Electrofrog point misbehaving

[NormanQ4]

 

My layout has 19 electrofrog points.  Eighteen that work without issue and one that is a bit rogue.

After having recent past issues using SEEP point motor switches to power the frogs I took the forum advice and installed additional DCD-ADS8sx / 2sx decoders instead, so all point frogs are now powered from the decoders.

This point is actually powered from a DCD-ADS8fx.

On first firing up the layout, when a loco doing a right turn (RM speak (actually straight ahead)) meets the frog of this offending point it stops and makes quite a loud buzzing noise although not actually triggering a short.

Sometimes at this time any other locos moving on the layout (has been up to four) reduce in speed from a cruising speed to a shunting speed, and any that are at a shunting speed stop completely.

On pushing the offending loco off the frog the other locos return to their previous speed.

The stalled loco issue can be temporarily resolved by doing one of two things.

1.   Flick the point to the left then back again to the right.

2.   Operating the right turn button again, even though the point is already set in the right turn position.

Using either of these options a loco can then cross this point freely without incident indefinitely.

That is, until there is a short on another part of the layout causing a power interruption and we’re back to square one.

The voltage to the remainder of the track stays constant.

From memory, I think this could also have been occurring when the frog was powered from the point motor switch.

I’ve tried reversing the polarity of the frog but that definitely causes a short when a loco touches the frog.

I could be wrong but it seems to me that the issue is not so much with the power supply to the point but with the messaging from the decoder to the point.

The only redeeming feature of this situation is that the point is located right at my control desk so at least I don’t have to go walkabout to move the loco.

Awaiting any words of wisdom.

This problem has been on the backburner while I was dealing with my FS issue.

[96RAF]

Hopefully I,have read and understood the problem correctly.

 

I presume the offending loco is not just the same one and could be any loco, else I would suspect B2B wheel settings for that one loco.

 

My initial thought was the point is backing off a tad and providing a high resistance across the frog, then the finger powered or repeat selection is snugging it up again and remaking the switch contacts - but you said you were using the Seep built in switch then went to the additional acc decoder switching. I presume you have by-passed the Seep switch - but as the problem has reappeared we can probably rule out the actual frog switching as the fault area i.e. Different switches in a different place.

 

The locos slowing is a sign of voltage drop - not enough to trigger a short cutout and not enough to drop the loco decoders below their operating voltage. Could you monitor the track voltage (AC volts will do) as you induce the fault scenario just for comparison.

 

My next thought is that that particular point has a fault. Points have been known to go bad. It may be that the links are not made or have not been disconnected into the right configuration. Not what you want to hear but replacing the point may be the next step.

 

Having correctly wired the other 18 points I won’t suggest this one is wrongly wired from the point back to the switch.

 

Rob

[NormanQ4]

@RAF96

Not the same loco, could be any loco, in fact the fault occurs before a loco even comes into play.

The loco just draws attention to the fault.

At first fire up and after any power interruption the frog polarity is reversed, but using either option 1 or 2 corrects the situation and thereafter the point performs perfectly.

As I mentioned before the track voltage remained constant.

I had an electrician friend, who's recently got involved with Hornby and RM, look at it yesterday and he couldn't figure out why this would be happening, other than that the decoder may be assuming an unusual default position with regard to the point / frog.

[Chrissaf]

I would temporarily swap this affected point motor wiring with a different working without issue port on the same or preferably different decoder and see if the fault stays the same. If the fault stays on the same point, then as Rob suggests, suspect the point itself, if the fault moves with the decoder port to the another point, then it looks like that particular decoder port could have a fault.

[St1ngr4y]

When the ADSn-FX decoders first appeared, I found that you could have a short immediately after powering on or after a short elsewhere had caused a loss of power. The reason for this was that the onboard relay which is used to switch the frog polarity, is spring-loaded in one direction. This means that power is applied to it to switch the polarity one way and power is removed to switch it the other way using the spring. In turn, this means that if the point itself is left switched to the direction which requires power for the latching relay, then if power is removed from the whole layout either by a closedown or a short elsewhere, then the spring switches the relay to the wrong polarity for the direction the point is facing. So when power is returned to the layout, the frog is fed the wrong polarity from the decoder's relay immediately.

I would hope that the newest decoders (-SX's) would have relays which remain in the same position during a power off.

 

I don't know whether this would explain what you are seeing or not.

 

Ray

[96RAF]

There was discussion on RM Web about odd behaviour of this brand of decoder, but not sure of the specific model type involved.

Rob

[St1ngr4y]

Norman,

 

On reading your initial post again, when you use option 2 to remedy the situation, this is switching the relay so that it matches the setting of the point. You may hear a quiet click from the decoder when you do this.

 

The way to avoid the short on a normal power up, is to switch the point to the left just before you close down the previous session.

 

Ray

[NormanQ4]

I think that the decoder brand is definately the issue, I've had so many problems with the 2fx in the past although DCC concepts say they've produced thousands of these units with very few failures. 

I've had three on my layout over the past couple of years and all have failed. The current problem is with the 8fx which powers points and frogs #8 to 12 with one port dead and two ports spare.

Here's the latest on what is happening now.

I had been thinking of using another port and when Chris suggested it I was convinced to do so.

Exactly the same bad result, but there's more.

This point is #8 from the mainline to the goods yard which contains points #9 to 12.

The mainline was activated by the green button (right) and this was the only direction affected.

Just by chance the loco ran further into the goods yard and shorted on point #11 which was set right.

I decided to check all of the other points within the goods yard by disrupting the power and then running the loco through each point in turn. 

To my amazement all of the frogs 8 to 12 failed (reversed polarity) when the green (right) button had been activated.

No shorts when the red (left) button had been activated

So to prove a point I reverted to the original port and got the same result.

I've really just been constructing my layout over the past couple of years and have only been using the main lines to entertain my grandson, I've barely looked at the goods yard and if a short occurred I just looked at it as an isolated incident.

I checked all the other points on my layout which are powered by the newer model decoders one 8sx and three 2sx with no problems.

I await your comments. 

[St1ngr4y]

Norman,

A couple of questions...

 

1. Are you using the latest electrofrog points where you can cut a couple of wires underneath and solder another two, and have you done that.

2. When you wired up the frog to the ADS8fx, did you remove the wiring of the frog to the Seep?

 3. Did you read my last but one post?

Ray

[NormanQ4]

Hi Ray

Yes, Yes and Yes, but I didn;t read your post prior to doing my testing, I've read it now for the first time.

Coincidentally I usually leave all my points "contrary" to their preset start up position which for "# 8 is left and #9 to 12 is right" which goes some way to explain why I didn't notice the problem on the other points.

[NormanQ4]

The only issue that I'm having with the sx decoders is when there is a short anywhere on the layout and I get the remove short circuit message from RM, I have to go around the layout and move any locos or carriages with pick ups for lights from all the frogs otherwise the short message and power cut off remains.

This is a real bummer since apart from the locos I currently have 18 carriages with lights and was planning to do another 7, so a lot of shuffling to be done.

What advice is available on availability of a similar type of accessory decoder with easy set up and frog switching?

[Chrissaf]

In that case what about leaving all your Accessory Decoders as are (to operate the points), and use the money that you would have spent on replacing them to buy Tam Valley 'Hex Frog Juicers'. These are totally 100% electronic without mechanical relays. Remove your frog wires from the current decoders and connect the frogs to the Tam Valley instead. The Tam Valley switches the frog power in less than a millisecond and I have found them (I use the two port version as a RLM) 100% reliable in operation.

.

The Hex Juicer board can independently switch the frogs of six points.

.

http://www.tamvalleydepot.com/support/whichfrogjuicer.html

.

PS - Configure the links on the board to use the 1.7 Amp trigger current.

.

The whole board connects to a single DCC feed, but only one single wire needs to go to the individual points to connect to the frogs. The six ports operate independently on the board and auto switch when they detect a short (thus there are no miss-wiring issues that can arise - unlike the relay based decoder products).

[NormanQ4]

@Chrissaf

Thanks for that Chris, having read the script it seems like a foolproof solution.

I was planning in adding 4 more points to my layout which would take the total to 23, which in turn would require 4 Hex juicers with a total cost of around $A400, just a bit pricey.

I wasn't planning on replacing my decoders as the SXs are all under a year old with the 8SX only a couple of weeks old at a cost of around $A110.

I was hoping for a more cost effective solution using the kit that I have with work-around solutions if possible.

[96RAF]

How about these...albeit using a mechanical relay.

http://www.gaugemaster.com/item_details.asp?code=DCC80

 

Not used them myself but forum chatter indicates they are fit for purpose at little money.

Rob

[NormanQ4]

@RAF96

Thanks for that Rob, that looks like it may fit into my budget.

I had seen them on the Hattons web site, but not being up to speed on that sort of stuff I didn't follow through.

Would anyone who is using or has used the Guagemaster DCC80 Autofrog like to give an opinion on their use?

Forum chatterers here's your chance to chatter, please be my guest.

[SteveM6]

I use them on my OO layout and have no problem with them at all. Three wires to connect and away you go.

Thoroughly recommended.

[Fishmanoz]

Seems to me the DCC Concepts product causing the problem is not fit for purpose. Have you spoken to them about it, told them exactly what is happening and asked for a solution?

[NormanQ4]

Hi Fishy

Yes, I have contacted them asking for a solution.

I don't expect to get any joy though given my past experiences with them.

My experience was that they blame the customer for everything and if that starts to look shakey just lie through your teeth.

It just cost me more money for postage to send the goods back and Consumer and Business Services SA were less than useless.

[NormanQ4]

Hi Fishy

I just thought I'd let you know that I got a response fom DCC concepts at 9.06 pm seeking further clarification so perhaps there's light at the end of the tunnel.

I'm so glad that the response wasn't from the clown that I had been dealing with in the past.

I did critically mention my previous problems with the fx decoders and I quote his reply. 

"Indeed the FX version had their faults which is why we replaced them."

The previous person denied there was any inherent problem with the fx and inferred that 99% + of faults were caused by customers, and that they weren't replaced because of their design faults.

It's refreshing to have someone tell the truth and gives hope for future dealings.

[Fishmanoz]

Good to hear Norman, matches my dealings with them, although going back a bit now. Mine were with the owner Richard Johnston. 

[NormanQ4]

Following is the solution suggested by DCC concepts to the shorting frog issue.  Any comments or advice would be welcome.

 

"I have been doing  some research for you into this problem.    

What is happening here is the Hornby system is still thinking there is a short when the frogs reset on the point decoders. When the Hornby system cuts the power to the track they are cutting power to the ADS decoders on your layout. As part of this the relays on the decoders will default to a centre position when they loose power. Now when they get power back they reset back to their last position but there maybe a very small power polarity swap as they do so which is being pick up by the Hornby system.  The short circuit protect on this system is very sensitive as I know from the fact that other DCC systems don't behave like this when resetting.   

My recommendation would be to split the track power bus from the track output on the Hornby system into two buses with one powering all the track and the other just the ADS decoders. I would then recommend that you use a dual circuit breaker like the PSX-2 unit from DCC Specialities. These units are by far the best out there and will in bonus provide a extra layer of protection for the system.  This unit will allow you to have a dedicated breaker for both DCC buses. This way if the track shorts then just that side of the of the power is cut at the PSX-2 and the system doesn't have to reset. This also means that the ADS decoder still have power and so you do not need to would about having to move the trains around till every frog is clear.  The PSX units are also adjustable in there sensitivity for when they trip.    This would be my recommendation for this scenario and I hope all this information is helpful.  I can draw a wiring diagram for this setup if this helps also."  

 

 

And here's the link to the unit in question. https://www.dccspecialties.com/products/pdf/man_psx1.pdf

[96RAF]

Using a split bus with circuit breakers seems to be a sensible suggestion and has been recommended before on forums with folk discussing shorting issues - allowing points to be reset then train running to continue. Being able to adjust the sensitivity of these breakers may be the final workaround to this one rogue point..

 

Is the centre off springing explanation contra to what Ray described earlier about these decoders, i.e. powered on and sprung off making me wonder if the same decoders are under scrutiny. Or did Ray mean sprung centre off.

 

Would we be seeing this problem in the same light if RM were not involved giving us an on screen indication. I am not fully convinced by the Hornby system persistent short rehandling explanation, especially if a 4-amp power supply brick is being used.

 

Had it not been for the amount of wiring disruption required I would have been trying the unit substitution fault finding regime previously suggested to see if the problem moved with the unit or stayed with the point, then we could be more sure of where the actual problem lay.

 

Personal opinion - pull the frog switching on that decoder and fit the cheap and cheerful auto-switching device to that frog.

 

Rob

[Colpatben]

I was planning in adding 4 more points to my layout which would take the total to 23,

I wasn't planning on replacing my decoders as the SXs are all under a year old with the 8SX only a couple of weeks old at a cost of around $A110.

I was hoping for a more cost effective solution using the kit that I have with work-around solutions if possible.

and

This is a real bummer since apart from the locos I currently have 18 carriages with lights and was planning to do another 7, so a lot of shuffling to be done.

 

Reading back to earlier posts plus the suggestion from DCC Concepts regarding overload protection, I believe that you are shortly going to have to consider how much power you have available for your complete layout.

 

One ‘Problem’ with the DCD-ADS8fx is that as they have a CDU for each output i.e. 8 off for each unit, on initial power up they will have an inrush (instantaneous) current of 350ma or approx 1/3 of an amp if you have 3 units then these alone would have an inrush of over 1.0 amp! This will settle down quite quickly to 90ma per unit which will be less than 300ma in your case when you have installed three DCD-ADS8fxs. Add to this your lit carriages probably at 80 to 100ma for each carriage (I am assuming that they are track powered and not fitted with individual decoders so are lit continuously), then several locos then your start up current may be such that you will trip your controller overload. I note from your footer that you have an eLink and I presume are using a 4amp PSU.

 

As an aside it is my opinion that it worth the investment in at least two current limiting devices on any small layout, one for accessories and one for the track.

 

Larger layouts with power boosters are split up anyway and different parameters are considered.

 

I wrote a piece regarding overload protection using cheap filament bulbs and published it as per link below which may be of interest.

 

http://www.ngaugeforum.co.uk/SMFN/index.php?topic=36637.msg431749#msg431749

Colin

[NormanQ4]

I would value further comment from anyone who has used the aforementioned DCC Specialities PSX-2 dual circuit breaker or similar.

It seems like a good bit of kit to my inexperienced eye.

It also works out much cheaper than 23 Guagemaster DCC80 Autofrog units and would be much easier and quicker to install.

[St1ngr4y]

Hello Norman,

I have a setup which gets around the problems which I described with ADSn-FX decoders. I use an Elite as my controller and I have an accesory decoder bus connected to the main track output from the Elite. I also have a track bus which is connected to a Power Booster unit which in turn connects to the Booster output of the Elite. This means that if and when a short occurs on the track, the Power Booster cuts out showing a red light, but the short isn't passed on to the Elite or Raiilmaster. In turn, this means that RM can operate points while the short still exists on the track. Sometimes (but not always) this allows an incorrectly switched point which caused a short by a train fouling the frog, to be corrected without even moving the train away from the point. Having corrected the cause of the track short, powering off then on again the Power Booster would restore power to the track. To do this it is necessary to unplug and replug the power input lead to the Booster, so to aid this operation, I inserted a single-pole push-to-break button into the power input lead to act as a reset switch.

Another advantage of this is that at the start of an operating session, the Power Booster gets a short circuit immediately on switch on of the Elite, but RM is still able to "set points at start up" if you have asked it to. After this has been done, the ADSn-FX onboard relays are then synchronised with their respective point settings, so a single push of the Power Booster reset switch restores power to the track and everything is ready to go.

 

Ray