High resistance leakage on hand laid turnouts

[bdennis]

Gents,
I have a problem that is driving me nuts on a number of my hand laid turnouts.. 
So am looking for some assistance..

I have had great success in building around 30 so far in terms of gauge and reliability etc..
I use a fast tracks jig to build each #6 turnout.
Then I uses a printed paper template (again fast tracks) to space out the wood ties in the right places. I use a simple glue stick to glue the ties to the paper template.
I then use Pliobond to glue the ties to the turnout to complete the assembly. I use a small blunt syringe to place the Pliobond to each tie. Some does run over between ties but wont be seen once painted.. I leave the paper attached to the bottom of the turnout.
I then test for gauge and use a bogie to test to make sure all works correctly..

Im having issues with 75% of my turnouts where they dont have a dead short but they do conduct (all be it at a high resistance) between some of the rails.
It is not uniform but it is a bugger to find. The high resistance does not cause any issues with my Digitrax DCC system in terms of dead shorts (yes a coin test in the location will cause the circuit breaker to trip) but plays hell with block detection as there is a high resistance present and is at the right sort of value that causes the BDL to think that the block is occupied.!..

I have tried making sure that the gaps I cut (using a round file) in the copper ties (fast tracks ones) is nice and wide and have run a knife over the edge (on a 45 deg angle) on both sides to make sure there is no copper left. I also file a gap in the copper on the bottom of every tie to safety before assembly..

I have tried to clean the turnouts before gluing the wood ties with a small wire brush to remove any left over flux or residue.
While this cleans the copper and the side of the rail it does not fix the high resistance issue.

I can eventually get each turnout to work but it is a real pain!..

Im wondering if the Pliobond is conductive or or the paper / glue stick / Pilobond is some how conducting..

I thinking that the issue MUST be with the copper ties and the way I have cut the gaps or some residue perhaps flux is still there that is causing the issue.
All turnouts are built on my workbench and then tested before installing.. (I found this the hard way as I have a few that I installed and now have the BDL permanently detecting occupancy!.. Im going to need to do some fault finding on the particular block / section to find the problem turnout..

Perhaps it is flux that is causing the issue?? I dont use that much.. Is there a flux remover?
If I have to remove all the wooden ties and dip the turnouts then thats what I will do but would prefer not to have to do that on 15 turnouts!.

Any thoughts / suggestions would be appreciated..

For any pictures see my blog..
http://dh2ndsub.blogspot.com.au/

[central.vermont]

Maybe  the "flux capacitor". 

Sorry Brendan, it just had to be said!!

I haven't heard of this problem before but was thinking that maybe when you file the gap the metal shavings are being embedded into the tie some  how? Really not sure on this. Not heard of flux being conductive though. Maybe peteski will chime in here.

Jon

[Sokramiketes]

Have you tried pumping 20 amps through one and seeing what melts?

[bdennis]

Have you tried pumping 20 amps through one and seeing what melts?

Mike,
I had not thought of that..
I guess it is worth a try!.. It will certainly fix the short..!

I have some 7.5A power supplies that are short protected..
The alternative is a 12v Car battery!.
Thanks.

[bdennis]

Hmm.. Tried a 16v 7.5A power supply with no luck.
The resiatance between the rails varies between around 4M Ohms and 12M ohms on the first turnout I treied..
Checked the current drain when the 16v 7.5A power supply is attached and no current registers on the digital multimeter.
Perhaps, if "shock treatment" approach is to work, more power / current is needed!...

[Chris333]

Plug it into the wall outlet 

[C855B]

4 meg ohms? And it's causing problems with a BDL168?

Frankly, I tend not to trust the accuracy of generic DMMs at these very high values. There are specialized leakage testers for more precise readings.

Can you be more specific about the problems with detection? I can't reconcile the recommended maximum 10K axle resistors with your experiencing false detection under turnout leakage in the >4M range. Something doesn't compute and it would be nice to get to the bottom of it.

Anyway, flux is nonconductive, and Pliobond shouldn't be conductive (been decades since I've used it, tho'). Jon's suggestion about metal shavings could be a lead, as the wooden ties, being an organic hydrocarbon with slight moisture content, are going to have the tiniest bit of microconductance. A dusting of shavings could provide just enough additional conductance to be measurable on a common DMM. But that's an edu-SWAG in the extreme.

[Jim Costello]

Brenden,  I wash my turnouts  before adding the ties in a warm/hot water detergent mix and scrub with a toothbrush and small wire brush as well ,and let to dry before adding the ties.
For custom turnouts, ie, those not made in the jig and where ties are individually applied, I apply pliobond directly to the back of the rail and using long lengths of strip wood ,pick up the turnout and apply directly
to the tie material,holding in place for a few seconds and cutting to length with a craft knife.I do up to 6 or 7 ties this way before completing by adding more pliobond. I use the micro nozzle onto the pliobond tube
to control the amount of pliobond. I also use the soldering iron to set the glue.
I also check conductively with a meter before installing, you will be surprised on how a small sliver of copper or rail burr somehow seems to make contact when you think it is ok

Hope this helps
Jim…Modelling the NH down under

[LV LOU]

4 meg ohms? And it's causing problems with a BDL168?

4Meg? Jeez..I'm sitting here with a Fluke meter,I have one probe in my right hand,one in my left,I'm getting 1.5Meg..If that much resistance is causing a problem in one turnout,with ballasted track on a damp day,your trains are gonna be at a standstill..

[bdennis]

Mike,
Thanks for that.
the reading I had taken earlier was just 1 turnout.

I have 10 or so that are playing up.
The test I completed was basically a set of aligator clips to the individual turnouts in isolation then connected to a actively detected block on the layout.. So Im only targeting the turnouts that when connected to powered / detected track then they show up as occupied..

When I check the turnouts, the problem can be just the frog, or both stock rails or the rails north of the frog or any combination of the mentioned rails...

If I can get them to test and not detect on the BDL then I will be happy..

[bdennis]

4 meg ohms? And it's causing problems with a BDL168?

Just on this one.
The readings I took from 1 turnout had different readings depending on the trails I tested.. This particular turnout has 10Mohms on 1 pair of rails and that is setting off the BDL.. The other rails with the lower reading are not setting of the BDL..
So my aim is to get the turnouts to a point where I can check each rail and not set off the BDL before installing. Ideally, each rail will have NO reading on the multimeter at all..

In terms of the throw bar for the tortoise machine. I have not yet drilled the hole or installed the turnout so there is no issues with the throw rod on the Tortoise bridging the gap on the throw bar.

[Chris333]

Look at the gaps close with a magnifier. Maybe there is hair thin copper that got pushed over the edge of the tie or something like that. I also wash/scrub all turnouts with dish soap and a toothbrush.

[bdennis]

Thanks for the suggestions guys..

Chris,
here is a pic of one of the problem turnouts..
As you can see, the gap in the copper ties in not small.!

[Chris333]

I see extra gaps so you should be OK, I don't know what is wrong.

[mmagliaro]

Wait wait wait... you still say that your BDL is tripping when you have a turnout with a "problem" resistance of
4 Megohm?   Something is not right.   Block detection should not be tripping with such a high resistance.
100k, maybe.  But 10k is more like it.

I highly doubt it's your gapped copper ties, but regardless.  Can't you adjust your block detector so it doesn't
trigger on such high resistances?  Heck, 4 meg, ... you will have problems with damp weather and an errant finger
placed across the rails.

And no, with resistance that high, you are never going to be able to "burn through" the problem with a high-current
supply, because the current draw will be so low at 4Mohm that nothing will happen.

What is this block detector that triggers at 4 Mohm?