I have never found a good source for describing what the numerous fans on a diesel electric locamotive actually perform. Say, for a example: an SD 40-2 has 2 fans grouped in the middle and 3 grouped toward the end of the long hood for a total of 5. Do some fans suck air into the hood body (and if so to where and why)? Do some air suck air out of the hood body (and if so why?). How about others (like an SD 75)?
Thanks.
On a SD40-2, the two center fans cool the dynamic brake grids while the three towards the rear cool the radiators.
VGN Jess I have never found a good source for describing what the numerous fans on a diesel electric locamotive actually perform. Say, for a example: an SD 40-2 has 2 fans grouped in the middle and 3 grouped toward the end of the long hood for a total of 5. Do some fans suck air into the hood body (and if so to where and why)? Do some air suck air out of the hood body (and if so why?). How about others (like an SD 75)? Thanks.
Thanks for the great information. I thought perhaps one of the fans sucked air into the engine for combustion. I suppose diesel-electrics just draw it in like in a car or truck? Turbos use exhaust gas back for combustion use, but fresh air gets into the combustion area how? Do the air filters connect to a port directly to the diesel intake manifold? As for the radiator fans, they are serving two purposes: 1) exhaust hot air out above the radiators, and 2) simultaneously draw cooler air thru the outside grills; right?
Thanks again.
I am amazed that nobody has commented on my post...
I should have said that the pressurisation fans, of course sucked air into the body. These were driven by the companion alternator, a D14 I think in the case of the A16C units. The pressurisation fans later received ducts covering them to prevent exhaust gas or dynamic brake outlet air being sucked into the body, since these units ran in the desert and it was pretty hot inside already.
A few hood units of export types G12C, G22C and some G6Bs had pressurisation fans. These all had engine driven radiator fans, and thus had no companion alternator. I hadn't thought much about what drove the pressurising fan until I saw one removed from a G6B and noticed that it had what looked like a standard air brake hose connected to it.... yes, it was driven by compressed air from the locomotive's main reservoir. That was a lot cheaper than fitting a companion alternator.
The discussion so far has been about EMD fans, because they are visible in a lot of cases, but GE locomotives have unseen radiator fans and dynamic brake fans that deserve some attention.
From the original U25 until the Dash 7 series, the GEs had engine driven radiator fans driven through a right angle gearbox. The U25 had two fans in line fore and aft, but they changed to a single fan from the later U28s onward. EMD actually built one locomotive with the U25 arrangement, two mechanically driven fans in line in the export GT16C for India, which was a sort of export SD24.
GE relied on the radiator fan(s) to cool the dynamic brakes in the U series and earlier Dash 7 series, locating the grids in the air intake below the radiator.
From the Dash 8 series onward GE adopted AC motor driven fans, initially two in line on the earlier units and later a single larger diameter fan under a wider, shorter radiator that projected further beyond the hood on each side. This remained current until the 2014 ES series (although they had two further high speed fans forward of the radiator for the separate air to air intercooler - deleted in the Tier 4 units built so far). I'd expect that the Tier 4 units have two radiator fans (again), as did the AC6000s and the ES44DCi and ES44ACi export units which had AC6000 radiators (and frame) together with the air to air intercooler.
From the late Dash 7 units onward, GE dynamic brakes were arranged with a fan in line with the resistor grids blowing across the locomotive, air entering and leaving through square openings at the top of the hood behind the cab. Generally, DC units have two dynamic brake units and AC units have two.
So the fans on GE units affect the appearance, even if they aren't themselves visible.
M636C
VGN JessI suppose diesel-electrics just draw it in like in a car or truck?
They draw it in through air filters in an 'air box', just as most modern cars or trucks do. In the 'old days' many of these filters were oil-bath filters (which were cleaned and re-used), and sometimes had an 'inertial' stage (turning the intake air around corners so the larger particulates are separated out, as in a cyclone filter). More modern applications that need better cleaning use replaceable filter elements. I think Mr. Clark knows far more about this than I do, and I welcome his comments in detail both on the historical 'timeline' and the evolution of use by specific builders.
Turbos use exhaust gas back for combustion use, but fresh air gets into the combustion area how?
I think you may be thinking about two very different things at the same time. Turbochargers use some of the heat in the exhaust gas to run a turbine, which drives a separate compressor that compresses the intake air to higher density. The exhaust from the turbine goes up the stack and is not returned to the engine. The intake air to the compressor, just as in turbocharged cars or trucks, comes through an air filter first.
Exhaust gas recirculation is the use of some of the exhaust to replace air in the intake charge. A specific use of this is to reduce the amount of nitrogen admitted to an engine's cylinder on the intake stroke, which can cut down the emissions of nitrogen oxides (the NOx that has been so troublesome to CAT and EMD). I don't know of any use of EGR that circulates the exhaust to a point ahead of the turbocharger, though -- it would have to be cooled down to a significant degree to preclude needlessly high post-turbo pressure, or involve a larger (and more expensive) intercooler to take out the extra heat. Again, Mr. Clark will have better detail information on where and how the exhaust gas is introduced in locomotive diesel engines.
Do the air filters connect to a port directly to the diesel intake manifold?
Essentially yes, the idea being to reduce the restriction of intake airflow as much as possible within design limits. On a turbo engine, the filtered air goes to the compressor inlet and not the intake manifold of the engine per se, but the compressed air then goes through to the intake manifold (via the intercooler) with as little flow restriction as possible.
Firstly, for anyone who didn't realise, I started my post before VGN Jess reponded and of course didn't see it until I had posted. I think I may have been distracted in mid post, which didn't help.
To answer the question about engine inlet air fans, in all of the pressurised export units the engine air was drawn from inside the body, having been drawn in by the fan (which looked just like an EMD 36" radiator fan, incidentally).
One of the innovations introduced by the GE U25 was a sealed carbody with what was called a "centralised air system", which was copied by both EMD and Alco. This meant that all air went through at least a primary filter, usually an inertial filter and then combustion air went through media filters, usually oil bath originally and later paper. A blower fed the primary air to traction motors. Before this time, locomotives usually had louvres in the hood doors adjacent to the engine air intake with oil bath filters mounted on the inside of the doors. The change occurred with the GP30 with EMD and the "Century Series" for Alco.
The only locomotive type I can recall with a visible inlet air fan is the Alco Century 636 which had a large fan set flush into the top of the intercooler box which fed both the engine and the intercooler radiators just below it (when the external louvres were open). The air then went through primary filters and into a very large main blower, where it went to the traction motors, but also through a duct formed by the main frame beams and sealed off by the fuel tank to the back of the engine and into an "air box" with secondary filters and into the turbocharger, which was on the rear end of an Alco locomotive engine.
One of MLW's simplifications in the M630 and M636 was to combine the intercooler circuit with the main radiator (which was 1/3 larger than on the M636 compared to the C636). But the combustion air still went under the engine. Some later modifications introduced primary air filters at the rear end of the engine.
GE, sensibly, always drew the engine air into the body at the rear, usually just forward of the radiator inlet.
Please don't confuse me with someone who actually knows what is going on with exhaust gas recirculation. I personally asked an EMD engineer and he didn't say anything about EGR. A colleague had a similar response from GE, who were actively trying to sell him the V250 (marine GEVO) at the time.
However looking at the excellent photo on the V250 Tier 4I brochure, there appear to be three heat exchangers: two are intercoolers, one for each bank of cylinders and I suspect the third cools the exhaust gas. I don't know if the exhaust is taken from before or after the exhaust turbine, but if after, I'd assume that it would need to go back through the turbocharger to be raised to the correct pressure.
My colleague suggests that the purpose of the exhaust gas is to reduce the oxygen content of the inlet air, which will reduce the combustion temperature and thus reduce the formation of NOx gases. I'm told that multiple injections of smaller quantities of fuel in a single stroke helps ensure the same lower temperature, and this can only be obtained with common rail fuel injection. Since air is largely nitrogen, adding exhaust gas wouldn't significantly reduce the nitrogen content enough to reduce NOx formation.
M636C My colleague suggests that the purpose of the exhaust gas is to reduce the oxygen content of the inlet air, which will reduce the combustion temperature and thus reduce the formation of NOx gases. I'm told that multiple injections of smaller quantities of fuel in a single stroke helps ensure the same lower temperature, and this can only be obtained with common rail fuel injection. Since air is largely nitrogen, adding exhaust gas wouldn't significantly reduce the nitrogen content enough to reduce NOx formation. M636C
EGR does not reduce the amount of nitrogen present in the intake charge. In fact, it actually increases it. Similar NOx reductions can be achieved by introducing pure nitrogen to the intake charge.
AnthonyVPeak temperatures are reduced because of the increased heat capacity of the intake charge due to the added mass of the exhaust gases present.
Can you explain this a little better. It doesn't seem to make full sense to me as written. The "mass" of each of the constituents of recirculated CO2 is very little different from the mass of nitrogen (and at 12 and twice 16 they bracket twice 15), and in any case a mass effect would have little to do with the 'heat capacity' of an admitted charge. I don't have my rubber bible handy, but is the heat capacity of CO2 that much greater than the equivalent number of moles of carbon and oxygen? I suspect you well know what you mean, with CO2 being a denser gas due to its structure, but I'd like to see it worded differently.
EGR does not reduce the amount of nitrogen present in the intake charge. In fact, it actually increases it.
My experience with EGR and FGR (the external-combustion analogue) is that just the opposite is true. To the extent CO2 displaces air in the mass of an intake charge, the amount of nitrogen is de facto reduced. To the extent it dissociates at high peak temperatures, it is much more likely to recombine than to preferentially react with a temperature-dissociated N (or allow N to react with available O/O2) to produce NOx. That is recognized as a fundamental element of a number of the 'cleaner coal' schemes.
That's not to take away from the essential point I would make (and I think you are making) -- that the major point of EGR is to slow down the oxidative "combustion" reactions taking place after ignition during the power stroke, and thereby decrease the peak temperature achieved during the reaction so that you'll get adequate expansion due to heat release causing the charge gases to expand (which is what produces the shaft power out of this kind of heat engine) while staying below the temperature range where oxidation of diatomic nitrogen (to the nitrogen oxide mix we call NOx) becomes critical. Pilot injection and modulated injection of the fuel charge do much the same thing*.
It can be interesting to compare the rate of effective-pressure development over a power stroke with heat release from oxidative combustion, when limitation of temperature-generated NOx is being factored in. This is much of what is behind development of genetic algorithms for complex injection mapping under different load conditions, for example.
You might want to mention that "nitrogen injection" as a substituent for EGR in NOx reduction would be limited fairly strictly to positive-displacement internal-combustion motors with very limited specific output, certainly not to Brayton-cycle turbines or most forms of coal combustion. But even so, this would result in so crippling a diminuation of peak achievable power in, say, a market-competitive design of locomotive diesel engine that no one sane would do it. (Let alone with the stringent Tier 4 final absolute limits on NOx in effect!)
An interesting associated issue involves aftercooling of the EGR charge. Theoretically it is desirable to retain as much of the exhaust heat as possible to maximize the Carnot (well, it becomes Rankine with EGR) efficiency. Ford experimented with this in the '70s (using thermal barrier coatings many places in the intake tract and engine, and overdesigning for high peak pressure); they were able to achieve very high sfc (and hence mpg) in small trucks with this approach. Even if modulating injection to reduce peak NOx generation, you can still develop significant fuel economy from 'hotter' EGR in many engines ... but be prepared for the kind of peak intake-manifold pressures that can be common to modded 6BTs. (The thought of such manifold pressure on a locomotive scale is not exactly reassuring... ;-} )
*Early pilot injection also was intended, at least in the engines I worked on, to provide chemical promoters for a more distributed light-off of combustion reactions in a lean-burn charge; in a CIDI engine this might produce up to hundreds of small ignition kernels even before the body of a modulated injectioln charge starts to be admitted. This is much the same effect as toluene or some of the other fancy diesel additives are supposed to generate when added to fuel in some light diesels, or the vaunted HHO "Brown's gas" does (they react 'early' in charge combustion to facilitate fuller reaction of the hydrocarbons in the injection charge plume, rather than just elevating temperature and post-intake charge pressure in the cylinder). This in theory would allow a much 'leaner' burn in terms of active oxidation taking place on the scale, and within the timeframe, required to get meaningful power out of an IC engine, and hence allow a higher percentage of combustion gas in an intake charge.
Overmod My experience with EGR and FGR (the external-combustion analogue) is that just the opposite is true. To the extent CO2 displaces air in the mass of an intake charge, the amount of nitrogen is de facto reduced. To the extent it dissociates at high peak temperatures, it is much more likely to recombine than to preferentially react with a temperature-dissociated N (or allow N to react with available O/O2) to produce NOx. That is recognized as a fundamental element of a number of the 'cleaner coal' schemes. That's not to take away from the essential point I would make (and I think you are making) -- that the major point of EGR is to slow down the oxidative "combustion" reactions taking place after ignition during the power stroke, and thereby decrease the peak temperature achieved during the reaction so that you'll get adequate expansion due to heat release causing the charge gases to expand (which is what produces the shaft power out of this kind of heat engine) while staying below the temperature range where oxidation of diatomic nitrogen (to the nitrogen oxide mix we call NOx) becomes critical. Pilot injection and modulated injection of the fuel charge do much the same thing*.
The number that I remember from my engineering thermodynamics class was 2700F - the professor was a member of CARB at the time. He mentioned that a very simple trick to reduce NOx from pre-1971 cars was to retard the spark a few degrees, though at the expense of power and mileage.
A neighbor of one of my high school friends did one science fair project variable EGR, one finding was that a typical car engine from the 1960's would tolerate a very large amount of EGR at cruising speeds. My recollection was that this was for the 1977 science fair and interestingly enough, he is now a EE prof at Stanford.
Keep in mind that the recirculated exhaust gas has a lot of water vapor as well as CO2, both having a higher constant volume heat capacity than nitrogen.
- Erik
erikem Overmod ... staying below the temperature range where oxidation of diatomic nitrogen (to the nitrogen oxide mix we call NOx) becomes critical. The number that I remember from my engineering thermodynamics class was 2700F - the professor was a member of CARB at the time.
Overmod ... staying below the temperature range where oxidation of diatomic nitrogen (to the nitrogen oxide mix we call NOx) becomes critical.
The number that I remember from my engineering thermodynamics class was 2700F - the professor was a member of CARB at the time.
Remember that this figure is thermal NOx under close to ambient pressure conditions -- as in FGR in a power boiler.
A neighbor of one of my high school friends did one science fair project variable EGR, one finding was that a typical car engine from the 1960's would tolerate a very large amount of EGR at cruising speeds.
This is essentially the principle of the Fish carburetor (and other wonder carburetors) -- a very lean, but hot mixture can still ignite (in part via polynucleate autoignition in the cylinder) to produce enough heat/pressure to generate power from an engine. The difficulty is that very small changes in demanded power can (and do) cause an engine operating that way to stall, and a carburetor system has little hope of responding correctly in the timeframe involved. More modern control modalities make this possible ... but they are almost always much better employed in injector modulation than in mixture control...
Thank you for reminding me -- I got carried away with the pure gases.
By the way, I was wrong. I was disparaging the idea of added nitrogen as a sensible way to decrease NOx. See this article and some of its references, notably 10-12, that appear to show exactly what he was claiming.
After skimming through the article (noting figure 7), it looks like the advantage of the membrane comes more from depletion of oxygen as opposed to enrichment of nitrogen. I'd wonder if more benefit could be achieved by enriching the oxygen content of the supply air and utilizing EGR to keep combustion temperatures under control (or maybe try argon injection...).
One take-away from figure 7 is that the NOx peaks about 35% load with standard air, which suggests the engine is running around 200% theoretical air. Idle is lower as the large amount of excess air quickly cools the combustion products, full power is lower as most of the oxygen is used up by the fuel.
erikemAfter skimming through the article (noting figure 7), it looks like the advantage of the membrane comes more from depletion of oxygen as opposed to enrichment of nitrogen.
I think you are right; the idea seems to be having an inert-at-prevailing-temp gas in the charge that slows down the oxidative heat release from the fuel without dropping parts of the charge mass below necessary transition temp for the combustion reactions. As I see it, we're not 'burning' the fuel, we're converting it to power gas as nearly at stoich as possible, and the nitrogen serves as a diluent to keep peak reaction temperature below critical.
I'd wonder if more benefit could be achieved by enriching the oxygen content of the supply air and utilizing EGR to keep combustion temperatures under control (or maybe try argon injection...)
A problem with argon injection is that you'd have to carry a pressure bottle of argon with you... or have a really, REALLY fancy arrangement to get 1/120 or so of the atmosphere separated out in realtime. The CO2 in EGR does this job every bit as well as a noble gas (remember what a good shielding gas it can be) and is not only free but serves a noble purpose (I couldn't resist the pun) in adaptive re-use. One point they stress in the nitrogen-enrichment application is that very little equipment is required to obtain the relatively slight enrichment of N2 they're using, and the supply of N2 is essentially limitless with no 'user service' being required (and of course no SCR rigmarole added on top of the EGR system).
My approach has always favored, as you say, enriching the oxygen content in the charge (so you need a lower charge mass of 'air' including a proportion of nitrogen, and then trying to run the charge as hot as possible to keep the thermodynamic efficiency up. Note that this usually involves higher peak cylinder pressure, for which the engine must be designed and maintained. The approach in the paper was to provide a low-cost add-on for an existing type of diesel engine. I had not thought about pursuing that type of approach, and I am now glad I have been introduced to it.
Idle is lower as the large amount of excess air quickly cools the combustion products, full power is lower as most of the oxygen is used up by the fuel.
I would wonder, though, if the engine-management system would be looking at a fixed idle-speed reference, and feed a bit more fuel to keep the idle 'on spec' rather than let it sag down to where standard idle mixture would let it.
Note the potential point in Table 2 (p.3) that 'hot EGR reduces engine power' (I presume this is because its heat reduces charge density) but also the implication that cooling the charge air and reducing "excess" oxygen will reduce particulates. (Which seemed to me to be somewhat opposed to your observation of their running at 200% theoretical air...) I had assumed that particulate generation was the result of some combination of premature combustion quench and insufficient oxygen for full carbon oxidation, so some of what they appear to be doing seems counterproductive in that very significant respect. I will read more carefully to see how this is addressed.
Be interesting to see if their installation on a C12 would scale to, ahem, a C18, wouldn't it???
Wizlish (and all others): Thank you for all the information. I felt my questions were initially "dumb", but in reading the responses to date, I understand all the EMD hood type fans are interesting to others as well and I'm glad I asked. I have a 100% better understanding now. I am a big TrainSimulator 2015 user and often wondered about the fans used on the EMD locomotives in that application, such as: GP7s, GP9s, FA7s, GP38-2s, SD 40-2s, SD45s, SD60s, SD70s, sD70ACes, and SD80MACs. Thanks again!!!
The next thing you might want to ask is why some of the fans have irregular blade spacing... and how the blades are shaped and flow-streamlined to minimize noise. Ask M636 about Q-fans, for instance.
Firstly to deal with YoHo's point about "Q-Fans".
There is a classic line about "wasted youth" in Pool Halls, but I have wasted much of my late middle age and early old age on a footbridge in Moss Vale where good photos can be taken in the early afternoon of north bound freight trains. However, locomotive for local limestone shuttle trains literally refuel under the bridge so I've seen a number of EMD fans (on JT26C-2SS) stopped literally a couple of inches below my feet (the bridge built in 1915 is right on the clearance limit).
The fans appear to be designed (from memory) with sixteen fan blade positions of which twelve are filled with four evenly spaced gaps. Even with gaps the fans need to be accurately balanced, of course.
What is the advantage of unevenly spaced blades? The blade passing frequency if constant, would excite a resonance which would form a major part of fan noise. By leaving every fourth blade out, the resonance does not build up and that source of noise is eliminated or substantially reduced.
I'm not familiar with the other airflow details of Q Fans. I did see a description years ago...
I understand the "tunnel" radiator arrangement was requested by Southern Pacific because of overheating in tunnels and snow sheds, particularly with the SD 45. The GE U36C and similar units did not suffer this problem, at least partly due to drawing the air from frame level where it was cooler and not mixed with radiator outlet air and exhaust.
By placing the fans under the radiator, the air velocity through the radiator is increased. This allows the radiator to regain normal temperature more quickly (between tunnels). The down side is that the air drawn from nearer the track has more dust and grit and the higher air velocity erodes the copper of radiator cores more quickly so the radiator doesn't last as long.
So there are swings and roundabouts....
EMD decided that their original layout was better. Had SP purchased straight SD45-2 units, they might have agreed. But EMD gave a very big customer what they asked for and it worked as designed.
But SP purchased and liked SD70Ms and when UP inherited them, they bought around 1500 in total for much the same duties.
Our community is FREE to join. To participate you must either login or register for an account.