12V Level crossing issue

[Heppeng]

Appologies for resurrecting an old thread but I need some help with a 12V level crossing that I have just bought. (see previous page for reference). Unfortunately the flasher does not work when connected correctly. The symptoms are that the lights stay on permanently when the barriers are down. Reading the above it seems that the circuit is designed to do this if the polarity of the input is the wrong way round. So I reversed the inputs (even though originally it was connected to the correct socket on the switch) and the results were no lights on at any time. I have dismantled the brick and cannot see any dry joints (my usual first suspect when a PCB does not work properly) My soldering skills are reasonable, so I was wondering if Mark may be able to advise on what could be wrong and the best route to getting it working again? My electronic knowledge is a little limited, but I have printed off the circuit diagram and as a last resort I could just replace all the components one by one until it works (hopefully)!!

Edited August 9, 2013 by Heppeng

[JopieK]

Hi Heppeng,

I have had the same issue. It is a 'dead' blinker. That happened when I was a teen and did not understand electronics at all. Can you share a picture and/or the schematics?. I suspect it has to do with a capacitor.

B.t.w. better to make a new topic for this, this is a useful question and we might also want to index it if the topic will yield any success of course!

Update:

it is called a "Flashlight Unit 12 V" / x543

[Heppeng]

The schematics, or links to them etc. were all in the previous topic which I posted in. Unfortunately as this has been split a lot of my references to it no longer make sense, so here is a link to the original topic:

http://www.eurobrick...989#entry728323

Incidentally the circuit is unusual in that it does not have capacitors, but instead uses chokes and a darlington pair.

Edited August 10, 2013 by Heppeng

[JopieK]

Problem was that that old topic was moslty about the whistle kit. Here is the schematic from the other topic:

http://en.wikipedia....tion_oscillator

After reading the explanation by Mark Bellis, I suspect that the inductors might be the problem. You can of course measure the circuit using a scope and see how it behaves and why it doesn't become a-stable anymore (that is in effect the problem of course).

Replacement inductors can be found here:

http://nl.farnell.co...mh/pg/110177985

[Heppeng]

Thanks very much for that. However I suspect there may also be more to it than that since when the polarity is reversed, the lights are off when it would appear that they should be on, This could mean that the diode on the right is open circuit, but would that necessarily interfere with the flashing and give the symptoms I have? I do not have a 'scope, just a digital multimeter. I don't expect the components are expensive, so I may just replace all diodes, inductors (as per your suggestion) and the two transistors and cross my fingers...

[JopieK]

Well that is very logical of course:

The right diode will conduct if you put 12V on plus and the GND to minus. So it basically by passes the whole circuit and current will flow through the lamps and the right diode. If you reverse it the diode blocks the flow so the lamps will stay off. So basically you now know that at least the most right diode works well :)

B.t.w. you can 'beep out' the diode of course with your multimeter. Of course you could replace all parts, but it would be best to find out what the problem is first ;)

[Heppeng]

The diode is there for circuit protection or to indicate an incorrect connection and conducts only when the polarity is reversed, resulting in the lights permanently on when the supply is connected the wrong way round. If you look at the circuit, if you reverse the input and make the bottom positive then current flows straight through the right hand diode, through the lamps and out what is now the negative at the top. When connected the right way round current will flow through the transistors and the other diode.

With mine, when connected the wrong way round you get nothing instead of permanently on lights if it was working properly, and connected the right way round permanently on instead of flashing. I will have to dig out my multimeter and test it but I am a little busy at the mo so it might have to wait!

[JopieK]

Yups, a diode is a 'one way street' for current... the inductors are just coils of copper, they the magnetic field works just like a motor so it can induce unwanted spikes that go into the wrong direction so to say. So it definitely is in either the inductor(s) or the transistors (be it BC547 or the darlington pair). It could be even the second diode, but I don't think that will be the case. A transistor or an inductor are much more vulnerable.

[JopieK]

so Heppeng, how is your 12V level crossing doing?! any progress?!

[Heppeng]

It might be a little while before I commence on this, might get a chance at the weekend.

[Andromeda]

How did you get on Heppeng?

I have been having a play with this circuit recently, and although Mark has done some excellent work in reversing this, there may be a couple of omissions.

The 18R resistor is actually a 15R (Brown,Green,Black). The signal diode (LHS) is a 1N4148, and the (RHS) diode is a 1N4007.

I have tried to recreate this circuit and found that the 22mH inductors that I purchased are the cause of the fault. Confirmed by connecting my 22mH’s to a working flasher cct and got the same results. I am going to investigate further and let you know.

[Heppeng]

Its still on the to do list! I really ought to get around to it!

[Andromeda]

Right, I've cracked it!

The two ‘inductors’ in the circuit are actually capacitors, 22nF to be precise. Confirmed by my newly purchased PEAK LCR meter, and by fishing out two Mylar 22nF caps out of the junk, and connecting them to a working Lego circuit.

Happy days!

Just to set the record straight: Mark’s online picture (http://www.brickshel...y.cgi?i=4448290) shows an 18R resistor, my circuit has a 15R.

Edit: And without Mark's inspiration, things would have been a lot harder.

Edit II: hopefully corrected images!

Edited November 19, 2013 by Andromeda

[Heppeng]

Looks like its just as well I did not get around to it sooner, sounds like I may have been frustrated!!! Many thanks for that!

[Andromeda]

Indeed a headache. I should have tested the component before ordering a half dozen 22mH chokes... !

I have a couple of broken flasher units on their way to me now. I think at least one of them exhibits the same symptoms as yours, I will let you know the out come.

Due to these fun and games I have recreated the circuit on stripboard, I won't hijack your thread though, I will start another !

If you don't mind I could put a link in here at a later date, and link your page from mine...?

[JopieK]

Hi Andromeda, no problem using this thread if it is about the same issue!

Maybe a good idea to also use a scope?!!

[Heppeng]

Likewise no problem, you have been very helpful and your contribution is welcome!

[Andromeda]

Maybe a good idea to also use a scope?!!

Indeed I wish I had one at times! My PEAK testers and DVM are my trusty companions. This is a very simple circuit, and made simpler my Mark’s disassembly. There were a few red herrings though!

Likewise no problem, you have been very helpful and your contribution is welcome!

Hi Andromeda, no problem using this thread if it is about the same issue!

Thanks. It’s good to keep information flowing! I have seen so many times in threads that someone fixes something, but omits to give any detail. Plus I have had my fair share of just ‘taking’ when it comes to forum threads…

[Andromeda]

My broken flasher unit arrived today. Contrary to the sellers listing, it did not light the lights at all.

After dissassembly, 4 alternate holes(top and bottom) need gently depressing and pulling apart, I started to test.

Using my DVM I tested all but two of the resistors in cct, they were both the 4M7's. Those I lifted a leg so to speak, with a soldering iron to test. My MPSA13 resistor was reading 12pF(out of range on all ranges with DVM) when it should have read 12R - yes 12ohms brown red black. OgeL have now used three different rated resistors for this unit, and I have seen two different signal diodes too.

The two diodes can be tested in cct too, they should read 0.5-0.7V. Similarly the two transistors can be tested in cct. You should get 0.5-0.6V between middle and one edge pin, and around 1.5v between middle and the opposite edge pin for the Darlington(MPSA13). And around 0.7V each way for the same pins on the BC547. Both of my transistors were toast.

Don't be afraid to reverse your DVM test probes when testing diodes/transistors, make sure you are in Diode mode.

The two large pink looking 1/2W resistors are in fact the 22nF capacitors. They can be difficult to test unless you have a decent DVM. I had to purchase a separate meter for this - see previous post.

Looking at the cct diagram that Mark provided, IMHO the most likely failures are the two transistors, the 1N4148 diode and the 12/15/18R resistor. I will however see how this proves out as I have another two broken flashers on their way to me for fixing as I type.

[Andromeda]

From three broken flasher units I have not found one broken 1N4148!

Two had a failed 12/15/18R resistor, center center component. When replaced, the lights illuminate constantly in either polarity.

All three had a blown MPSA13, outer pins showed a dead short, even when center to outer showed 0.7V.

One had a blown BC547B/BC237B - yes ogeL really had a variation of parts!

I have, in my excitement to fix these, omitted to take any photos! I have one more coming from FleaBay, and I have at least one fried resistor, so I'll endeavor to to upload a few pics soon...!

@Heppeng

I can pm you a care suspected package. Have you been able to follow any of my testing from post the above post?

[Heppeng]

Thanks for the PM! I think I can just about follow, from your description it sounds like I have at least a blown MPSA13 since mine illuminates constantly on either polarity. Hopefully once you have sent me the bits I need I can get busy with the soldering iron and all will be well!

[Heppeng]

Yippee!

Half an hour last night with a soldering iron replacing the transistors and diode and I now have a working flasher!

Many thanks to all who contributed to this thread, particularly Andromeda without whom it would have been much more difficult - if I even would have suceeded at all.

The most difficult part was removing the old components - it was a bit too much of a challenge to try and unsolder all three legs of the transistor simultaneously with one iron so i could pull it out, so I just cut the old transistor as near to the case as I could, and was then able to unsolder the legs one by one. You need to be careful that the new components are soldered in at the same height as the originals, I put them in a lttle high because I was worried about heat damage during soldering, and longer legs = less heat in to the component. This meant that it was a very tight fit back in to the casing, and it only just went back together!

The barriers certainly got a work out last night - you just can't help playing! :)