DCC Frog Juicers and Electrofrog Points

[Despatcher]

I have used electro frog points for years, however I have come up with a problem on my New Gauge O, built during Lockdowns railway, The layout is about 24' long and 2' wide. It is a steam diesel stabling yard. It has 9 points. I have a Z21 controller that works fine. All the points are DCC Concepts Cobalt Digital IP points. They are all controlled by the Z21 without any fuss. I have used O Gauge Peco Electrofrog points. I have wired the blade rails to the running rails. I have cut the two fine wires near the frog, I have insulating rail joiners on the two inside rails leading from the frog. The long thin wire from the frog is soldered to some 16/0.2 dropper wire and attached to 2 Hex frog juicers. Four to one and five to another. All the locos potter up and down the layout. No short circuits, smooth slow speed running. This was also the system I used with the layout this Gauge O has , which was a modern image OO gauge diesel depot and storage facility with over 35 electro frogs wired exactly the same. All my locos have high capacity stay alives. So ... here is why I am scratching my head.

 

I have a small Class 03 Gauge O diesel shunter. It can be stopped exactly on the frog of an O gauge point. However, if I leave it there for any time, it won't move again. The stay alive runs out of power and I have to push it off the frog where it soon charges up and carries on pooling up and down my railway. I was puzzled by this. My big OO Gauge Class 66s, 60s, 67s, 33s etc were not big enough to stop on an OO Gauge frog and most of my Gauge O locos are similar. Although I have just bought w wonderful Minerva 0-4-0 Peckett that certainly will.

 

So, not really being into electrics, I got out my multimeter that I rarely used as I have not much idea how it works to be honest. I found that all my track was fine with a decent current every where. Which is what I expect as every piece of track, points and all has two dropper wires going to the track bus. However, I then discovered that the "live" frogs only had about somewhere between 1.6 and 3.4 of whatever I was measuring by putting one terminal on the outs running rail of the point and the other on a corresponding part of the frog. I was expecting the frog to have the same current as the main track ie 11.6 of what I was measuring. So, I scratched my head some more. I checked videos on wiring electro frogs. No, I've done this correctly. I tried using the frog socket in the Cobalt Digital IP points. Same result. I tried disconnecting the wire from the frog ie not plugging it into the hex juicer or the cobalt IP digital and because of the stay alives the locos ran across what was a dead frog.

 

I am assuming that if my locos did not have stay alives they would all stop on every point as at the moment the stay alives are doing their job? So my questions out there for help please are:

 

1. Should the frogs have the same power as the rest of the point and the running rails? I am assuming so!

2. What have I done wrong.

3. Is there a remedy without digging all my points up, which I really, really, really don't want to do? It has taken me 14 months to get this far.

 

My layout will actually run just fine as long as I am careful where I stop the small wheel base locos. My Class 25 and Class 35 and the wheelbase Prairie will never have a problem. I would like to learn and understand where I went wrong though.

 

Thanks

[KingEdwardII]

2 hours ago, Despatcher said:

I then discovered that the "live" frogs only had about somewhere between 1.6 and 3.4 of whatever I was measuring

Despatcher,

 

Can you explain these measurements a little more, please. I don't really follow what you measured and what you observed, but it sounds important.

 

Yours,  Mike.

[ITG]

Not quite the same issue, and I certainly would class myself as still learning about DCC, but I once tried to measure the output of frog juicers into frogs with little success. I was advised that you cannot really reliably measure DCC current with a conventional meter, and that measuring continuity is a far more reliable test of what is happening through points.

But don’t Cobalt Digital IP point motors have frog switching capability without your Hex devices?

[Despatcher]

Many thanks for replying. OK. With my multimeter I measured what I believe were the track amps. Everywhere on the layout, putting a probe on one track and one on the other I get a reading of about 11.6, which I assume is amps. When I put a probe on the outside running rail of a point and one a running rail of the frog, I got a reading of around 1.6 to 3.6. I hope you can see this in the photos. There is also a photo of the hex frog juicer. 

[Despatcher]

The frog juicer

[Despatcher]

Looking at the last photo of the frog juicer, I realise that the yellow and blue wires coming into the right hand side of the Frog Juicer are from a Z21 extension. I am wondering that if I power it from the main track bus which is providing 11.6 (amps?) directly from the Z21 it will cure the problem. I will play in the morning this problem has scrambled my brain enough for one day. It would be good if it was that simple! And it would make me pretty stoopid. But we all live and learn!

[WIMorrison]

what you show on your multimeter is 11.6 'volts'. It is not really 11.6v as DCC is not a sine wave therefore your multimeter cannot accurately measure what the voltage is, however it will probably be somewhere around  and actual 14-16v .

[Junctionmad]

We use lots of tam valley juicers on our O gauge club layout. You should see the same voltage measurement as on plain track on the frog ( after a loco has run through the point ) 

 

 

[brossard]

It looks like a short to me.  I had a similar problem a while back, but it cleared itself for some inexplicable reason.  One reason for it could be that the underside jumpers are still touching despite being cut.  I took my turnout up and made absolutely sure there were no errant wires.  Otherwise, Sod's Law says that the fault will return at a show.

 

As said above, if using Cobalts (mine are Tortoise), I don't see why you would need a juicer but I'm sure you have your reasons.

 

BTW, I can't tell from the photo, but the joiners coming off the Vee need to insulated.

 

John

[KingEdwardII]

You are measuring AC voltage with your multimeter. This is OK for showing you what is going on.

 

This is showing 11.6V for the voltage between the rails on your regular track. This is not an accurate measurement of the actual voltage, since the DCC signal is a high frequency one which is a variable square wave rather than the 50/60Hz sine wave of household AC power for which multimeters are designed.

 

However, 11.6V is not an untypical reading for a multimeter measuring the AC voltage of a DCC system.

 

The fact that the same equipment gives a reading of only 3.5V between the frog and the opposing rail indicates a problem with the power feed to the frog. I would expect to get the same 11.6V reading you observe on the regular track.

 

3.5V is quite low and I would expect this to cause problems for a loco traversing the turnout once its wheels get to the frog.

 

So, why is the voltage so low? I am not familiar with the Tam Valley Frog Juicers and I can only work on the basis of the manual I can read on the Web:

 

http://tamvalleydepot.com/images/Hex_Frog_Juicer_Manual_v2.1.pdf

 

Things worth checking:

- what's the voltage on the inputs to the Frog Juicer (terminals on the right with the black & yellow wires in your picture)

- what's the voltage between each of those inputs and the output terminals on the Frog Juicer (the 6 way connector block on the bottom in your picture)

 

Basically, I'd try and work out where the voltage loss is occurring. It could be caused by the wiring and/or the connectors anywhere between the DCC track bus and the frogs.

 

I use MTB MP1 point motors on my layout and these have a built-in switch that is used to supply power to the turnout frog - I don't use a frog juicer. However, I expect the frog voltage to be the same as the voltage for regular track in the direction the turnout is switched - and that is what I get, unless there is a problem. I typically only get problems when first installing a new turnout, for example if I wire the connections to the switch the wrong way round.

 

Yours, Mike.

[Despatcher]

Many thanks for the replies so far. All useful stuff.

 

The joiners coming off the "eve" are insulated.

 

I'm glad that you have confirmed that the voltage in the frog or "vee" should be the same.

 

I am fairly certain the underside jumpers are not still touching as all 9 points are the same. That would be very careless workmanship!! I also cut them with a thin slitting disc on a Dremel power drill and made sure there were no stray wires and a gap by using a small blade in the gap. Also, f I disconnect the wire from the frog (vee) there is no short and my locos will cross the point, and the now dead frog (vee), on power stored by the on board loco stay alive.

 

I will connect the Frog Juicers to the track bus from the Z21 tomorrow and not the bus from the Roco booster. It would seem logical that that could be the possible cause of the problem. Incidentally, I only used the Tam Valley Frog Juicers with the Cobalts as I knew no better. We all have to start somewhere.

 

The power for the Cobalts is from the Roco Booster bus wire and not track bus from the Z21 too. I might need to change that too and totally do away with the Booster bus wire.

 

It is an old Booster and not compatible with the Z21. On my last layout, I was going to have "power districts" but when the locos ran across the insulated joiners from a "district" powered by the Z21 to a "district" powered by a booster, there was an instant short. I contacted Roco who told me the boosters were not compatible with the Z21. I assumed they would be ok to use for an accessory bus not realising they may have this effect on the power in the frogs. Tomorrow will be interesting.

[Hamburger]

Did I understand correctly that you are using 2 separate power sources?  One for track, one for the frogs?
Is it possible that one of them is set to around 11.6 volts and the other is set to around 15.1 volts? 

AND that the frog polarity is wrong? 

If yes you should connect both frog and track to the same bus and change frog polarity.

[Dominion]

The frog juicer relies on the loco bridging a piece of rail with a definitive polarity to the frog. When a longer wheelbase crosses the gap, it “knows” which way to set the frog polarity. My hunch is that when your short wheelbase loco is purely on top of the frog, the frog juicer no longer has the absolute reference to the known polarity rail it is supposed to match, and “forgets” what state it is supposed to be in (red or green on the leds)
 

If this is indeed what is happening you could confirm by looking at the leds on the juicer. Watch which light switches from red to green and back for the turnout you want to test. Drive the short loco over it without stopping and see the colour of the light.
now drive it on to the frog, stop. And if it won’t restart, see if that specific led has changed colour.  
 

They should all stay the same as their last state normally until a loco crosses them in a different direction, but can change and be different if you power off and on again. I wonder if the stay alive also has some effect too, as the juicer relies on the dcc signal, and perhaps it is more likely to change state and get confused while not anchored to a known rail with a capacitor load in the loco. 
If the light does change while the short loco is stationary on the frog, then at least you know the rest works and it is not really indicating a fault, just a quirk !

 

I use Tam valley frog juicers in some spots on my layout, and find them very useful, but where I have Cobalts I use them to drive the frog polarity. As you are using a different power source to track power to drive the motors, don’t use the “frog” outlet on the cobalt digital ip.  Instead use the entirely electrically separate switch built in to the cobalt. Put your track power into the 2 ports for the built in separate switch, and connect the frog wire to the one marked common. Different cobalt motors have the ports in different sequences, but the 3 ports I mean will be next to each other and one should be marked C. Look in the DCC instructions for your type of motor if in doubt.

 

Hope this helps, Tom

[Nigelcliffe]

9 hours ago, Despatcher said:

 

I am fairly certain the underside jumpers are not still touching as all 9 points are the same. ............

 

I will connect the Frog Juicers to the track bus from the Z21 tomorrow and not the bus from the Roco booster. It would seem logical that that could be the possible cause of the problem. Incidentally, I only used the Tam Valley Frog Juicers with the Cobalts as I knew no better. We all have to start somewhere.

 

The power for the Cobalts is from the Roco Booster bus wire and not track bus from the Z21 too. I might need to change that too and totally do away with the Booster bus wire.

 

It is an old Booster and not compatible with the Z21. On my last layout, I was going to have "power districts" but when the locos ran across the insulated joiners from a "district" powered by the Z21 to a "district" powered by a booster, there was an instant short. I contacted Roco who told me the boosters were not compatible with the Z21. I assumed they would be ok to use for an accessory bus not realising they may have this effect on the power in the frogs. Tomorrow will be interesting.

 

OK, quite a lot of useful detail in the above posting.

 

First - possible cross-connections of the frog wires, jumpers, etc..   If you are sure nothing can touch, then fine.  But its something that should be checked.  Its quite hard to check it with all the peripheral stuff (Juicers, Cobalts, etc) also connected.   Ballast can also be a conductor, if it is glued with PVA, though that usually goes away when it is thoroughly dry (PVA is an acid, so conducts electricity).  

 

 

If you have your Frog Juicers on the Roco-booster, and the main track on your Z21, then you've done what you did before and that didn't work the last time (connected the Roco-booster and Z21 to two power districts, and get problems when a loco crosses from one to the other).   

Your voltage measurement from frog to running rail (3.5v) is only telling you "something wrong" - the frog has power from the Roco-booster, the running rail the Z21, and the voltage is just "something via the complexity of connecting the two together".  
As the Z21 and Roco-booster are not compatible for adjacent power districts, then one of them should NEVER be connected to rails via anything. Hopefully this hasn't damaged any of the Juicers (they're pretty robust as electronics goes), but I'd not be 100% certain of no damage.  

 

 

The Cobalts themselves are fine on a secondary booster (Roco-booster).  But only so long as you don't use the "frog" output of the Cobalts.  Otherwise, you're connecting the Roco-booster and Z21 to the track again.   
As currently setup (Roco-booster and Z21) and if wanting to use the switches on the Cobalts, then use the second switch (with three terminals):  two terminals take the track bus (Z21) and one out to the frog.   

I'd be using the Cobalt switches and not using Frog-Juicers;  there being no reason to use Frog-Juicers if a turnout motor has reliable frog switching.   

 

 

As a general principle, the second booster (as currently arranged) isn't adding much to your layout's power handling or reliability.  A handful of Cobalt motors doesn't draw much current.   So, it may be simpler all round to remove the Roco-booster from the system.   

 

 

- Nigel 

 

 

[Despatcher]

Many, many thanks for all of that. I am going to try powering one frog juicer from the track bus first, as that is an easy fix. . If that works it will mean just changing the power connections to the two Tam Valley Frog Juicers. Four wires. If that does work and solve the problem I will take the bus circuit from the Roco power booster out. That will mean changing the power feeds to to the 9 Cobalts to the track bus. Fingers crossed!

[Despatcher]

Thanks Nigel, I think that is what I slowly worked out too. All started by looking at the photo of the Frog Juicer that I posted and realising it is powered by blue and yellow wires (the bus circuit from the Roco Booster) and not red and black wires (the track bus circuit from the Z21). I now realise that when a loco was crossing the points and got to the frog, one set of wheels was collecting power from the Z21 and the other collecting power from the Roco Booster via the Frog juicer. 

 

Interestingly I had my previous OO gauge layout wired up like that for about 7 seven years without noticing a fault. It was only a pure chance that I stopped my O Gauge Class 03 with all 6 wheels on the frog that I became puzzled. The chances of that happening are slim. I should have bought a lottery ticket.

 

Many thanks. It is amazing sometimes how as experienced (ish) modellers we cannot see the absolutely obvious! And thank goodness fro RM Web!

[Pete the Elaner]

13 hours ago, ITG said:

Not quite the same issue, and I certainly would class myself as still learning about DCC, but I once tried to measure the output of frog juicers into frogs with little success. I was advised that you cannot really reliably measure DCC current with a conventional meter, and that measuring continuity is a far more reliable test of what is happening through points.

But don’t Cobalt Digital IP point motors have frog switching capability without your Hex devices?

 

A slight digression but it is true that you cannot read accurate DCC voltages with a normal multimeter. The meter will be configured for sine waves, which you get from mains or direct from a transformer.

DCC uses square waves. Because the wave pattern is different, the meter will mis-read it. creates different results, some prefer not to call this AC.

It should be consistent around the layout though, so a low reading in 1 section will indicate an issue. From a troubleshooting point of view, this is more important than the value itself.

 

I saw a mention of voltage & current.

Voltage is the electrical pressure across 1 or more components. Current (amps) is the flow through it. This is a similar concept similar to water pressure difference across the ends of a hose hose & the flow through it.

An ammeter needs to be inserted into the circuit (unless you have a fancy, expensive inductive one). It can be useful to see how much current is flowing but it can be a nuisance to set the meter up to read it.

[Dominion]

The Hex frog juicer should definitely be fed from the same bus that feeds the power to the track. Good luck with the switch :-)

[Despatcher]

Thanks to everyone for your advice. As ever very much appreciated. However obvious that was I could not see it initially. Stupid or what?? 

 

Yep, altering the power feed to the Hex frog juicer solved the problem. Here it is connected to the red and black track bus from the Z21. The disconnected blue and yellow feeds from the power booster can still be seen. I have changed the feeds to both frog juicers now so the problem has totally gone.

 

There is also a photo of the booster sitting next to the Z21. It is going in the bin!!!!!!

 

I may "wire out" the bus wire from the booster now by simply disconnecting it from the booster and attaching it to the end of the track bus from the Z21. They are about 30cm apart. I will then have the option to wire the Cobalts to the frog wires.

 

Problem solved. Job done. Again, I could not have done this without RM Web. Thanks.

[Pete the Elaner]

24 minutes ago, Despatcher said:

Thanks to everyone for your advice. As ever very much appreciated. However obvious that was I could not see it initially. Stupid or what?? 

 

 

Not silly at all.

I am sure all of us have got involved with a problem & not noticed the obvious.

 

It helped that you explained it carefully. Explaining the problem to somebody else can sometimes help you to realise what is going wrong without them having to reply.

[Nigelcliffe]

1 hour ago, Despatcher said:

Thanks to everyone for your advice. As ever very much appreciated. However obvious that was I could not see it initially. Stupid or what?? 

 

Yep, altering the power feed to the Hex frog juicer solved the problem. .......

 

I may "wire out" the bus wire from the booster now by simply disconnecting it from the booster and attaching it to the end of the track bus from the Z21. They are about 30cm apart. I will then have the option to wire the Cobalts to the frog wires.

 

Problem solved. Job done. Again, I could not have done this without RM Web. Thanks.

 

 

 

As Pete said, not silly or stupid.  Sometimes someone else looking will spot the obvious mistakes in stuff I've done.

 

If you do decide to keep the yellow/blue bus for the cobalts (probably the simplest wiring option), then I suggest wiring them into an easy-to-disconnect switch or connector block, before the combined wires end up at the Z21.   Then its quick to disconnect one or other bus should it be necessary in future.   
If you do use the "frog" outputs of the Cobalts, make sure you put some clear labels on your wiring (eg. a taped tag around the blue-yellow where it connects near the Z21) that there is a cross-connection from blue-yellow to red-black at the frog outputs.   Otherwise, years away, same mistake can happen!

 

If doing lots of re-wiring underneath, consider investing in a "bootlace ferrule" crimp kit (mine was well under £20 from Amazon).  Makes very tidy ends of wires for putting into screw connector blocks.  Once used, I've wondered why I didn't have one years ago. 

 

 

- Nigel

[WIMorrison]

There is an issue with using the frog output from the Cobalt if you are powering the Cobalts from a separate bus, which is effect an accessory bus.

 

When you get a short circuit due to a train overring a turnout set against it then you will short the main and the accessory bus therefore losing the benefits from having the separate bus to power the Cobalts.

 

Not a problem if this is what you expect, however if you hoped to be able to operate the turnouts when there is a track short then you will have an issue. It is better to use the SPDT switch on the cobalt fed with the track bus to switch the frog then the issue will not happen  

[Despatcher]

Thank you all for the support, the patience, the time, the kind words and the knowledge. Hopefully all sorted and done now. Phew!!