[superheatedsteam]

 railguy, on 28 September 2021 - 12:42 PM, said:

I did a quick test of this in Unstable release 0321 earlier this afternoon. It appears that everything is working, with only one possible (and maybe unrelated) exception. It appears that applying the Independent brake does not deplete air from the main reservoir as it should on an independent brake application. Either that, or the flow rate is so slow that depletion goes almost unnoticeably. The compressor sound triggers, both for electric and mechanical compressors, worked correctly for me.

Yes, confirming that use of independent brake on v1.4.rc2 or v1.4.rc3 does not drop main reservoir pressure. The pressure drops, as expected in v1.3.1 and x1.3.1-373

[pschlik]

 steamer_ctn, on 29 September 2021 - 12:13 AM, said:

The existing recharge rate is equated to Max RpM, and then a linear relationship is adopted to vary the charging rate.

Do you have manufacturers graphs or tables for the RPM vs output air volume?

I’m not sure this is something anyone bothered to measure. If you are lucky, a locomotive manufacturer may have measured the CFM at idle and at max RPM-what happens in between? Who knows. All I ever heard is that wabtec compressors are close to linear but other brands like Gardner Denver exhibit subtle quadratic behavior. (Using a linear relationship is not a horrible approximation, but it is a bit idealistic.)

Really the main reason I’d like to enter in my own values for X rpm and Y psi/s is to model the behavior of GE locomotives which is neither linear nor constant with respect to RPM. It would also be handy should someone ever find graphs for mechanical compressors.

Of course, the real best case scenario would be if I could enter in values for cubic feet per minute instead of using psi/s :p

[steamer_ctn]

 superheatedsteam, on 29 September 2021 - 03:27 AM, said:

Yes, confirming that use of independent brake on v1.4.rc2 or v1.4.rc3 does not drop main reservoir pressure. The pressure drops, as expected in v1.3.1 and x1.3.1-373

I will have a look at this.

[steamer_ctn]

 pschlik, on 29 September 2021 - 08:53 AM, said:

I’m not sure this is something anyone bothered to measure. If you are lucky, a locomotive manufacturer may have measured the CFM at idle and at max RPM-what happens in between? Who knows. All I ever heard is that wabtec compressors are close to linear but other brands like Gardner Denver exhibit subtle quadratic behavior. (Using a linear relationship is not a horrible approximation, but it is a bit idealistic.)

After a quick internat search, I managed to find this document. It tends to support your suggestion that the mechanical compressors were linear (even though the graphs in this document are not for the air delivered). This makes sense when I think about it, as the air cylinder pumps a certain amount of air each cycle, so increasing the cycles will increase the air delivered by the same increase in the cycles (linearly), and hence the pressure increase will also increase linearly.

I would be interested to see why the GD compressors were quadratic, this implies that the drive shaft was not directly connected to the engine drive shaft, and some difference in rpm existed.

 pschlik, on 29 September 2021 - 08:53 AM, said:

Really the main reason I’d like to enter in my own values for X rpm and Y psi/s is to model the behavior of GE locomotives which is neither linear nor constant with respect to RPM. It would also be handy should someone ever find graphs for mechanical compressors.

I assume that the first bit is about electric compressors? Do you have manufacturer graphs showing the variation in air output to engine rpm?

In regard to the mechanical compressors, if it is a linear change, then there is probably not a great deal of advantage in adding a table.

 pschlik, on 29 September 2021 - 08:53 AM, said:

Of course, the real best case scenario would be if I could enter in values for cubic feet per minute instead of using psi/s :p

I agree, however this will need a significant rewrite of the code to ensure that the full pneumatic operation is modeled. In other words, we would need to define all the relevant volumes associated with the braking system (brake cylinders, auxiliary reservoirs, etc), and accurately calculate the amount of air used during braking operations so that we could work out how much air we need to replace, and hence based upon the compressor size, how long it would take to recharge the reservoir.

Perhaps somebody will be interested in taking on this task. http://www.elvastower.com/forums/public/style_emoticons/default/drool3.gif

[steamer_ctn]

 superheatedsteam, on 29 September 2021 - 03:27 AM, said:

Yes, confirming that use of independent brake on v1.4.rc2 or v1.4.rc3 does not drop main reservoir pressure. The pressure drops, as expected in v1.3.1 and x1.3.1-373

Hopefully there should be a patch correcting this issue in rc1.4 shortly.

Can you please test and confirm.

Thanks

[Weter]

There are rotor/screw compressors design... Maybe them are quadratic?

[pschlik]

 steamer_ctn, on 29 September 2021 - 08:18 PM, said:

I assume that the first bit is about electric compressors? Do you have manufacturer graphs showing the variation in air output to engine rpm?

Ah so GE's methods with electric compressors are...interesting. From page 75 of this ES44DC manual, "The air compressor speed is twice engine speed when the engine speed is below 525 rpm." Though this isn't quite the full story, as the air compressor speed will not be twice engine speed when the throttle is idle.

Might sound odd that the electric compressor RPM depends on the engine RPM but it runs on AC, the frequency of which depends on the RPM of the engine. So it behaves like a mechanical compressor except for that 2x speed thing below 525 rpm (which is notch 2). An (approximate) plot of this behavior is something like this. (If I could, this is the data I'd give to the simulator to work with).

https://imgur.com/2i9TWtP.jpg

From 0-450 RPM (basically, engine off to throttle idle) the flow rate is pathetic, as one could expect from a mechanical compressor. But once you set the thing to notch 1 the potential flow rate shoots up, in stark contrast to a mechanical compressor. It's an unusual configuration, but it entirely avoids the need to waste fuel and make noise just to charge the train. The fact that notch 1 and notch 2 power have noticeably higher charging rates is just something that can't be faithfully represented right now.

[ErickC]

That's not too different from traction motor blowers, which also are driven by the auxiliary alternator on most locomotives and are thus tied to engine RPM (which is why most locomotives power up in dynamic braking).

I'll try the revised build over the weekend. Work is keeping me very busy at the moment.

[superheatedsteam]

 steamer_ctn, on 29 September 2021 - 10:24 PM, said:

Hopefully there should be a patch correcting this issue in rc1.4 shortly.

Can you please test and confirm.

Thanks

Hi Peter,

Sorry for the delay. I missed 1.4rc4 but have just tested 1.4rc5 and issue persists. Main reservoir pressure unaffected by multiple independent brake application/releases.

Cheers,

Marek.

[railguy]

 pschlik, on 30 September 2021 - 03:28 PM, said:

From 0-450 RPM (basically, engine off to throttle idle) the flow rate is pathetic, as one could expect from a mechanical compressor. But once you set the thing to notch 1 the potential flow rate shoots up, in stark contrast to a mechanical compressor. It's an unusual configuration, but it entirely avoids the need to waste fuel and make noise just to charge the train. The fact that notch 1 and notch 2 power have noticeably higher charging rates is just something that can't be faithfully represented right now.

That explains something that I've noticed on the GE's. When the compressor fires up (with the characteristic "whoop"), you will often hear the prime mover blip up slightly in RPM--probably to Run 1, then drop back down when the compressor shuts off. Now that makes sense, the prime mover "blip" is the engine picking up enough RPM so the compressor goes into higher air production mode.