a vast scientific edifice with 10 minutes of Googling, you might want to consider the possibility that you're wrong.

Offline jfb

  • Mars
  • ***
  • Posts: 396
Re: Merlin and RD-180 engines
« Reply #5 on: April 30, 2022, 08:47:08 PM »
Another way reusability cuts down on launch costs - it dramatically cuts turnaround time, allowing them to fly missions every couple of weeks (including two manned missions, one launching the day after the previous on splashed down).  That cadence gives them plenty of opportunities to streamline operations on the pad, further reducing costs.

Offline smartcooky

  • Uranus
  • ****
  • Posts: 1959
Re: Merlin and RD-180 engines
« Reply #6 on: May 01, 2022, 03:20:13 AM »
Another way reusability cuts down on launch costs - it dramatically cuts turnaround time, allowing them to fly missions every couple of weeks (including two manned missions, one launching the day after the previous on splashed down).  That cadence gives them plenty of opportunities to streamline operations on the pad, further reducing costs.

Correct.

Starlink 4-16 launched just over a day ago on April 30. It was atop booster 1062 - the same booster used for the Axiom 1 Crew Dragon on 9 April. That's 21 days, including the 4½ days it took to tow it back to port. SpaceX have stated that the actual process time of refurbishing B1062 was just 9 days.
If you're not a scientist but you think you've destroyed the foundation of a vast scientific edifice with 10 minutes of Googling, you might want to consider the possibility that you're wrong.

Offline smartcooky

  • Uranus
  • ****
  • Posts: 1959
Re: Merlin and RD-180 engines
« Reply #7 on: May 01, 2022, 04:57:27 AM »
You can't overestimate how important the transition from Block 4 to Block 5 has been, especially in the area of booster maintenance and refurbishment. Here's just four things as an example.

Landing Legs: These suffer a lot of aerodynamic heating during re-entry, and they are also closest to the engines during landing.
- On Block 4, they were coated with a white, ablative, heat-resistant cork which, if it wasn't peeled off during reentry, had to be stripped off and replaced before the next flight. The legs themselves had to be unbolted and removed for processing after each landing.
- On Block 5, anyone who has been paying close attention will have noticed that the landing legs are now black. This is not for aesthetics - the black coating is a type of heat-resistant semi-crystalline organic polymer resin called polyvinyl cyanide. Its lighter, cheaper and more heat resistant than the previously used ablative cork, and most importantly, it does not need to be replaced after every  flight. Also, the landing legs can now be retracted in-situ, at the port making transport to the maintenance facility easier and quicker. Similarly, any maintenance to the legs can be carried out with them still attached to the booster

Grid Fins: These also suffer a lot of aerodynamic heating during re-entry.
- On Block 4, they were aluminium, and since they stick way out into the air-stream, they were coated with with the same white, ablative, heat-resistant cork as the landing legs. It too had to be stripped off and replaced before the next flight. Some aluminum grid fins caught fire during the entry, especially on high mass payload missions which had longer downrange landing distances and therefore higher re-entry velocities.
- On Block 5, they are made of titanium, which has a melting point of 1650-70 °C (compared with 660 for aluminium). The titanium grid fins are uncoated and have much higher survivablity. They require no more than inspection during refurbishment, only being replaced if they show cracking or damage.

The Octaweb: Suffers more heat exposure than any other part of the booster (except the engine bells)
- On Block 4, the entire octaweb was welded. This made it hard for the maintenance crews to replace it since welding literally made it part of the entire booster shell. It also needed to be cut off and refitted by welding if an engine had to be changed. Also, it is more difficult to maintain due to the brittleness of the joints - every joint has to be crack-tested
- On Block 5, the octaweb is bolted. No welding cracks to check for, but I imagine the bolt holes would still be checked. If an engine change is needed, it can be easily be unbolted, removed and refitted. Bolted octawebs also make it easier for Falcon Heavy. They are strengthened to withstand the additional stress the thrust. Previously, the centre core would require a one-off manufactured octaweb, and modifications to those on the side boosters. Now, they just have a set of three strengthened octawebs, one for the centre core, and two for the side boosters, and just swap them over.

Thermal protection for the engines
- On Block 4, this was achieved by coating the bottom of the octaweb with a composite thermal protection system using a  mixture of the same ablative cork used for the legs and grid fins, and "Phenolic-Impregnated Carbon Ablator" a.k.a. PICA-X; the same material that formed the heat shield of the Cargo Dragon capsule. Unfortunately, the cork gives rise to similar problems faced with the legs and grid-fins. It peels off on exposure to excessive heat.
- On Block 5, they use Inconel, a far more heat resistant material, that again, does not need to be coated. Its heavy (3½x that of aluminium), but its heat resistance outweighs that disadvantage when it is such a comparatively small part of the booster shell.
« Last Edit: May 01, 2022, 05:01:26 AM by smartcooky »
If you're not a scientist but you think you've destroyed the foundation of a vast scientific edifice with 10 minutes of Googling, you might want to consider the possibility that you're wrong.

Offline bknight

  • Neptune
  • ****
  • Posts: 3107
Re: Merlin and RD-180 engines
« Reply #8 on: May 01, 2022, 11:26:29 AM »
...

The Octaweb: Suffers more heat exposure than any other part of the booster (except the engine bells)
- On Block 4, the entire octaweb was welded. This made it hard for the maintenance crews to replace it since welding literally made it part of the entire booster shell. It also needed to be cut off and refitted by welding if an engine had to be changed. Also, it is more difficult to maintain due to the brittleness of the joints - every joint has to be crack-tested
- On Block 5, the octaweb is bolted. No welding cracks to check for, but I imagine the bolt holes would still be checked. If an engine change is needed, it can be easily be unbolted, removed and refitted. Bolted octawebs also make it easier for Falcon Heavy. They are strengthened to withstand the additional stress the thrust. Previously, the centre core would require a one-off manufactured octaweb, and modifications to those on the side boosters. Now, they just have a set of three strengthened octawebs, one for the centre core, and two for the side boosters, and just swap them over.

...
Interesting name they chose, I would have thought that nanoweb would have been the choice.
Truth needs no defense.  Nobody can take those footsteps I made on the surface of the moon away from me.
Eugene Cernan

Offline smartcooky

  • Uranus
  • ****
  • Posts: 1959
Re: Merlin and RD-180 engines
« Reply #9 on: May 01, 2022, 04:51:20 PM »
...

The Octaweb: Suffers more heat exposure than any other part of the booster (except the engine bells)
- On Block 4, the entire octaweb was welded. This made it hard for the maintenance crews to replace it since welding literally made it part of the entire booster shell. It also needed to be cut off and refitted by welding if an engine had to be changed. Also, it is more difficult to maintain due to the brittleness of the joints - every joint has to be crack-tested
- On Block 5, the octaweb is bolted. No welding cracks to check for, but I imagine the bolt holes would still be checked. If an engine change is needed, it can be easily be unbolted, removed and refitted. Bolted octawebs also make it easier for Falcon Heavy. They are strengthened to withstand the additional stress the thrust. Previously, the centre core would require a one-off manufactured octaweb, and modifications to those on the side boosters. Now, they just have a set of three strengthened octawebs, one for the centre core, and two for the side boosters, and just swap them over.

...
Interesting name they chose, I would have thought that nanoweb would have been the choice.

Yeah, I think its because its an eight-sided web
If you're not a scientist but you think you've destroyed the foundation of a vast scientific edifice with 10 minutes of Googling, you might want to consider the possibility that you're wrong.

Offline cjameshuff

  • Mars
  • ***
  • Posts: 372
Re: Merlin and RD-180 engines
« Reply #10 on: May 02, 2022, 10:41:05 AM »
...

The Octaweb: Suffers more heat exposure than any other part of the booster (except the engine bells)
- On Block 4, the entire octaweb was welded. This made it hard for the maintenance crews to replace it since welding literally made it part of the entire booster shell. It also needed to be cut off and refitted by welding if an engine had to be changed. Also, it is more difficult to maintain due to the brittleness of the joints - every joint has to be crack-tested
- On Block 5, the octaweb is bolted. No welding cracks to check for, but I imagine the bolt holes would still be checked. If an engine change is needed, it can be easily be unbolted, removed and refitted. Bolted octawebs also make it easier for Falcon Heavy. They are strengthened to withstand the additional stress the thrust. Previously, the centre core would require a one-off manufactured octaweb, and modifications to those on the side boosters. Now, they just have a set of three strengthened octawebs, one for the centre core, and two for the side boosters, and just swap them over.

...
Interesting name they chose, I would have thought that nanoweb would have been the choice.

...why? What about it is "nano"?

The Merlin engines also use a somewhat unusual pintle injector. The Merlin engines have a quite clear lineage, with many design features in common with the FASTRAC engine (which itself never got developed into a flight-ready engine) and extensive development between the Merlin 1A and the current Merlin 1D. The Merlin 1A and 1B weren't even regeneratively cooled, they used ablative chambers and nozzles. The notion that they just copied their engines from someone else is complete nonsense.

Offline Peter B

  • Saturn
  • ****
  • Posts: 1268
Re: Merlin and RD-180 engines
« Reply #11 on: May 03, 2022, 05:06:35 AM »
Thanks for those analyses of the engines, excellent summaries!

Offline bknight

  • Neptune
  • ****
  • Posts: 3107
Re: Merlin and RD-180 engines
« Reply #12 on: May 06, 2022, 01:21:17 PM »
...

The Octaweb: Suffers more heat exposure than any other part of the booster (except the engine bells)
- On Block 4, the entire octaweb was welded. This made it hard for the maintenance crews to replace it since welding literally made it part of the entire booster shell. It also needed to be cut off and refitted by welding if an engine had to be changed. Also, it is more difficult to maintain due to the brittleness of the joints - every joint has to be crack-tested
- On Block 5, the octaweb is bolted. No welding cracks to check for, but I imagine the bolt holes would still be checked. If an engine change is needed, it can be easily be unbolted, removed and refitted. Bolted octawebs also make it easier for Falcon Heavy. They are strengthened to withstand the additional stress the thrust. Previously, the centre core would require a one-off manufactured octaweb, and modifications to those on the side boosters. Now, they just have a set of three strengthened octawebs, one for the centre core, and two for the side boosters, and just swap them over.

...
Interesting name they chose, I would have thought that nanoweb would have been the choice.

...why? What about it is "nano"?

The Merlin engines also use a somewhat unusual pintle injector. The Merlin engines have a quite clear lineage, with many design features in common with the FASTRAC engine (which itself never got developed into a flight-ready engine) and extensive development between the Merlin 1A and the current Merlin 1D. The Merlin 1A and 1B weren't even regeneratively cooled, they used ablative chambers and nozzles. The notion that they just copied their engines from someone else is complete nonsense.

I was thinking that the first stage has 9 engines, thus the nano.  If that is incorrect then whatever the next increment from octo then.
Truth needs no defense.  Nobody can take those footsteps I made on the surface of the moon away from me.
Eugene Cernan

Offline cjameshuff

  • Mars
  • ***
  • Posts: 372
Re: Merlin and RD-180 engines
« Reply #13 on: May 06, 2022, 03:38:35 PM »
...

The Octaweb: Suffers more heat exposure than any other part of the booster (except the engine bells)
- On Block 4, the entire octaweb was welded. This made it hard for the maintenance crews to replace it since welding literally made it part of the entire booster shell. It also needed to be cut off and refitted by welding if an engine had to be changed. Also, it is more difficult to maintain due to the brittleness of the joints - every joint has to be crack-tested
- On Block 5, the octaweb is bolted. No welding cracks to check for, but I imagine the bolt holes would still be checked. If an engine change is needed, it can be easily be unbolted, removed and refitted. Bolted octawebs also make it easier for Falcon Heavy. They are strengthened to withstand the additional stress the thrust. Previously, the centre core would require a one-off manufactured octaweb, and modifications to those on the side boosters. Now, they just have a set of three strengthened octawebs, one for the centre core, and two for the side boosters, and just swap them over.

...
Interesting name they chose, I would have thought that nanoweb would have been the choice.

...why? What about it is "nano"?

The Merlin engines also use a somewhat unusual pintle injector. The Merlin engines have a quite clear lineage, with many design features in common with the FASTRAC engine (which itself never got developed into a flight-ready engine) and extensive development between the Merlin 1A and the current Merlin 1D. The Merlin 1A and 1B weren't even regeneratively cooled, they used ablative chambers and nozzles. The notion that they just copied their engines from someone else is complete nonsense.

I was thinking that the first stage has 9 engines, thus the nano.  If that is incorrect then whatever the next increment from octo then.

That would be a "nonaweb". The "nano" prefix means "small", usually on the order of molecular structures, or "one billionth" when used as a SI prefix. A nanoweb would be something you image with an electron microscope.

Offline bknight

  • Neptune
  • ****
  • Posts: 3107
Re: Merlin and RD-180 engines
« Reply #14 on: May 07, 2022, 08:49:22 AM »
...

The Octaweb: Suffers more heat exposure than any other part of the booster (except the engine bells)
- On Block 4, the entire octaweb was welded. This made it hard for the maintenance crews to replace it since welding literally made it part of the entire booster shell. It also needed to be cut off and refitted by welding if an engine had to be changed. Also, it is more difficult to maintain due to the brittleness of the joints - every joint has to be crack-tested
- On Block 5, the octaweb is bolted. No welding cracks to check for, but I imagine the bolt holes would still be checked. If an engine change is needed, it can be easily be unbolted, removed and refitted. Bolted octawebs also make it easier for Falcon Heavy. They are strengthened to withstand the additional stress the thrust. Previously, the centre core would require a one-off manufactured octaweb, and modifications to those on the side boosters. Now, they just have a set of three strengthened octawebs, one for the centre core, and two for the side boosters, and just swap them over.

...
Interesting name they chose, I would have thought that nanoweb would have been the choice.

...why? What about it is "nano"?

The Merlin engines also use a somewhat unusual pintle injector. The Merlin engines have a quite clear lineage, with many design features in common with the FASTRAC engine (which itself never got developed into a flight-ready engine) and extensive development between the Merlin 1A and the current Merlin 1D. The Merlin 1A and 1B weren't even regeneratively cooled, they used ablative chambers and nozzles. The notion that they just copied their engines from someone else is complete nonsense.

I was thinking that the first stage has 9 engines, thus the nano.  If that is incorrect then whatever the next increment from octo then.

That would be a "nonaweb". The "nano" prefix means "small", usually on the order of molecular structures, or "one billionth" when used as a SI prefix. A nanoweb would be something you image with an electron microscope.
That is precisely why I have in my post
Quote
If that is incorrect then whatever the next increment from octo then.
Truth needs no defense.  Nobody can take those footsteps I made on the surface of the moon away from me.
Eugene Cernan