G Scale Model Train Forum banner
1 - 20 of 23 Posts

·
Premium Member
Joined
·
112 Posts
Discussion Starter · #1 ·
I just acquired a Wada Works GP-9 with maybe one hour runtime on it And I am running into a problem that I think has been discussed before. At about a half tank of fuel it will lose fuel prime, or if I try to run it lean it will run great for a few seconds but then will lose prime. Is this a problem with the muffler system or the tank itself? Any advice would be a great help, rather then tinker with a almost new Loco and mess things up that need not be messed with.
Thank You.
 

·
Premium Member
Joined
·
126 Posts
Hi,
I'll preface this by telling you that I don't have a Wada GP-9, but I do have some experience with glow engines in locomotives.
1) If there are any air leaks in the fuel or exhaust system the engine will not run properly and will be a nightmare to tune. You can test this by blowing into one tube while you pinch off the other one. Any leaks should be evident. It could be a loose connection, a fuel tank lid seal gone bad, or a pinhole in a fuel line.
2) Run the engine rich - with plenty of blue smoke coming out. If you lean it out too much it will produce less (or no) smoke, sound fantastic, and run entirely too hot. Try running it as rich as possible.
3) If you can't remember when you replaced the glow plug then put in a new one. Often this is the first thing to try since it is simple to do. The only way to tell if a plug is bad is to put in a new one and try it out.
I hope this helps!
Bob Pope
 

·
Premium Member
Joined
·
491 Posts
Since you indicate that your engine is running fine for the first half of the tank, there is no question that the problem is a fuel-related one.

When a glo engine unexpectedly and repeatedly shuts down with half a tank of fuel, there is the possibility that the fuel pickup line inside the tank may have a hole in it. Also, if the glo engine is using high nitro fuel (25-40% nitro), over time that fuel can eat thru the brass in the clunk. The clunk is the heavy piece of metal on the end of the fuel line inside the tank. If that corrodes away, the clunk will fall off and the fuel line will pop up to the middle of the tank and can't draw the half of the fuel remaining in the tank.


So check for these sources of leaks as well as the air leak tests recommended earlier in conjunction with the muffler. Also insure that the muffler is mounted securely. It would probably be wise to go ahead and replace all these pieces of fuel tubing while you are at it. And if you do, be sure you have the right size tubing and make sure that the line inside the tank is flexible enough and/or the clunk is heavy enough to hold the tip of the fuel line in the bottom of the tank so it can pick up all the fuel. One last item is to be sure that there is no trash in the bottom of the fuel tank.


If all else fails, take your engine and visit a local RC flying field on a busy weekend. There will undoubtedly be someone there with plenty of experience with small glo engines who can quickly diagnose and resolve your problem. The guys flying small glo helis may be best because your engine and fuel setup will be most like what they are using too.

Good luck,

Ross Schlabach
 

·
Premium Member
Joined
·
112 Posts
Discussion Starter · #4 ·
Thank you all for the information. I will check everything you have listed. This Wada had less then one hour on it when I got it, so it dose need to brake-in. But right now I am running it way to rich just to get it to run. (3-3 1/2 turns) At that setting I am getting about 5 min run to a half tank. I will post what I find thanks for the help.
 

·
Premium Member
Joined
·
131 Posts
Hey Ross, great suggestions, but unfortunately, there's no clunker in these tanks. Its the type that drains from the bottom and has a push button primer, which I hear RC guys no longer like cause they tend to leak.

I sent Bill a pretty long email about checking everything, as the exhaust system has many parts. I haven't ran into this particular issue, but have ran into a few other doozies, that I hope isn't contributing to the issues.
With the half tank issue, the drain is at the bottom of the tank and the fuel line runs out of it straight up and level over to the carb. This level line and carb is about halfway up the tank, so I think when the fuel is above this line, its naturally feeds by gravity, and once the fuel drops under this, there's not enough back pressure to keep the fuel moving. But basically I had the same suggestions of checking the lines, rerout it temporarily as an experiment, use the soapy water/bubble check for leaks, check to make sure the carb needle is clean, and cross your fingers you didn't just run into the stress cracking problem in the muffler as I recently did (although its a failry easy fix).

Oh and Bob (sweet new live diesel btw!), good point. I have to press my fuel lid down a little to get a good seal, so Bill, double check that as well.

Good luck!
-Ray
 

·
Premium Member
Joined
·
112 Posts
Discussion Starter · #6 ·
Ray,
I got your email thank you, sorry for not replying sooner. Thank you very much for the information. If I get any time over the weekend I will work on the Wada again and see what I can find. The one thing I did notice is the fuel tank lid did not seem very tight but I could not tell with the body on it. When I can get back to it I will post what I find. For the time being do you know who makes or who I can order a replacement fuel tank from that will not require allot of mods.
 

·
Lord of the Steam
Joined
·
620 Posts
Another thing to take a look at is the fuel filter. There could be an accumilation of gunk on the screen. There should be one in the fuel line just after the line leaves the fuel tank. Its silver and can be unscrewed to get to the screen. I usually have to clean mine after three or four runs.

Also, check the lubricant trap (fake gas tank) under the engine. There have been reports of stress fractures which may lead to loss of adequate exhaust pressure to move the fuel along. There are two screws that hold it in place, you may want to drop the tank and give it a good once over to make sure there are no cracks that are hidden by the chassis body.

I personally have never had a stress fracture, but a couple of former customers did have that happen, and Wada sent me a replacement for them. His take on it (don't know how accurate this is - we did have a bit of a language barrier) was that the tank had excess lubricant built up in it that clogged the tank and caused the problem. I doubt that would be the case if the GP9 only has an hour on it, which would be only two runs or so since I normally get about 20 minutes if the engine is tuned correctly. I get less on the Zephyr just because it runs hotter and faster.

Best wishes,

Scott
 

·
Registered
Joined
·
13 Posts
Posted By scottemcdonald on 08/20/2008 12:08 PM
Also, check the lubricant trap (fake gas tank) under the engine. There have been reports of stress fractures which may lead to loss of adequate exhaust pressure to move the fuel along. There are two screws that hold it in place, you may want to drop the tank and give it a good once over to make sure there are no cracks that are hidden by the chassis body.
I personally have never had a stress fracture, but a couple of former customers did have that happen, and Wada sent me a replacement for them. Scott

Hi Scott,
Long time, no speak etc. I think I have this stress fracture in the lubricant trap on my GP9. It is visible as a hairline fracture on the side of the tank and a slight amount of exhaust smoke emits from it. What can I do about it - is it easily repairable? My engine also sometimes gives up with half a tank of fuel. I have replaced one of the fuel lines and will follow the other suggestions.
The main problem I have had with my engine is its habit of tripping out the CPU under load which requires a frequent reset. Any suggestions here? The good news is that I finally got the automatic throttle control to work (ATC) okay/DesktopModules/NTForums/themes/mls/emoticons/laugh.gif" border=0> The recent video I posted here shows my GP9 in action.
One final point, after great success with the Spectrum 2.4gHz receivers in other models, I tried the diversity receiver system in the GP9. Strangely the throttle servo became very jittery and unstable, so I am at a loss to understand why this otherwise foolproof system will not work in this application. I've gone back to the FM PCM system which seems okay, but for convenience I would have preferred the Spectrum system.
Perhaps we should be discussing all this elsewhere - this is the live steam forum after all/DesktopModules/NTForums/themes/mls/emoticons/tongue.gif" border=0>
Regards, Geoff (in the UK)
 

·
Premium Member
Joined
·
131 Posts
Bill,

No problem. This appears to be a similiar tank, but I'm not sure if the mounting holes line up:
http://www3.towerhobbies.com/cgi-bin/wti0001p?&I=LXWZ45&P=7
A few manufacturers also sell replacement O-rings for the cap (need to double check the measurements), and if the built in primer is leaking, you can take it out, seal the tank, and add an in-line bulb primer to the fuel line if required.

Scott, good call on the fuel filter, but mine didn't come with one, so I assumed it wasn't standard. I did add one for safety though.

Geoff,
That hairline fracture will probably grow, and 2 others will probably start, if they haven't already. I just had this happen to me about a month ago and reparied the tank. It started with the side/top fracture emitting exhuast. So I degreased the tank and stripped it. I found that this stress crack was a little worse then first thought and even goes through a brazed seem. I also found another stress crack starting on the bottom rounded portion of the tank where the drain is. So I drilled a few small holes in the stress crack lines and silver soldered it up with the 450 degree stuff. After this, I noticed another stress crack had started on the other big curve on the bottom of the other side of the tank, so soldered that one up as well.

I could be wrong about these theories, but I think what happend is on the first side/top crack, where the crack starts on the top of the tank is exaclty in the corner of the rectangular indentation where the fan shroud comes down for extra clearence. I think the rectangular indent was punched after the tank was assembled since it bowed in the side walls of the tank in a little, and probably pre-fatiged a stress line that shook loose over time with heat/cooling/vibration. The two cracks along the bottom of the tank on the large curves may also be pre-fatiged stress lines if the tank shape was cold-rolled. This compressed the metal on the inside of the curve and stretches them on the outside. Now add heat/cold expansion/contraction and vibration over time, and these would appear. I think I formed the second crack here faster as I heated the metal up to much more that it normal gets and cooled it down faster from silver soldering the first set of cracks, so basically accelerated the process. I'm hoping doing this, however, exposed all the fatigue points ahead of time, so hopefully the tank will be fine now, but I'm still putting it through its paces. The reason behind my theory is I happen to have a second used and broken wada tank that has the fatigue cracks in the same exact locations that mine started on. Here's some pics if you want to take a look:
http://1stclass.mylargescale.com/conrailray/wada_tank/

I've also had 2 problems with the on-board circuitry.
One, I could get the system to idle so slow (and under other specific conditions) which would produce low enough voltage and the voltage regulators would cut out. This would shutdown all the on-board electronics, and even the DX6-receiver, which takes a second or two to reset, and would make the throttle servo jump, usually choking the carb even more, and couldn't get out of that state and/or would stall the motor. Of course the easy answer is to not let it idle so slow, but what fun is that?!? ;) So since I already happened to add on-board starting batteries for the starter and glow plug (this has been really convenient btw, and I highly recommend to do it), I built a circuit to power the DX6 receiver off this battery so that in can never reset and an opto-coupler circuit to get the DX6 receiver to talk to the current on-board electronics to the traction motor control.
Two, I run into your problem as well when the on-board eletronics turn off the PWM driver and/or CPU. I think this has to do with heat build up I believe the PWM drivers will turn themsoelves off under too much load and/or heat, but need to doulbe check the specs of the ICs used), as it happens for the most part when I have the shell on, but will look into this more in the near future.

I also use the specktrum DX6 system, but haven't had any jitter problems. I couldn't get my older FM system to work, probably because the brass shell is causing interfearence. Interesting that we have opposite experiences!!

btw, I think the general live steam community has accepted the discussion of live deisels in the live steam forum (which I am thankful for as there's no where else to do it!)

Take care!
-Ray
 

·
Lord of the Steam
Joined
·
620 Posts
The main problem I have had with my engine is its habit of tripping out the CPU under load which requires a frequent reset. Any suggestions here?


This is usually the result of what Mr. Wada referred to as "over revving" which means that the rpm's of the glow engine are higher than they need to be. This means that the circuitry which sends pulses of dc energy to the wheel traction motors, trips out because it can't provide output at the rate of the input from the generator. It somewhat happens on the lower end when there is insufficient energy to power the circuitry, the receiver and lights only that instead of tripping out, you don't have enough energy at idle speed to power everything, so the speed controller doesn't activate when you push the starter button over to the run position, (i.e. little green light doesn't start flashing at an even interval).

Its a balancing act between the lowest and the higest useable rpm's. As to how that relates to actual rpm's and the cutoff points at each end of the spectrum - I have no idea what those points are.

Several years ago at Diamondhead we tried to use an RCS system to control a GP9. The gentleman, (so sorry I can't remember his name - Last name started with a Z?), who was one of RCS's technical wizards, had to program a chip specifically for the GP9 by doing a mod to the programming in the receiver in order to compensate for the dc voltage pulses. I vaguely remember something about the pulses not being a full 5 volts. Not sure if that helps or not as insight into the electronics of the Wada circuitry.

Scott
 

·
Registered
Joined
·
13 Posts
Posted By ConrailRay on 08/20/2008 11:29 PM


Geoff,
That hairline fracture will probably grow, and 2 others will probably start, if they haven't already. I just had this happen to me about a month ago and reparied the tank. It started with the side/top fracture emitting exhuast. -Ray




Hi Ray,

Thanks for your detailed reply - this is really helpful stuff! I'll get the tank out and see if I can repair it as you suggest. I may also try another DX6i receiver to see if it behaves any better.

Regards, Geoff
 

·
Registered
Joined
·
13 Posts
Posted By scottemcdonald on 08/21/2008 11:26 AM
The main problem I have had with my engine is its habit of tripping out the CPU under load which requires a frequent reset. Any suggestions here?

This is usually the result of what Mr. Wada referred to as "over revving" which means that the rpm's of the glow engine are higher than they need to be. This means that the circuitry which sends pulses of dc energy to the wheel traction motors, trips out because it can't provide output at the rate of the input from the generator. Scott




Hi Scott,

Thanks for your reply. I don't think I over rev the engine (the ATC takes care of that?), but perhaps you could take a look at the video I posted recently (http://www.youtube.com/watch?v=CYruU1O7-0Q) which shows my GP9 in action. In the third shot of the GP9 it actually trips out after passing the camera, which is why the last cars are seen grinding to a halt. Would you say from the exhaust note in the video that the engine was over-revving?

Regards, Geoff
 

·
Premium Member
Joined
·
740 Posts
Geoff,

What a great video. I had to forward it right away to a friend of mine who LOVES the H8. Was the Daylight an Aster or Accucreaft model? And did I see an S-12 in there too?
 

·
Lord of the Steam
Joined
·
620 Posts
Geoff,


I watched the video, and it doesn't sound to me as though the engine is over revving, it sounds like it is just above the threshold for power generation. The exhaust however is really dirty. Smokes like an Alco, not an EMD. So I have to ask - what kind of fuel are you using? The reason I ask is that if the fuel is too cool (low nitro percentage) the ability of the ATC to react in time to kick up the RPMs when needed might not be as effective because the engine doesn't respond quickly enough to the extra fuel. This means that the engine doesn't quite choke out because of low rpms, but also doesn't generateenough juice for the circuit. You might want to use the servo trim to slightly bump up the overall idle RPMs.

The fuel mixture might be also too wet, which will cause the engine to run too cool as well. Wada always insisted that I use a 20% nitro "sport" fuel. I've had some tell me that what I use is two stroke helicopter fuel.

Scott
 

·
Premium Member
Joined
·
131 Posts
Scott,
I was wondering how the ATC worked in theory myself, as I don't have this on my unit. From what I gather above, it sounds like the governer will adjust the engine carb servo to control the rpms/voltage generation output for the traction motors so the loco stays at a constant speed?
If so, this is totally opposite of how a real loco works. A real loco has neutral and 8 throttle notches. Each notch is a set increase in engine rpms and it will never deviate inbetween like you can with a car. The part that is automatically governed is how much power is supplied to the traction motors. For a bad example ;) , if you put the loco in notch 1, it will produce say 12 volts. The govorner will start to feed this 12volts to the traction motors and at the same time monitor the engine RPMS. So if 12 volts isn't enough to get the train moving, it will realize the traction motors are trying to pull more current, which in turn, starts to slow the engine rpms down and could stall it. The governer quikly realizes this and starts lowering the power to the traction motors so the engine rpms stay at a constant speed. So technically, if your train is heavy enough and you go to notch 1, it might not move at all becuase the governer knows it will stall out the engine and wont give power to the traction motors, which is why you usually see engineers crank up a few notches to get going and then back down.

Anywho, once I get things running smoothly, I'm planning on building a circuit that will monitor the enigne rpms with either a hall effect sensor or IR sensor. Then I'll program a uProc to take these readings and govern the power to the traction motors so that I'l have neutral and 8 throttle notches, each with a pretty constant engine rpm. I have to figure out how to design in a way to adjust this, as depending on the outdoor temperature can really changes your idle rpms. So when its cold out, I dont want notch one to really be idle!! I'll get to this sooner or later!

Take care!
-Ray
 

·
Lord of the Steam
Joined
·
620 Posts
This may be a dumb question but how does the ATC work?


Not a dumb question at all, and I'll do my best to relate how Wada told me.

The ATC adds a small spinning disk with holes in it on the front of the fan, which is attached to the shaft that exits out the back of the glow engine. Wada uses a LED sensor, the same that are used on disk drives to monitor rpm's, to do the same thing for the glow engine. The new circuitry monitors the engines rpm's and when the rpm's start to drop, the circuit tells the servo attached to the glow engine to increase the rpm's to avoid engine stall. When you don't have the ATC, you have to remember to slightly kick up the rpm's of the engine before increasing speed (more juice) to avoid stalling the engine cause the added request for dc energy from the generator puts more strain on the glow engine. So the ATC takes care of that for you. Mr. Wada said that the ATC can react faster than the operator, so if you have a track with grades, the ATC will ensure you don't stall as soon as the load on the engine changes from level to up. It will also allow you to go from full fast forward into a sudden change of reverse without stalling the engine, if the engine is tuned correctly.

That's how I understand that it works. As to all of the internal circuitry settings and thresholds of the rpm's, thats in the Wada vault.

Scott
 

·
Premium Member
Joined
·
126 Posts
Thanks for the explanation Scott.

Something I've always wondered - why did Wada go through all that complexity? Why couldn't he have just controlled the speed by controlling the engine throttle? The only thing I can think of is that the engine and generator put out enough current at idle to make the locomtove move.

Thanks,
Bob
 

·
Lord of the Steam
Joined
·
620 Posts
Something I've always wondered - why did Wada go through all that complexity? Why couldn't he have just controlled the speed by controlling the engine throttle?


That's a question for Mr. Wada ;-)

The very first 'prototype' of the GP9 that Mr. Wada brought over in 1994 to Diamondhead had two servos. One for the glow engine and one that controlled an rc car knife switch controller with a resistor as a heat sink. He was never really happy with that arrangement as he thought the car controller to be too rough of a method of controlling the engine. His son is the electronics expert in the family and developed the electronic speed controller and ATC.

This is just a guess, I think the complexity was to be able to control the speed smoothly. I really enjoy the ability to run the GP9 and do switching operations. There was a steamup several years ago at Dr. Rivets where I just stayed in the yard and played around moving everybody's rolling stock. It was a lot of fun, and the electronic speed controller and ATC made it very easy. I think that if it was direct dc to the motors based on engine rpm's only, you would either have speed demon that would be running much faster (like my Zephyr) and more difficult to control or choke out the engine because of putting too much torque on the engine at low speeds. The prototype with the rc car controller took a lot of 'practice' to get to know the engine sounds in order to be able to manually kick up the rpms to avoid choking out the engine. The electronics really made it easier to run from a beginner's point of view. You have to remember that he was trying to make it easy for a non-glow experienced person to be able to run the engine. There are only a few in our hobby, let alone the sparkie world who have ever played around with glow engines that I have met.

In retrospect, I think those who have had the best success with the GP9 are those who have had previous experience with glow engines in airplanes or cars. I know that even to this day, I still seek out the advice of those who have more years than me in understanding the little engines.
 

·
Premium Member
Joined
·
126 Posts
Thanks for the explanation Scott - I saw a video of the Wada doing some switching and it was extremely smooth and controllable.

And thanks to the live steamers that allow us to discuss diesels here - I'm not sure there is anywhere else we could go that would provide this amount of help and information.

Bob Pope
 
1 - 20 of 23 Posts
Top