ULA Innovation: Integrated Vehicle Fluids (IVF)

[Damon Hill]

One detail I don't recall being addressed is the crankshaft torque of the engine; I wonder how that will be handled?

[kch]

One detail I don't recall being addressed is the crankshaft torque of the engine; I wonder how that will be handled?

Maybe the same way it's handled on longitudinal-crankshaft motorcycles -- the "peripheral equipment" (alternator, coolant pump, etc.) rotates in the opposite direction to cancel the torque reaction.  No reason why it wouldn't work for this.

[tj]

Countering torque -- The 2012 ULA white paper illustration might seem to indicate the 6 cylinder engine rotation is counter rotating 3-piston halves. I seem to observe a connection to the generator exiting midway between the 6 piston block.

[Damon Hill]

kch: Good point.  Seems like the PTO from the crankshaft could be easily geared to accomplish that counter-torque task.  Simple solution!

[Patchouli]

I wonder could the engine from IVF be used on other in space applications such as a high performance Lunar or Mars rover?

[Ronsmytheiii]

So with IVF an upperstage is basically using Hydrogen boil-off as a source of both power and fuel, that essentially gives you unlimited firings of both the main engine and the thrusters right?

If that is the case, that would make interesting delivery options for GEO craft.  Imagine being able to trade efficiency versus time for a satellite. A normal launch would use two burns, but a customer could also settle for a shorter but more efficient delivery method using many thruster and/or main engine firings. Kind of like SEP on the cheap.  That seems promising for commercial clients.

[jg]

I wonder could the engine from IVF be used on other in space applications such as a high performance Lunar or Mars rover?

The engine might be, but it's complex; but a lot of the magic of the IVF work is that the energy used is leveraged N ways to Sunday, for thrusters, for pressurization of the tanks, for thrust to keep the propellants settled (ullage), etc. I suspect you'd be best off thinking things from first principle.  Dunno if Frank Zegler has spent any time thinking about that possibility.

[Jim]

I wonder could the engine from IVF be used on other in space applications such as a high performance Lunar or Mars rover?

Not really.  It is not a high performance engine.  It is specifically for the IVF and not to drive wheels.

[Jim]

So with IVF an upperstage is basically using Hydrogen boil-off as a source of both power and fuel, that essentially gives you unlimited firings of both the main engine and the thrusters right?

If that is the case, that would make interesting delivery options for GEO craft.  Imagine being able to trade efficiency versus time for a satellite. A normal launch would use two burns, but a customer could also settle for a shorter but more efficient delivery method using many thruster and/or main engine firings. Kind of like SEP on the cheap.  That seems promising for commercial clients.

Not really.  The IVF thrusters are not as efficient as the RL-10.

[Patchouli]

Not really.  It is not a high performance engine.  It is specifically for the IVF and not to drive wheels.

I was thinking in a hybrid drive setup where the engine charges the rover's batteries for operations during the lunar night or for higher then average power then would be possible with solar panels.

[sanman]

Has any other way ever been proposed to harness hydrogen boil-off? Like maybe a Proton Exchange Membrane, or Thermoelectric Material or something?

[TrevorMonty]

I wonder could the engine from IVF be used on other in space applications such as a high performance Lunar or Mars rover?

Masten Xeus lander which is based on modified Centuar/ACES. The ICE would enable the lander and crew to survive a lunar night, not only does it provide power but also heating and water. There is a thread which talks about using them in rovers, look under Missions to Moon section, feel free to kick the thread back into life.

[Jim]

I was thinking in a hybrid drive setup where the engine charges the rover's batteries for operations during the lunar night or for higher then average power then would be possible with solar panels.

A fuel cell is better for that.  Higher efficiency and less conversion losses.  But either way, not a good idea.  It would be wasting solar power carrying around the the mass of the ICE or fuel and not to mention fuel when not in use.

[Jim]

Has any other way ever been proposed to harness hydrogen boil-off? Like maybe a Proton Exchange Membrane, or Thermoelectric Material or something?

read the documents

[gin455res]

okay, this might be obvious but not for me so:

why use a piston engine (+generator i'd assume?) instead of an fuel-cell?
i'd see no problem designing a system of multiple cells, churning out amps from excess Hydrogen/Oxygen for a much lower weight and no moving parts
(much less failure modes, there's a reason gemini/apollo didn't use piston engines for power)

I could see why a fuel cell was a poor choice, but wasn't seeing the problem with a gas turbine, so I read the docs.
http://tinyurl.com/ula-ivf2012

....

.....

As for turbines:

A turbine could be used for such an application but it would be exquisitely small with extremely high rotating speeds to produce only 20kW with low density hydrogen as the working fluid. Provisions for heat and shaft power extraction could be made but the overall developmental complexity of cooling, lubrication, ignition, control and power take off at this very low power level seemed daunting compared to the IC engine. The use of such small turbines on ground based installations is virtually unheard of. Virtually a whole new technology would have to be developed at substantial cost and risk.

So the turbine would be smaller. But too small, too high RPM, hard to get the heat out, and not off the shelf technology.
But note that if IVF were used on a methane launcher, 20kW natural gas micro-turbines are commercially available, so that may be an option. (But of course those turbines burn 80% nitrogen, not pure O2)
For hydrogen, the ICE came out ahead.

Yes the docs are a very good read.

There is a type of gas-turbine that is described here:
http://www.agileturbine.com/publications/Small%20Scale%20Combined%20Heat%20and%20Power.pdf
that uses a sub-atmospheric cycle -- combustion, expansion, cooling and then compression.

In the link above it is suggested as the engine in residential scale Combined Heat and Power unit. In this application it is claimed to have some advantages; a) the combustion is at slightly subatmospheric pressure and this eliminates the need for a natural gas fuel pump; and 2) although having lower power density than a conventional (compressor, combustion, expansion) turbine, it scales to small powers better.

In the outlined (IVF) flat head piston engine the enthalpy comes indirectly through the cooling system. Perhaps a sub-atmospheric turbine would supply the enthalpy (heat output)  more directly (/simply) , much as in the above CHP application.

(makes one wonder if a sub-atmospheric staged combustion rocket engine is feasible and possibly simpler - no feed pump, lower pressures, and better scalability - though using the propellant as a heat sink before burning it upstream sends logical circularity warning Klaxons in my head - though I think it works?)

-----------------------------------------------------------
Hi, This question is for Frank Zegler (see below)
------------------------------------------------------------

Dr Sowers,

When eliminating turbine based units from the IVF system design, were sub-atmospheric 'inverted brayton cycles' such as outlined in the linked pdf, considered.  This is a scheme for a residential scale CHP (high enthalpy) micro-turbine system, that reverses compressor and turbine sequence to  produce very-low-power turbo-generators.

Or would continuous combustion require too high a fuel flow to keep combustion temperatures sensible, (what is the peak combustion temperature in the IC engine anyway)?

http://www.agileturbine.com/publications/Small%20Scale%20Combined%20Heat%20and%20Power.pdf 

thanks
Toby

Post this question on the IVF thread.  I'll get Frank Zegler, the inventor, to answer.  (Another way of saying: I haven't the foggiest...)

[Designvis]

Frank Zegler of ULA here to answer a few questions about the IVF architecture and how we got to where we are. 

We considered every type of prime energy convertor at the outset of IVF. The original design was a strictly thermodynamic system without moving parts.  This had the regrettable feature of being impossible to test on earth.  But maybe someday.  We just need an orbital lab.

We considered turbines including devices with integral thruster function.  This was ultimately rejected as having several significant risks which could undermine the system operation.  We looked at old fashioned and PEM fuel cell systems as well as hybrids of these chemical systems and solar power. 

To understand our thinking we need to talk about how power is consumed on existing and future in-space vehicles. These are long-duration missions and we have very long periods of minimal power output punctuated with short-duration high-power demands.  In other words we have a very high power turn-down ratio- on the order of 30:1 or greater.  Most power systems do not like this sort of thing.  Fuel cell systems are excellent at sustained low-power demands but if you try to pull 20kW out at their low cell voltages you will run into rapidly increasing losses that ultimately degrade efficiency to the point that standard heat engines begin to look attractive.  This is not to say it can't be done.  It can be- generally at a large mass penalty.  The question is what is the cost and complexity of such a system and how flexible and extensible is it to missions we don't even know about now.

We considered turbine based systems for quite a while.  An innovative design called a single rotor turbine which combined a centrifugal compressor and turbine into a single wheel was considered.  Some sort of turbine-based system can almost certainly be developed but there was a lot more risk.  One of the key issues is that we are burning pure hydrogen and oxygen without the traditional nitrogen diluent.  We run at low MR and the excess hydrogen thus acts to suppress combustion temperatures.  But this means you have to pump a fair bit of hydrogen up through a reasonable pressure rise.  Pumping liquid hydrogen centrifugally is a nuisance but pumping gaseous hydrogen that way begins to fall in the category of very hard.  Making the H2 hot makes this problem worse.  You are facing multiple stages running at extremely high speeds.  We considered running at very low pressures but this does not push you towards compact, light, easy-to-build stuff.  While we could cool the turbine rotor with hydrogen we recognized that we were feeding it with a potential cutting torch.  Even a few milliseconds of high MR operation would turn the rotor to slag.  Certainly we could address this with superalloys and sophisticated burners and elaborate active cooling passages and control systems.  The question is: can you make something like this for $40,000? We didn't think so.   

Importantly we had to drive both generators and compressors and these don't want to spin at 50,000 RPM.  So we were facing multiple high reduction (likely planetary) gearboxes with significant losses.  We were also concerned about the power/time demand for starting such a turbine.  This stuff starts to mount up against a tight cost and mass budget. 

Based on prior art done by Vickers & NASA in the early 1960's we were pretty sure that a simple IC engine could produce this kind of power and do it cheaply.  Plus we had the benefit of decades of evolution.  We built a single cylinder engine and tried it- it worked great.  Then we tried a flight-weight, gas-cooled Wankel engine and it worked even better- except that we learned a lot about H2/O2 combustion and cooling that reinforced our judgement about a turbine.  Long story short we decided to dissipate the energy over a broader area with the 6-cylinder.  It is simpler to cool and lubricate and we can run at much higher chamber mixture ratios. 

After racking up about 250 hours over multiple engine evolutions and builds we have concluded that the multiple cylinder system fits our needs very well.  Its super-redundant, cheap, has a huge turndown ratio and is fast to build.  It starts fast and easy and has resisted all our attempts to destroy it via ignorance- a strong recommendation.  The thing is incredibly tough.  Weirdly all of our testing has pointed us more and more towards even more conventional ICE designs.  Though we were only asking for 20kW we have come to realize that we  could readily extend power up to in excess of 80kW.  Meaning that extremely large vehicles and all kinds of unknown demands could be supported without any mass change to the ICE.  That is a huge benefit. 

I mentioned in a previous post that IVF works very well with fuel cells and solar electric systems and here's why.  A fuel cell system on a long-duration cryogenic vehicle has to deal with widely varying inlet reactant conditions and in general the cells want feed pressures that are above what vehicle tank pressures run at during coast.  Much of IVF is a thermal control and heat balance system.  It has recirculating coolant, cryogenic heat exchangers and compressors that can act to support a fuel cell just as well as the ICE  It has the controller to manage all this and the power handling electronics to take low-voltage fuel-cell generated electrical power and boost it up to the 300V that allows us to move it around efficiently.  And it has a high-voltage battery to handle spike loading and permit the fuel cell to operate at peak efficiency.  The incremental mass delta to add the PEM cell stack into IVF is very small compared to a stand-alone system.   You run the ICE when you need and the fuel cell the rest of the time.  This extends usable mission time even further.  It takes all the good things from fuel cells, addresses all the risks and supplies all the gizmos required to get those cells to run properly for weeks.   

In general the same applies to solar.  You can keep array size down and suppress mass and cost by sizing to just the sustained average load.  This can mean the elimination of deployment mechanisms which are the biggest pain.  The power handling electronics are already on board the IVF controller. 

The six cylinder engine halves rotate together in the same direction.  Consider the engine and its loads together- net torques are very small except during speed transitions.  Vibration is very low.

Anyway I'm sure all this will spur further questions.  But at least you can now see the path that lead us to where we are.  It's a classic evolution based on learning from testing. 

z

[Kansan52]

Thanks! I understood!!. The only problem was extrapolating (hopefully correctly) Internal Combustion Engine from ICE.

[jimvela]

Frank Zegler of ULA here to answer a few questions about the IVF architecture and how we got to where we are. 

Thank you for taking the time to comment.
Kudos to ULA for engaging through NSF- you will find the signal to noise ratio very high here...

[Lee Jay]

I mentioned in a previous post that IVF works very well with fuel cells and solar electric systems and here's why.  A fuel cell system on a long-duration cryogenic vehicle has to deal with widely varying inlet reactant conditions and in general the cells want feed pressures that are above what vehicle tank pressures run at during coast.  Much of IVF is a thermal control and heat balance system.  It has recirculating coolant, cryogenic heat exchangers and compressors that can act to support a fuel cell just as well as the ICE  It has the controller to manage all this and the power handling electronics to take low-voltage fuel-cell generated electrical power and boost it up to the 300V that allows us to move it around efficiently.  And it has a high-voltage battery to handle spike loading and permit the fuel cell to operate at peak efficiency.  The incremental mass delta to add the PEM cell stack into IVF is very small compared to a stand-alone system.   You run the ICE when you need and the fuel cell the rest of the time.  This extends usable mission time even further.  It takes all the good things from fuel cells, addresses all the risks and supplies all the gizmos required to get those cells to run properly for weeks. 

In such a scenario, why run (or even have) the ICE at all?  Batteries are quite capable of supplying massive amounts of peak power (I have batteries at home capable of 2kW/kg sustained and 4kW/kg peak for 30 seconds) and I thought that was the whole reason to have the ICE in the first place.

Thanks so much for your presence and your detailed post on the IVF!

[Coastal Ron]

Frank Zegler of ULA here to answer a few questions about the IVF architecture and how we got to where we are.

Great explanation.  IVF is certainly an exciting technology, and we're all looking forward to the possibilities it can enable.

I have one question:
 

...The original design was a strictly thermodynamic system without moving parts.  This had the regrettable feature of being impossible to test on earth.  But maybe someday.  We just need an orbital lab.

Obviously the ISS was envisioned to help with such experiments, and no doubt you considered it.  Was there any particular reason that using the ISS at this time did not fit your needs?