Electrofrog point misbehaving

[NormanQ4]

Here is the latest update on this sad saga.

I ordered the DCC Specialities PSX-2 circuit breakers from Traintek in the US that was suggested as a solution by DCC Concepts which took a while to arrive.

I isolated the frogs from the main track power and connected one CB to them with the other connected to the main track bus.

This seemed to solve the problem of the frogs having to be cleared of locos and lit carriages when shorting occurred.

Unfortunately it created a whole new problem. When a loco came to my diamond crossings it tripped the circuit breaker which automatically reset after 2-3 seconds resulting in the loco moving forward about 20mm when that little bit of power came back to the track.

The loco kept moving forward like this until all of its pick up wheels crossed the IRJ of the next point. A multi meter shows no fault with the wiring.

This also happened to all the locos on the track that were moving at the time resulting at one stage with 5 locos bunny hopping around the track until the offending loco was clear of the diamond crossing and adjoining track piece.

This even happened when the locos were travelling at cruising speed.

Prior to the CB being connected intermittent shorts occurred when the tyres of some locos bridged across the very narrow (probably less that half a millimetre) insulating plastic between the two crossover rails on two of the four diamond crossings on my layout. I hadn’t noticed this obvious fault when I fixed the crossings and ballasted them otherwise I would have contacted Peco. The insulating plastic of the other two crossings is around one millimetre wide. 

I have now disconnected the CB to the track but left the other one connected to the frogs. Shorts then only occurred intermittently and only at the places where the plastic was very narrow. Even though I had created a groove between the rails with a Dremel type tool.

When a short occurred (only when a loco was going at shunting speed) I had to move any locos or lit carriages from all the frogs as before to permit a restart and sometimes I did not get the “short “ warning and then had to reset the decoder with the little red button before restart would be allowed.

I have been getting so frustrated with the continual problems and have walked away from the layout for days at a time in case I lost the will to live.

I’ve spent a couple of years building this layout but have recently felt like permanently walking away from it, but unfortunately I couldn’t even sell it in this condition.

I feel that my only option now is to remove the diamond crossings completely and hope for the best.

I’m sure I’ve forgotten to mention something but that will do for now. I’d hoped to be reporting back with good news but unfortunately that was not to be.

[walkingthedog]

Should have used DC. No problem with diamond crossings.

[Chrissaf]

Norman, this was my solution to your issue using Peco diamond crossings. It works flawlessly like a dream. I'm using the TamValley DFJ003 (in RLM mode), but any RLM will work in this configuration. After cutting the crossing rails to electrically isolate the frog area, I joined the two frog rails together to form one single electrically conductive frog area at each end of the diamond as indicated by E & F on the drawing.

.

/media/tinymce_upload/75889fc5031a20deb7ebd4e3eaa02859.jpg

.

[Fishmanoz]

Good solution Chris. What you have done of course is adapt best practice DCC frog switching on points for use on crossings. In  either case, the isolation of the area around the frog between the IRJs and the rail cuts means that opposing polarity/phase only appears where the rails are well separated (unlike at the frog itself) and so there is no chance of a short from wheels bridging the gap. 

[NormanQ4]

Hi Chris and Fishy

Thanks for your responses.

Is adopting this method likely to be the end of my troubles, or could it create another problem elsewhere?

Am I right in believing that I would need a Duo and a Hex juicer to serve my four crossings?

Should I still use the CBs that were recently connected to the frogs and track bus?

I have to say that I am reluctant to throw more money and effort at this, just to take another step backwards, given my recent experiences.

[SteveM6]

Norma, before you start hacking your track around, go back to the instructions for the PSX2. You will find that the trip values are adjustable and I suspect you will have to adjust them to stop your issue.

Follow the instructions exactly though, they are not exactly clear.

[NormanQ4]

Hi Steve

Firstly thanks for changing my gender, and I never felt a thing.

I contacted DCC Specialties and they said there was no way to adjust trip values on a computer screen system such as Hornby Railmaster/Elink. Like me they checked out the info on the Railmaster guide and found that there is no provision for programming on main.

The default value of the PSX is 3.81 amps which is not much lower than the Hornby 4 amp controller.

Using the jumpers system of the PSX the next step up is 6.35 amps.

I don't pretend to know much about things electrical, but this doesn't seem like it would solve my problem.

[SteveM6]

Sorry Norman, didn’t press the second n hard enough. 

I had a similar issue when I first installed mine but the adjustment allowed everything to work just fine. If you can, give it a try.

[Chrissaf]

or could it create another problem elsewhere?

.

Unlikely, the solution is very local to the diamond crossing itself and doesn't impact on the track layout surronding it. I have four diamond crossings with this solution distributed across my layout and as stated it completely cured my short circuit on the diamond frogs issues. However I would only connect the frogs to the RLM unit (see reply part B below), the frogs need to be electrically isolated from the rest of the layout. So you would need to remove any secondary frog connections made to other switching products.

.

Am I right in believing that I would need a Duo and a Hex juicer to serve my four crossings?

.

The solution ideally requires a RLM (Reverse Loop Module) to function efficiently. This is because the two frogs on the crossing are always at opposite phase (polarity). If using the TamValley products the Duo has a switchable RLM feature option on it. The TamValley Hex frog juicer does not have an RLM option.

.

Technically, the Hex juicer would work as each of the two frogs on the diamond crossing are in effect treated as two individual frogs (the same as any point frog). But the benefit of using an RLM function is that both frogs get switched to their correct phase (polarity) simultaneously, putting less stress load on the DCC power supply.

.

An RLM with a relatively low current detection level is preferred, again to minimise stress on the power supply. My TamValley DFJ003 (Duo) switches at 1.7 amps.

.

As previously said, any RLM will do the job. Hatton's have Hornby RLM R8238 in stock and special clearance discount price of £12 each, plus postage to Australia of course. If you use the R8238, put the switch position in the 'Select' position. This makes the switching trigger current about 0.8 amps. Which should be more than adequate for this application.

 .

PS - Changing your current DCC Specialties product switching current to 6.35 amps definitely won't work.

[NormanQ4]

Hi Chris

I'm sorry I missed the point that the gist of your post was towards the RLM not the Frog Juicer.

I have to confess that I know absolutely nothing about RLMs and since I don't have any Wyes on my layout I haven't bothered reading any posts regarding them in the past.

Would I need one for each crossing (4) or one for each end of each crossing (8)?

Am I right in believing that the track power supplying the the RLMs should still be protected by the PSX CB?

[Chrissaf]

Would I need one for each crossing (4) or one for each end of each crossing (8)?

.

An RLM has a DCC input pair and an output DCC pair (protected side). The function of the RLM is to monitor the output DCC pair for a rise in current caused by a short circuit (each leg is monitored individually and doesn't need to be a rise in current between the pair due to a short). Once the trigger current is reached, the RLM reverses the phase (polarity) of the output to remove the short circuit condition and the current falls down to the nominal background level again. As this switching occurs in micro-seconds (when using all solid state RLMs) the power is never actually broken as far as locos on the track is concerned. My sound locos do not even exhibit a glitch or click in the sound when they trigger my RLMs.

.

If you refer back to my diagram in my earlier post. You will see that one single RLM is deployed per diamond crossing. The output pair has effectively an A and a B leg. Each leg of the DCC output pair connects to an individual frog on the crossing.

.

Am I right in believing that the track power supplying the RLMs should still be protected by the PSX CB?

.

I would replace the word "should" with the word "could" in your sentence. My understanding of the PSX CB (CB as in Circuit Breaker) is that it is like having a very sophisticated electronic fuse that breaks power and then restores it again. This being the reason why ALL your locos stopped and then went through a series of 'bunny hops' (power connecting & breaking in pulses) until the condition creating the short was removed i.e the offending loco passed over the crossing frog.

.

If this understanding is correct, then I see no technical reason why the DCC input of the RLM could not be connected to a PSX CB protected supply. Perhaps overkill, but not a technical issue.

.

IMPORTANT CAVEAT. It would be very important though for the track leading into and out of the diamond crossing to be powered from the same PSX CB protected supply as the 'DCC input' of the RLM. If the PSX CB output is currently connected directly to the frogs, this connection needs to be removed. The frogs MUST only connect to the DCC Output of the RLM.

.

Don't forget that in my earlier drawing I showed that the rails on the diamond crossing were cut (I used a Dremel with a very thin cutting disc) to create electrical isolation of the frogs. This isolation is essential, the solution will not work without it. The isolated rails of the frog were then also electrically bonded together with a wire strap to create a single electrical frog area. As indicated by the grey coloured rails in my previously posted drawing.

.

I feel that you may have been given bad advice to use the PSX CB product for your diamond crossing frog shorting issue. I can only assume that the person(s) giving that advice didn't fully understand the technical crux of the issue. I developed my solution as a result of having the exact same issue you documented on my own PECO diamonds.

[NormanQ4]

I feel that you may have been given bad advice to use the PSX CB product for your diamond crossing frog shorting issue

 

Hi Chris

DCC Concepts had recommended the PSX CB to deal with the aforementioned problem I was having with frog shorts on the points.

Unfortunately as a consequence of fixing the frog shorts the diamond crossing shorts went from an almost unnoticeable blip to the wild bunny hopping, I presume due to the ultra sensitivity of the unit.

With this unit disconnected from the main track the frog short problem has returned to the points so I will have to reconnect it.

That is why I was asking about protecting the RLMs with the PSX CB.

I suppose what I was saying is would the PSX CB create a conflict with the RLMs and if so should I provide a separate bus for the RLMs independent of the PSX CB.

 

 

[Chrissaf]

I suppose what I was saying is would the PSX CB create a conflict with the RLMs and if so should I provide a separate bus for the RLMs independent of the PSX CB.

.

If the PSX CB is protecting the point frogs, then I would assume that the whole layout power is in effect being fed via the PSX CB unit. In which case do not put the the RLM on a separate bus that is independent of the PSX CB.

.

In my previous reply I wrote:

"IMPORTANT CAVEAT. It would be very important though for the track leading into and out of the diamond crossing to be powered from the same PSX CB protected supply as the 'DCC input' of the RLM."

[NormanQ4]

If the PSX CB is protecting the point frogs, then I would assume that the whole layout power is in effect being fed via the PSX CB unit.

 

In my post just prior to your drawing, I incorrectly wrote that one of the CBs of the PSX-2 "protects all of the point frogs".  It actually protects all of the DCC Concepts accessory decoders which in turn power all of the frogs.

The other CB protects the track.

Accessory decoders on one bus and  track on another.

I've ordered four of the R8238 RLMs from Hattons and the VAT saving almost covered the cost of the postage.

[Chrissaf]

Put the DCC Input of the R8238s on the track CB. Remember to use the R8238 'Select' switch setting too.

[NormanQ4]

Thanks Chris, will do.

[96RAF]

Norman

You may like to read this just out of general interest, nothing particular to point out.

http://www.halton96th.org.uk/page19.html

There is also a writeup on there for the R8239 Booster unit which coincidentally acts as an RLM. Also a good bargain from Hattons if they still have any, as it includes a 4-amp PSU and they were only £25 when I got mine - less than the discounted price of the PSU Alone.

Rob

[Chrissaf]

R8239 now sold out at Hatton's, probably because people realised it was cheap way to get the 4AMP PSU for nearly half price.

[NormanQ4]

Hi Chris

I'm a little bit confused by this guidance.

Remember to use the R8238 'Select' switch setting too.

And this lifted from the RAF96 link.

The unit can be set for sensitivity by way of the Select/Elite switch - in effect this switch is selecting whether to be aware of a 1-Amp (Select) or 4-Amp (Elite) power supply unit and react accordingly.

I'm using a 4-Amp PSU.

Is your advice to use the "Select" a typo or is there a special reason for selecting that over the Elite.

[NormanQ4]

Thanks Chris for deciphering Hornby logic.

I was confused that two normally reliable advice givers were giving what on the face of it looked like conflicting advice.

[Chrissaf]

If you read Rob's comment really carefully, he is in essence saying the same as me (as I said, perhaps poorly phrased). I just go that extra mile to put the detail in my replies with the aim of removing all 'elements of doubt'. This results with my replies invariably being a page of text vs one liners. Despite the criticism I receive for my long replies, there is a need for the detail sometimes.

[NormanQ4]

Absolutely 😆

[96RAF]

I know what I mean...

Rob

[Chrissaf]

No Norman, not a typo.

.

The R8238 switch naming by Hornby is a bit misleading. The intention by Hornby was that the 'Select' position would be used when you were using a 'Select' and the 'Elite' position would be used when using the 'Elite'. This was before the eLink came along.

.

Rob's statement is perhaps lacking in clarity (perhaps could have been phrased with a little more detail), in that the 'Select' typically uses the 1 Amp power supply and the 'Elite' uses the 4 Amp supply. However, the Select (and the eLink) can both be upgraded to the 4 Amp supply. Hence Rob's comment referring to the power supplies, because it would be useless to put the R8238 in the 'Elite' switch position if you were only using the 1 Amp supply.

.

In reality, the Select / Elite switch position defines the RLM trigger current that will invoke the phase (polarity) switching. The 'Select' position implements a detection trigger current around 0.8 Amps give or take a bit and the 'Elite' about 3 Amps. By using the 'Select' position you are doing two things. Firstly you are making the switching decision more sensitive, secondly you are putting less electronic strain on the DCC Controller, Power Supply & R8238 components due to the lower short circuit currents involved.

.

Finally, because the R8238 is only powering the diamond crossing frogs and NOT a whole section of track. It only needs to provide enough power for a single loco as it traverses the diamond crossing. It is unlikely that a loco will need to draw more than 0.8 Amps, if you did happen to have a really old Triang motor type loco with weak magnets that was drawing more than 0.8 Amps, then just put the R8238 in the 'Elite' switch position. But personally, I feel that the 'Select' position is preferred if you can use it for the reasons stated in the paragraph above.

.

If you go back to page 4 and review my post made at 12:27 on the 20th, you will find the text reproduced below. Note the highlighted sentences. The reply above just expands, in more detail, the highlighted text.

.

/media/tinymce_upload/00dff981166b3633e58c1d9e2a312f00.jpg

.

[Fishmanoz]

I know Rob knows what he means too. 

 

And as for Chris going the extra mile - what he means is that in general his text laid end to end will be a mile long, all relevant to the subject.