|
Post by llxlocomotives on Aug 5, 2015 12:23:21 GMT -8
The reason for showing the U36C is it is simiar to the Atlas Master train master in every respect in the measured performance. Yes it is a red box. Rightly or wrongly, I put all Atlas China engines in the post 2000 era. Drawing a line like that is always difficult and will not satisfy all.
Ed, These points are where the motor just stops raising the weights off the floor. A gram of so less and the engine pulls it up slowly. Neither the SD40-2 or the C-636 spin the wheels. The other engines on the chart do at all points. The SD40-2 is the average of four and five readings at each point because it showed to be so different. The others are an average of three readings. The variation is at any weight point is a few grams.
In regards to how much weight it takes to move a non powered engine, it does vary. Weight and wheel type are the major factors. The NS wheel engines move with less weight on the pulley.
At this juncture it is hard for me to see that the SD40-2 and the C-636 have the same model motor. They may, but they do not respond the same. One unique difference is the decoder on the SD40-2. Not sure why yhat would matter.
|
|
|
Post by Spikre on Aug 5, 2015 12:28:29 GMT -8
?? Larry, are the Gear ratios the same in the IMRC SD40-2 and Bowser C636 ? does that make much differeance ? always liked that Hobbytown of Boston had Optional Gear Ratios back in the bronze ages. Spikre
|
|
|
Post by edwardsutorik on Aug 5, 2015 13:26:03 GMT -8
always liked that Hobbytown of Boston had Optional gear Ratios back in the bronze ages. I am enjoying figuring out what to do with a couple of Hobbytown SW switcher drives with low speed gearing. SO MUCH FUN!!!!! Ed
|
|
|
Post by Judge Doom on Aug 5, 2015 14:46:21 GMT -8
At this juncture it is hard for me to see that the SD40-2 and the C-636 have the same model motor. They may, but they do not respond the same. One unique difference is the decoder on the SD40-2. Not sure why yhat would matter. It seems that many have forgotten that the IMRC units with only DCC decoders (non sound) were intentionally made to run slower on DC, so that they could run/MU at the same speed with DCC/sound equipped IMRC units, which typically take more voltage to get going on DC because of the sound installed. It seems rather stupid in the grand scheme of things, since now the basic non-sound units won't run well on DC with anything else offered unless they're run on...DCC. The Bowser units don't have this, so on DC power a DC Bowser unit will run more typical than a DCC (non-sound) IMRC SD40-2. While some tests performed on the IMRC unit on DC may still hold true, others like a speed test or starting voltage test will obviously be crippled by the decoder.
|
|
|
Post by llxlocomotives on Aug 5, 2015 15:07:03 GMT -8
I covered the voltage function impact earlier. At that time the start voltage was pointed out. The question is why does it not recover? It is still way short at at 16 volts on the supply. The implication is that the decoder is impacting the voltage by 2 volts. The curves indicate that when the SD40-2 is at 16 on the supply, the C-636 is at 14 volts.
So taking that possibility a little further, in looking at the drawbar force at 12 on the SD40-2 would it be like a running a little over 10 on the C-636?
It is becoming apparent that a project is at hand to better understand the problems that are impacting the SD40-2.
The initial step is to run the C-636 at 10 volts and see if it replicates the SD function shape. At that point, this comparison will have run its course. The project on the SD40-2 to better understand the underlying cause and demonstrate improvements is an activity of its own.
|
|
|
Post by southeastern on Aug 5, 2015 22:46:21 GMT -8
I have several of the Intermountain SD40-2's in the GATX scheme, performance was not real good straight from the box. After a bit of running, they got more sluggish to the point that I became quite concerned. Not even enough power to spin the wheels. Had a bit of a tinker with the worst one, tried another decoder (NCE DASR), checked the trucks for binding, cleaned the wheels, lubed everything, still no good. My intention all along was to strip the paint, try and fix some of the flaws on them but now they need re-motoring. The wheels aren't much better with the plating already lifting off. Think they will become non powered dummy locomotives very quickly. Now I'm worried about all the U18B's I've got sitting in the cupboard, I hope they don't have the same motor.
Dave
|
|
|
Post by WP 257 on Aug 6, 2015 5:27:57 GMT -8
Summary: Plating lifting off wheels (unfortunately relatively common in HO with several manufacturers) = Good for coefficient of friction and tractive effort = Bad (or less optimal) for electrical conductivity and DCC.
Sometimes you have to take the good with the bad. We can't keep the plating on the wheel sets if the original plating process was not as good as it should have been.
I've actually seen HO diesel wheel sets that were plated more than once by the manufacturer in China. There's an outer layer of plating that quickly peels off in less than 10 or 20 hours of run time, but beneath that still another layer (or partial layer) of plating. Perhaps sometimes the manufacturer realized they had a plating problem and then ran the wheelsets through the plating operation a second time rather than scrap them. The importer would unfortunately be none the wiser, so I can't blame the importer for that particular situation, as unless they were physically right there they might have no idea something was plated more than once.
Except that sometimes you can kinda tell if you look. If the wheel tread has slight surface irregularities, is other than perfectly smooth out of the box, I have found that those wheel sets are the ones where plating will quickly come off. I think if it's been plated more than once, the wheel surface out of the box tends to have a "rougher" (but not pitted like the sintered wheelsets) look to it.
I have yet to see Kato wheelsets that had anything less than fantastic plating--it takes forever to wear by comparison to the Chinese stuff--so clearly the Chinese cut corners on their plating operation.
It also seems to me that Athearn has changed either their plating metal or the process as the Genesis series of engines has progressed along. The earlier ones, at least some F units, had plating that peeled in no time and utterly trashed my track. The later, more recent Genesis units like F45's, FP45's, and the recent GP-7's seem to have much better wheel plating--and I did not have the issues with horribly dirty track from plating coming off in no time at all. The GP-7's and -9's are actually pretty sweet runners. These more recent wheelsets also seem to my eyes to be a slightly different color--darker--than the earlier Genesis wheelsets.
John
|
|
|
Post by llxlocomotives on Aug 6, 2015 15:05:41 GMT -8
You all know what a eureka moment is. Sure enough, running the C-636 at 10 volts, it hit me over the head. The draw bar force fell between the second and third weight increments. I am not ready to display the chart, yet. I am going to run some more data on the 636 and SD40-2. This result does support the notion that decoder voltage loss is a significant factor in the performance we are seeing.
We all knew that there would be some loss through the decoder. This implies that that loss is approaching 20 percent. Since voltage is the driving function for the motor this makes comparing the decoder equipped SD40-2 or any other engine with a decoder difficult.
This highlights the problem with using motors fundamentally sized for a DC in a DCC system. It is understandable that sound, low speed and current draw reduction are impacting the motor sizing. The reality is that a significant portion of the hobby is not buying into sound and DCC for that matter. Yes there is a large market for them, but at what price, performance and cost? That is the subject of a different discussion.
I believe this is implying that more high end voltage needs to be available to the motor. Less drop in the module. More voltage to the module. I suspect that the wireless DC module systems will have an advantage here. To highlight this, the additional testing on the SD40-2 will examine the DBF at supply settings above 12 volts. The irony here is you can slow the engine with PWM by reducing the time it is on full voltage, but full power is on 100% of the time at the voltage out of the decoder.
This testing will take a few days, with warm up each run takes about forty five minutes. I'll post the results As soon as possible.
|
|
|
Post by bnsf971 on Aug 7, 2015 2:52:51 GMT -8
Larry, Intermountain will supply a "dummy plug" to replace the stock decoder, so you can remove the decoder from the equation. All you have to do is call them.
|
|
|
Post by llxlocomotives on Aug 7, 2015 4:49:54 GMT -8
Terry, Thanks. I have all but decided that this unit should be involved in a restoration activity. The extent of that is still in the planning stage. Part of that will be to take the electronics out of the system in some fashion. That plug will be just what is needed in a first step in that activity.
I will likely define the plans on my blog. My practice in posts on a forum like this is to present this is what I have done and here are the results.
|
|