G Scale Model Train Forum banner

1 - 20 of 41 Posts

·
Premium Member
Joined
·
127 Posts
Discussion Starter · #1 ·
I have sound working in my K-27 with a Phoenix 2K2 sound card.
The only problem left is that the sensors on the left side (I think) are not positioned exactly correct and I get 4 chuffs and a slight space.

I got my K-27 at the end of last week and spent all my time trying to get the optical sensor to work. The engine was going to run at the Del Mar Worlds Greatest Hobby on Saturday  so I ended up running this weekend with voltage controlled chuff.

The engine performed very well, I was pulling 24 cars around the DelOro layout without any slippage.

The optical sensor puts out a low going high voltage on chuff and the Phoenix expects to have it's high pulled low.
The solution is to install a NPN transistor (Radio Shack 2N2222 works fine) with its collector connected to the Phoenix chuff input (pin 15) and its emitter connected to ground (pin 16 and the K-27 ground (J1-5). Now connect the base to the chuff signal from the optical sensor (J1-7). This inverts the signal so the Phoenix board responds.

The optical sensor has another problem though. It doesn't respond with valid chuffs until the track voltage gets up to 4 volts. The engine starts moving on about 1.5 volts and runs at prototypical speed at about 9 volts so the engine has no sound as it starts. To get around this, as a test,  I installed 5 diodes in series with the motor. These diodes drop about 4 volts and require the track voltage to be above 4 volts before the motor starts turning.

I cut the trace on the plug-in board between one of the chokes and connector pin.
If this becomes permanent, I will have to install 5 diodes in the opposite direction as well or see if I can find an appropriate Zener.

Now I have to take the cylinder area apart and see if the optical sensor board can be moved to get a correct sounding sequence.
If this can't be accomplished, I'll have to fall back on installing magnets.
 

·
Premium Member
Joined
·
1,704 Posts
Thanks for the info Dave. I'm hoping that the start voltage issue won't be a problem with DCC or battery?
 

·
Premium Member
Joined
·
2,093 Posts
Dave....Stan Ames said on chat that the optical chuff electronics shifts from 2 chuffs per rev to 4 chuffs per rev based on motor RPM and vice versa.
 

·
Premium Member
Joined
·
127 Posts
Discussion Starter · #4 ·
The optical sensors are located in the cylinders, one on each side and as the piston moves back and forth, the sensor beam is broken by a plastic extension of the piston giving a chuff.
I don't see any shift from 2 chuffs to 4. I get four chuffs at all speeds but they go chuff chuff chuff ....chuff. It sounds bad.
 

·
Premium Member
Joined
·
2,093 Posts
The chuff sensors are in the cylindars....the chuff electronics are in the boiler...and my understanding was that was where the 4 chuffs per rev changed to 2 chuffs at some speed and that this "fixed" chuff signal was what was sent to the pins on the socket. I'm just the messenger....but I do remember this discussion. The discussion involved opinions that with the socket board, NO electronics should be anywhere but on the plug in board...and that all that should come out of the engine were wires from a device to a pin on the socket. Was a good discussion.

I gotta say though....life would be a WHOLE LOT EASIER if folks would post, versus email, the solutions to these issues. Chat is a nice place to discuss these issues...and get straight skinny....but it's NOT a repository of modeling information. We've been told that the "solutions" will be posted once they are approved by the manufacturers of the electronic boards...WHEN is the issue.
 

·
Premium Member
Joined
·
2,537 Posts
Mike.
Here is the actual chuff circuit sent to me by Stan Ames.



Perhaps you could explain how that circuitry will automatically change from two chuffs to 4 chuffs per revolution.
There is a jumper from the output of D3 to D4 that (allegedly) selects whether or not it is 2 or 4 chuffs.

Thanks Dave.  You learn something every day.  The symbol looked like an envelope to me.
 

·
Premium Member
Joined
·
127 Posts
Discussion Starter · #8 ·
Tony, here's a shot of the board. There is a jumper JP1 that when removed (according to the documentation) will take you down to two chuffs per revolution. It comes jumpered as delivered.
 

·
Premium Member
Joined
·
2,537 Posts
OK Dave.

That is what the circuit seems to show.

Looks like today is a brain fade day for me.
Thanks for the help with the pic.
 

·
Premium Member
Joined
·
1,262 Posts
Looks like a well thought out design by Bachmann here (or whoever does their engineering for them)! /DesktopModules/NTForums/themes/mls/emoticons/crazy.gif
 

·
Premium Member
Joined
·
451 Posts
Anyone using a phoenix: If you are using a real time whistle, can you blow the whistle BEFORE you move the engine? I have heard a rumor that even in manual mode, nothing works until you hit the first chuff. That's not how realtime is supposed to work (and, even for automatic stuff, most engines whistle off before they start moving ... that's the point.)

So, anyone try this?

Matthew (OV)
 

·
Premium Member
Joined
·
2,537 Posts
Del, 
The chuff  would be a well thought out design if it actually worked without modifications and additions.

The sound chuffs do not start working until about 5 volts are on the track.  As the loco starts moving at 1.5 volts there is no chuff when the loco starts moving.
Whilst not a problem for battery R/C and DCC users, this is not good for regular DC users.  
Stan Ames/Roger Cutter overcame the problem by powering the Bachmann electronics with the Sierra battery.  That is likely to be a potential point of failure as the Sierra battery will also be powering the smoke unit and classification lamps.
I hate to think what might happen to the Sierra charge circuits if they cannot keep the battery charged under such a load.

Never mind that the chuff output is the wrong polarity for almost all sound systems and requires an inversion transistor.
Now we read that the chuff is incorrectly timed.
you may get: chuff, chuff, chuff, chuff, pause!!!  chuff, chuff, chuff, chuff, pause!!! etc. 

I have no qualms re the actual socket pcb design. I think Bachmann have got that almost right.
 

·
Premium Member
Joined
·
2,537 Posts
Matthew,
I could of course be wrong, but the way I read that circuit diagram of the chuff, indicates to me the chuff output is on all the time and the power is interrupted by the optos when the chuff timer actually does its thing.
It does not create the output voltage. It cuts it off.
That might mean the sound system has automatically cut itself off until reset by the next chuff signal stopping.
If that is so, it is a really stupid design.
 

·
Premium Member
Joined
·
127 Posts
Discussion Starter · #15 ·
Tony, your post of the optical sensor circuit explains why I saw 0.6 volts on the chuff output when the engine is stationary. The voltage out increases to around 2.8 volts when the sensor is interrupted.

The problem I am seeing with the chuff timing seems to be that one of the sensors is sitting a little too far forward in the cylinder. The sensor should be in the middle of the motion of the moving part. If it is too far forward, you don't get symmetric chuffs.
 

·
Registered
Joined
·
1,579 Posts
This is nutz.
I would not at all be surprised if Stanley designed this chuff circuit.

I have fixed mine, and am doing others on-site.
NO electronics, no polarity issues, works with all 2-wire chuff sound systems.

I am taking photos (after disassembling it for that purpose) and might just post them, or a linque to them.

I locked my #1 and #4 drivers from free-float, so I know where the thing is going to point.
Under load, the keyed axle bites the slots in the driver insulator, and the teeny little springs in no way can center it until you remove the load.
Once you take the play out (and it still trverses 1600 curves and switches, TIGHTER than the 8' minumum stated by the manufacturer), it appears they accidently placed a proper mounting pad right where it needs to be.

Oh, and last time I checked, I haven't seen anyplace in Sierra or Phoenix where the LENGTH of the chuff can be externally modulated while the unit is in operation, so any long/short chuff seems a tad odd.
Maybe I missed it.

Anyway, mine works with parts provided, no rat shack visits.

Anyway, if anyone wants it, I'll put the photos and details up.

TOC
 

·
Premium Member
Joined
·
127 Posts
Discussion Starter · #17 ·
Dave, I would like to see how you did it. If these sensors remain a problem, I'll be doing some other modifications.
 

·
A Steamed Elder
Joined
·
3,857 Posts
I can't believe that Bachmann "accidently" found a way to make a realistic sound of an engine "out of time"!:D I have a DVD of a K37 going up the Animas River that sounds identical to what you are getting. Dave, hopefully you can get this fixed.
 

·
Premium Member
Joined
·
2,537 Posts
Hi Dave.  
Again I could be wrong.
I read it differently.  
Perhaps, even though the loco is stationary, the chuff could still be high at 2.8 volts and drop to .6 of a volt when the sensor works.  It depends on exactly where the sensor is when stopped.  
It might be sensing the position and dropping the voltage down when stationary.
Then again maybe not.
The only way to rresolve this is for Bachmann to tell us exactly how it is supposed to work.
Unfortunately, trying to get answers from Stan Ames, is like trying to get blood out of stone.
 
1 - 20 of 41 Posts
Top