Skip to main content

Looking for our electronic and electrical experts help and opinions. I'm looking for a way to easily detect a blown TVS Diode for use for Track Power or even in a Locomotive. Does this seem like it would work? I know that we connect a TVS Diode between the hot and common but no power passes through it unless a short happens, the TVS absorbs the power spike, if the TVS blows it results in a short through it connecting the hot side to the common if I have this right. Normally no power should be passing across this part of the circuit so I want to try using a TVS, Dropping Voltage Resistor, and 12 Volt LED in series between the Hot and Common Transformer Leads under the layout before power feeds to the Track Connections above. My theory is that if the short happens on the track, then corrected (like a derailment, train put back on the track correctly), and power is applied and Track Feeders TVS that shorted should now complete the circuit sending power through the shorted TVS, Dropping Resistor, and LED lighting it to show which TVS Diodes need to be replaced. I figure if this would work that it would be easier than having to test every track feeder, and also tell that's the issue in a Locomotive if it doesn't run afterwards. Does this sound like it should work without any adverse effects? Just wondering. I have enclosed an attachment drawing. Thanks for looking, I would definitely appreciate any help or suggestions.

Attachments

Images (1)
  • 20230202_120358
Original Post

Replies sorted oldest to newest

Longer explanation.

You created a series circuit with the TVS. When the TVS conducts- now that current is going through a dropping resistor and LED- neither of which can handle much current or spike in voltage.

Net effect- the LED blinks when a spike is conducted- until it blows the LED and then the TVS isn't doing jack squat. It wasn't doing anything useful before in the circuit for protection and now less than nothing when your circuit blows the LED.

Drawn vertically which better shows light in parallel to fuse. When the fuse blows- circuit is no longer protected= bulb lights when track is powered indicating lack of protection at that point.

Replace both the fuse and the TVS when the fuse blows.

*****The fuse is NOT protecting the track from over current in this diagram****

The fuse is there so when/if the TVS fails shorted, then the fuse blows from the sustained high current, providing an indicator via the light bulb. Also, you want to use a relatively higher amperage fuse. One that would blow if the TVS remained shorted, but not so low as to blow prematurely. Again recommending probably 10-15A, maybe even 20A fuse value.

Attachments

Images (1)
  • mceclip0
Last edited by Vernon Barry

Using and thinking about TVS as protection to prevent damaging voltage spike a good thing. They are relatively cheap and effective. However, putting them at lockons all over the layout- potentially overkill. Worse, too many TVS can load down the DCS signal and be a nightmare to find the one that died shorted somewhere on the layout. This is where you sit- extreme overkill IMO with them placed all around your track where you cannot easily get to them.



Conversely: Putting anything in series with the TVS would tend to reduce it's effectiveness at shorting the voltage spike and preventing damage to the things we are trying to protect.

As a compromise, the circuit I presented is a way for a hard to get to track connected TVS, using a low impedance larger rated fuse in series with the TVS should not prevent it from dampening and killing off most spikes. This is again where fuse size (rating) is important as we need a brute force higher current value fuse to maintain that connection through even the most serious spikes. It's only when the TVS fails dead short, and your power supply has enough brute force against this sustained short, then the fuse blows open. When that happens, now the light bulb is in parallel to the open circuit fuse, and the TVS is failed shorted- now the light bulb lights indicating we have lost protection. Again the light on indicates a failure- but only valid when the track is actually powered.

Further- using a light bulb rather than a fuse offers a slight advantage IMO. In the event it was a spike that blows the fuse, but did not kill the TVS, a light bulb is (relatively) lower impedance in the cold state. So effectively the TVS is still in the circuit when the fuse is blown. Granted this is not ideal, but compared to the LED and dropping resistor, I feel this might be a better circuit.

Again this drawing, minus transformer is more like your lock on setup. Key is, use a 10A or higher fuse- maybe even "slow blow" type.

Attachments

Images (1)
  • mceclip0

Thank you Vernon, I appreciate your honest comments and better solution.  I run conventional off a transformer so I don't have to worry about compromising any DCS signal. I thought it was a starting point to try to come up with something that would work and not have to crawl around and check every TVS, I honestly didn't think the resistor and LED could handle the power but hadn't thought of using a regular 18V light bulb and fuse circuit. Thanks!

Last edited by Gary P
@Gary P posted:

Thank you Vernon, I appreciate your honest comments and better solution.  I run conventional off a transformer so I don't have to worry about compromising any DCS signal. I thought it was a starting point to try to come up with something that would work and not have to crawl around and check every TVS, I honestly didn't think the resistor and LED could handle the power but hadn't thought of using a regular 18V light bulb and fuse circuit. Thanks!

No problem, I had this idea in my mind for a while of this fuse and light indicator, but never previously shared with the forum.

Again, this is not my first choice. Best protection is the TVS as shortest and nearest direct path across the rails. This circuit represents a compromise, where it provides protection, and in the event of catastrophic failure provides a way for the fuse to blow and an indicator that could be seen remotely.

The other experts may chime in and shoot holes into my idea.

Vernon Barry, Gunrunnerjohn, and Adrian, you are all amazing! Such brilliant minds, and thank you all, and so many others, that help everyone here on this forum when we have questions that you all are willing to take the time to help us out. It's greatly appreciated! I will have to give Vernon's idea a try, and I will definitely have to put TVS Diodes in my Locomotives.  Thank you all!

Gary thanks for asking a great question.  Vernon, as usual a clever solution.

I agree with Adrian’s and John’s advice about installing TVS at/on the locomotive’s board where track power connects to it.

Additionally, on a layout without DCS, I am of the opinion that installing TVS diodes across transformer outputs, relay coils and other electromagnetic coils (uncouplers) connected to track power will help more quickly attenuate the oscillating spikes generated by derailments as they bounce around the conductors on the layout.

In my experience, the control boards in FasTrack Switches, if fed by Track power instead of Aux power, seem to be vulnerable to voltage spike damage from derailments.

Last edited by SteveH

SteveH, thank you. Yes, I have TVS Diodes wired across all my track feeders. I'm using Aux Power to power my Remote Fastrack Switches and Fastrack Uncoupling Track, with TVS Diodes wired across them, my two MTH Z-1000 Transformers output terminals. I still need to put the TVS Diodes in my Locomotives.

Thanks Gunrunnerjohn, I run conventional with 2 MTH Z-1000 Transformers for my 2 tracks on small layout 8 foot x 45 inches (to clear the door opening and gain 3 extra inches to squeeze through the narrow walkway!) with a 6 foot x 4 foot "L" extension that fits my "L" shaped room. No need to worry about DCS Signal or Command Control at Phillips Junction in Slackerville. 😁 The owner says we have to run on a tight budget.20230125_115542

Attachments

Images (1)
  • 20230125_115542: Bandit, the Boss

No, you created a shoddy spike detector with that circuit.

It will not protect anything and will just blink when there is a spike.

Vernon,

I think you're right on the money.

Actually the LED might only blink once, in the process of frying, when the first spike hits (during the period before the TVS actually clamps).

This of course depends on the TVS's specific  rated clamp voltage.

Mike

I think there's way too much emphasis on attempting to detect the TVS failure.  A shorted TVS (the most common failure mode) will be fairly obvious.  An open TVS is difficult to impossible to detect in circuit, so that's a somewhat moot point.  Also, the TVS parts really don't fail that often, standing on your head to detect the occasional failure doesn't make much sense.  There are tons of other failures of electronics that are as frequent, but we don't attempt to have indicators for all of them.

I suggest you use protection devices in a proper manner and deal with the failures as they happen.

Add Reply

Post

OGR Publishing, Inc., 1310 Eastside Centre Ct, Suite 6, Mountain Home, AR 72653
800-980-OGRR (6477)
www.ogaugerr.com

×
×
×
×
Link copied to your clipboard.
×
×