Frog Juicer Issues

[Maletrain]

It takes two juicers to make an auto-reverser.  It's most easily done with a Dual Frog Juicer, and there is conveniently a jumper on the board to allow this to be.  I have installed the jumper.  This Juicer then cannot be used for anything else.

This is what I think may be going on, i.e., the AR1 was slower.  I think I need to slow down the Dual Frog Juicer, but I haven't figured out how to do that yet.  I'm not sure I understand your discussion about two reversing sections.  I essentially have just cut one turnout, one side of the wye's triangle, and one tail track from the wye out as the reversing section.  There's no way to avoid a situation that the frogs are within (and abut!) the reversing section.

Thanks,
DFF

Re Dual Juicer question:  Doooh!!!

Regarding "I essentially have just cut one turnout, one side of the wye's triangle, and one tail track from the wye out as the reversing section."  I am pretty sure that is your problem.  The frog Frog Juicer sees a short to one side of the turnout or the other, and both it and the Frog Juicer for that section of track try to resolve it at the same time. 

There is a method for slowing one of the juicers, to let the other one win every time.  On the other hand, is there any way to avoid that conflict by setting both that turnout frog and the reversing section to run off the same juicer circuit?  In other words, if the turnout is set to go to the left on the wye, can the same frog juicer that makes the left leg of the wye match its adjoining track also set the frog polarity to go left on the turnout?  And, of course, it has to work the other way, too.  I know that some of the PSX-AR reversers can also throw a turnout, but I don't see anything in the Digitrax AR1 or Tam Valley literature on doing that, and I am not clear that is needed for your case.

I have not thought-through your situation, partly because it is not easy to do without a track diagram.  But it occurs to me that a wye is likely to have cars on the turnout and at one of the ends of the wye reversing section at the same time, unless that wye is very large.  So, there will probably be situations where a train is going through one leg with part of it on the boundary of the reversing section and another part of it on the frog of the turnout that joins the two legs of the wye.  In that case, the section may reverse in a manner that makes its phase inconsistent with the phase of the turnout frog, and then it only takes a single metal wheel to activate the juicer for both that frog and the reversing section,unless the frog and the reversing section can always be made to be in-phase with the end of the wye that the turnout is aligned to.  So, getting that alignment phasing would be my goal.

[davefoxx]

I'll try to get you a diagram this weekend.  I'll add, for now, when I first discovered the short, the locomotive was completely within the reversing section and came into contact with the frog.  That turnout is also entirely within the reversing section.

DFF

[RBrodzinsky]

If the turnout is entirely within the reversing section, than the frog-juicer for that frog needs to be fed from the dual-frog juicer.  You cannot feed directly from the main DCC bus. You probably survived with the AR-1 simply because it is a slow reverser.

Yes, if you can slow down your Dual-Juicer, you would probably be "ok", but think about the circuitry: if you are feeding the frog from the main DCC (through the Hex-Juicer), you are bringing the power in from a different circuit.

[Maletrain]

I don't think that feeding one Frog Juicer through another Frog Juicer is the answer, because both still see the same short circuit at the same time, and race each other to fix it by switching phase.  However, the idea of having the frog reverse when the reversing section reverses is the proper concept - but it would really best be done by using a set of contacts on the switch machine to power the frog, and feeding those contact from the reversing section reverser. 

Making the reversing section reverser slower than the frog reverser will probably work, because that allows the frog to match the reversible section without changing the reversible section, while allowing the reversible section to reverse to match the other track and then have the frog reverser match it before the reversing section reverser can reverse back.  That is what the AR1 was probably allowing to happen.

But, I have to ask, since you already have an AR1 and it worked, why are you replacing it with a Dual Frog Juicer in the first place?

[mu26aeh]

I don't think that feeding one Frog Juicer through another Frog Juicer is the answer, because both still see the same short circuit at the same time, and race each other to fix it by switching phase.  However, the idea of having the frog reverse when the reversing section reverses is the proper concept - but it would really best be done by using a set of contacts on the switch machine to power the frog, and feeding those contact from the reversing section reverser. 

Making the reversing section reverser slower than the frog reverser will probably work, because that allows the frog to match the reversible section without changing the reversible section, while allowing the reversible section to reverse to match the other track and then have the frog reverser match it before the reversing section reverser can reverse back.  That is what the AR1 was probably allowing to happen.

But, I have to ask, since you already have an AR1 and it worked, why are you replacing it with a Dual Frog Juicer in the first place?

See below:

https://www.therailwire.net/forum/index.php?topic=41814.msg625012#msg625012

[davefoxx]

If you need a diagram because this isn't clear, let me know.



In the picture above, the wye is obvious.  If you look close, you will find six (unfortunately clear) push pins next to the rails: two near the center and four at top right in the picture.  That's the rail gaps and show the reversing section between them, i.e., the leg of the wye at right.

All of the four turnouts near the reversing section have had the frog wire disconnected from the Hex Frog Juicer.  The locomotive will run around the wye.  I had one small stutter at the bottom gap, but I haven't been able to repeat it after multiple times around the wye in both directions.

You may note that two of the turnouts are outside the reversing section and two are inside.  Each one of them has caused a problem at the frog.  So, it works right now, but the frogs are all disconnected. 

DFF

[davefoxx]

So, here's the culprit.  You may notice the small wire at right on the board that has been cut.  After supper tonight, I installed that jumper, which delays the change in polarity current setpoint.  But, then the locomotive shorted at the first rail gap it came to, which was not a frog.  So, that didn't work.  Cutting the wire restored the proper timing current setpoint.  The locomotive runs around the wye, but, as noted in the previous post, the nearby frogs are all disconnected.

Yes, there are two wires going into the outputs, because I have two feeders in the reversing section on each rail.  These do not cause a short, or else the layout would shut down immediately.  They are not part of the problem.



DFF

[Maletrain]

The picture helps, but the configuration seems a little strange on the upper right corner, where the 2 turnouts are included in the reversing section.  It looks like one of those turnouts is half of a crossover, with the insulating gap between the two parallel tracks.  I can't see if the nearest of those tracks is a dead end, or if it connects elsewhere to the other visible layout tracks.  If it connects elsewhere, then it will need another set of isolation gaps to completely isolate the reversing section.

I don't understand why you set it up that way.  I would have put just one set of isolation gaps on the other side of the switch that is that end of the wye, rather than using 2 gaps to add 2 turnouts to the reversing section at that end.  I guess you have a reason for doing it the way you did it, so now I am wondering if that reason is part of your problem, whatever it is.

[davefoxx]

That track to the right of that crossover is a dead end, but I intend to lengthen it with a stub on a fold-down shelf.  It will still be a dead end, though.  I included those turnouts (the crossover) in the reversing section, because I was trying to lengthen the reversing section for the longest cut of cars to go through there.

What you suggested for gaps was what I originally had when I first built the wye.  The reversing section was too short.

Thanks,
DFF

[Maletrain]

OK, then I think it should work if you can make the Dual Frog Juicer reverse the whole section slower than the Juicers on the actual frogs.

I am not taking the time to look up the Dual Juicer instructions, so I am not sure what that jumper wire you added and then cut really did.  You posted that it "delays the current setpoint" and that cutting it "restored the proper current setpoint."  That sounds to me like you are confusing the current at which the circuit trips with the speed at which it trips.  You want to keep the current setpoint low so that some breaker upstream of the Dual Juicer doesn't trip before the Juicer  tries to switch phases (which sounds like what happened), but you still want the Dual Juicer to trip faster than any upstream circuit breaker but slower than the Juicers on the turnout frogs. 

Since the relatively slow AR1 worked, it seems that there is a timing for the Dual Frog Juicer that should work, too.  Th question is whether that particular timing (delay) value can actually be set into the Dual Juicer by jumpers or maybe CV programming.

[Maletrain]

A couple more questions about your set-up:

At the crossover that is split by the boundary of the reversing section, are the two frogs on those turnouts controlled together by one or separately by two of the Quad Juicer's outputs?

What type of turnout actuators do you use?  Could you use switching contacts on those to control the phases of those frogs, instead of the Juicers?

Have you thought about moving the crossover to the left (in the photo) to get it out of the reversing section?

[davefoxx]

OK, then I think it should work if you can make the Dual Frog Juicer reverse the whole section slower than the Juicers on the actual frogs.

I am not taking the time to look up the Dual Juicer instructions, so I am not sure what that jumper wire you added and then cut really did.

That wire jumper is to cause a delay, per the instruction manual.  But, it delayed too much, and my locomotive shorted at the first rail gaps it came across.  So, I cut the jumper to make sure that was the cause of the problem, and, bingo!, it was.

You posted that it "delays the current setpoint" and that cutting it "restored the proper current setpoint."  That sounds to me like you are confusing the current at which the circuit trips with the speed at which it trips.  You want to keep the current setpoint low so that some breaker upstream of the Dual Juicer doesn't trip before the Juicer  tries to switch phases (which sounds like what happened), but you still want the Dual Juicer to trip faster than any upstream circuit breaker but slower than the Juicers on the turnout frogs.

You are correct.  I should have written that it delays the speed in which it changes the polarity.  It's been a long week.  Sorry for any confusion.  I'll correct that post for clarity.  (But, I also installed two Tam Valley supplied jumpers to bump the current setpoint up from two amps to four amps on each individual juicer, since my DCC system is greater than five amps and I run consisted units with sound.)

Since the relatively slow AR1 worked, it seems that there is a timing for the Dual Frog Juicer that should work, too.  Th question is whether that particular timing (delay) value can actually be set into the Dual Juicer by jumpers or maybe CV programming.

Like I wrote above, I added the wire jumper unsuccessfully.  I am not aware of any CV programming as something that it possible on a juicer.

DFF

[peteski]

I don't know what the instructions state (I don't own any juicers), but as I see it the tripping current should be set as low as possible (not as high as possible).  Using high current seems counter-intuitive to me. 1 or 2 Amps should be plenty high. 4 Amps seems like trouble.

And yes, I understand what you are thinking when setting the current higher, but I think it is not the way to go.  There will likely be another solution. The track photo was not helpful (at least to me).  I think we need to see the diagram (track, all the gaps, and wiring) of the problem area to get to the bottom of this problem.

[DKS]

(I don't own any juicers)

Then you'll just have to get one...



Sorry for the OT post... I just couldn't resist. Carry on.

[peteski]

OK, I lied, I have one of  those stashed somewhere.

But it is DC-only.