Use of TVS electrical surge protection

The TVS units I typically recommend are bi-polar devices, so they don't require adding two for an A/C output.

1.5KE36CA at Digikey

I designed avionics for many years, and we had very harsh protection criteria for many of our boxes.  I learned a lot about transient protection, and we used a lot of TVS devices in many of the designs.  I also have an impressive collection of ferrite beads of various sizes that got generously sprinkled around on boards, primarily to prevent emissions.  It is rare for a new design of a complex piece of avionics to make it through all the environmental testing without some tweaks, and they usually were in the EMI emission and protection arena.

Even the guys that are EMI specialists don't claim to have all the answers as far as EMI and lightning protection is concerned, and I don't either.

You may consider a TVS across the pickups of a locomotive to be overkill, and you have every right to hold that opinion.  My opinion is the TVS is a valuable transient protection device, and for the 50 cents I spend on them in quantity, I plan on continuing to use and recommend them.  I disagree with your opinion that they're not useful in a command environment, and I'm a bit confused how you come to that opinion.

Even if you have a surge protector in use on your 120 volt sources I still recommend using a TVS for your trains, especially the newer electronics. It is better to err on the side of caution with electronics then to be complacent and say it won't happen to me.

The price for a TVS unit compared to a repair job for a circuit is very low! Retail price for a TVS might be around $2.50 compared to a $150.00 repair bill.

Lee Fritz

An assumption was made in your post above that the engine manufacturers install all the proper protection in their products. By thinking they had all the bases covered, or thinking something might not be necessary, what if they have missed something that really was needed? The end users don't always use the products as the manufacturers assume they will, users are very creative and come up with all kinds of stuff no one has thought of before. No matter how much thought and expense is put into a product, it can not be made perfect and protected against all possible means of failure. Compromises are made in the design and manufacture of all products.

I'm not an electronics expert, I just dabble in the hobby with a lot of help from others here on the forum. I have no scientific data, but I have read many posts here on this forum about someone's engine electronics being fried after a derailment or short on the tracks. I have not witnessed this happening first hand, all I have to go on is what others have posted. Sometimes the manufacturers make the repairs and sometimes the engine owners get stuck with a $150-300 repair bill. If a 50 cent TVS will help prevent any of this this, I'm all for it and think it's a pretty good idea. It might save me and a manufacturer quite a bit of $$$.

Just for some counterpoint...  

Although the TVS is cheap, even for quantity 1, the cost of repairing an engine from operator-inflicted problems can be very high. There's a potential hidden cost every time a less-than-skilful operator unbuttons an engine, whether it's to add a TVS or replace a (perfectly good) battery with a BCR.

BCRs are fine, however, new batteries don't need replacing until they're too old to take a full charge.

Even dropping an engine's trucks to replace a traction tire can have unanticipated consequences.

I never recommend opening up an engine unless it's necessary to effect a repair or do something else that is absolutely necessary.

 but I have read many posts here on this forum about someone's engine electronics being fried after a derailment or short on the tracks.

Odds are that any boards that get fried due to derailments are victims of poorly-resonding or nonexistent over-current protection, rather than voltage spikes..

Barry, I certainly agree that someone that doesn't have the skills necessary start modifying their equipment, that applies to virtually any device.  I also don't routinely open up a locomotive just to install a TVS, it happens if I'm already in there for other work.

Not opening up a locomotive unless there's some repair needed is certainly not a point that I'm going to argue.

I have TVS's on track and power, but I do not have them on any engines yet. I'm probably semi-skilled when it comes to working on trains, many here are much more skilled than me. I am also a little hesitant to open an engine just to look around or install a TVS, but as GRJ says, if I have to get into one, I think I will be adding a 50-75 cent TVS while in there.

Barry,

I think I agree with too much current also, especially with the older transformers.  There have been posts here that have said the spikes are the most dangerous or most damaging to the electronics. I'm definitely not the expert, I just try to protect my stuff as best as I can with information gathered from reading things here on the forum and an occasional article or two elsewhere. I happen to believe they are both equally bad and I don't want spikes or over current if there is any way I can prevent it from happening.

 To quote GRJ...You may consider a TVS across the pickups of a locomotive to be overkill, and you have every right to hold that opinion.  My opinion is the TVS is a valuable transient protection device, and for the 50 cents I spend on them in quantity, I plan on continuing to use and recommend them. 

   My question is could i install the TVS across the red and black wires (the ones with wire nuts) above the boards or at the trucks?..across the wire nuts all kinds of room..at the trucks not much room.

At the wire nuts would be fine.  They just have to be directly across the wheels and center roller leads.

I am now understanding the debate and feel that this is what possibly happened to my MTH railking ES44ac locomotive.  The only thing is i did not notice any derailment that might have triggered this event.  I have the DCS system and feel confident that i have wired and hooked up my system correctly.  I am relatively new to the world of O-scale trains.  This TVS talk has certainly opened my eyes.  On my layout I have added two additional sidings.  My plan was to goto the DCS system for the purpose of turning on and off the engines I wanted to use at any time.  I guess my plan is flawed because I haven't added any power switches to those sidings.  So if any issue arrises I possibly could have a runaway train on my hands???  I will admit i got confused with one member saying you don't need TVS and another prominent member saying it can't hurt.  I am far from a handy guy with regards to train repair.  Although I am willing to learn.  I just don't know if my confidence level is up to the challenge.  Once again thank you OGR members for opening up a nice healthy debate that I am sure more people are learning from like myself!

Jim,

Your TIU already has a TVS on each channel.

  If a TVS is inop in your TIU how would you know?

Here is an article that shows a burned TPC caused by a voltage spike during a derailment. A solution is offered also for the circuit in question.

Larry

Yes, I made some assumptions:

TIU or the Lionel equivalent includes TVS's in each channel that would protect the track from spikes coming from the power source. Is that true????

At no time is the locomotive motor connected directly to its pickups in a digitally controlled locomotive....There is a circuit board between the motor and the pickups. Installing a TVS on the pickup offers no protection on a DCS type locomotive. If  a TVS is added at the pickups to protect the circuit board how does a motor generated spike get suppressed on the other side of the motor controller by a TVS at the pickups??? 

Once spikes are suppressed at their source no more suppression is necessary.

MTH and Lionel followed normal industry practices for filtering power to circuits at the circuit. I know, a big assumption but I have not read of wholesale loss of any of these electronic components so they must have done something right (???).

Look, these are your trains. Do as you wish. I only offer my take on the TVS situation. I choose to not use them in my all DCS layout. We'll see if that is a bad decision.

LDBennett

Barry,

I ordered the TVS and there not expensive.

My son and I are in the process of upgrading a PS Doodle Bug to PS-2.  Do you have a diagram onto where to install the TVS.

It is quite a job to get everything to fit in the shell. Modifying includes drilling, cutting, and machining.  I was surprise that we needed to change the couplers?   In which we did.  Were to the stage of changing the lights.  I assume these changes are due to the battery voltage.  I understand the lights, but not the couplers.

After catching up on the post for the TVS I found it is more complicated than expected.  My problem is this.

I have found in the derailment problems with locomotives.  Not so much with the diesel engines but with the steam. The Transformer breaker normally takes care of the problem.   The trailing and forward trucks often derail.  I had to stretch or get stronger springs.  I still get a occasional derailment.  On the  4-8-4  the short took the shortest route to spring.    The short to the spring happened so fast I did not realize it was the spring and was cooked before the power was shut down at the transformer.  So the cow catcher arced on a slight grade, when put back on the track.

My question is this.  Will two thin plastic washers on each side of the spring prevent this?

Well I have to disagree with not opening up an engine if it is not broken. because Q/C is all but non existent and assuming the builder did a good job is a mistake. I open all my engines up and usually find something like this going on. future mishaps can be avoided by correcting them early on. it takes some skill to open up an engine, but not much. 

     To quote JohnS...Well I have to disagree with not opening up an engine if it is not broken. because Q/C is all but non existent and assuming the builder did a good job is a mistake. I open all my engines up and usually find something like this going on. future mishaps can be avoided by correcting them early on. it takes some skill to open up an engine, but not much. 

    I completely agree. Out of all 11 of my ps3 and 2 engines 7 needed some attention from repositioning wires to defective 2rail 3 rail switches. I now pull the cab shell and inspect and service before it sees the track.

That is inherently not true as was pointed out by the QSI study and by others.

Did that study (and "others") actually state that there were more damaged boards due to transient spikes rather than over-current anomalies, or did it state what could happen due to a transient spike?

If the former, let's see the studies.

One article Trainman9 refers to is here Surge Protector.

I would be very interested to know what the forum electronics folks think about this if they would be so kind as to take a look at the article and the circuit they came up with and provide their comments here.

I think this just illustrates the point that some of us are trying to make.  There is a potential benefit to having TVS protection, and really there is no downside to adding it.

Let me be clear:

• I am not disputing that TVS's are good things to have
• I am not disputing that transient surges can cause electronics in engines to fail
• I understand that there are lots of published findings that show the damage that a transient power surge can do to electronics

Rather:

• I believe that the majority of failed engine boards are due to poor or nonexistent circuit breaker/fuse protection on train layouts
• I don't see where any of the studies determines, other than possibly as opinion, that that the damage from transient power surges outweighs the damage form over-current situations
• If there is such study that provides proof - not opinion - that power surges more often cause damage than do over-current situations, I'd appreciate it if someone could show that exact language in such a study.

Barry, can you point us to the study that says a majority of failures are due to circuit breaker/fuse deficiencies?  To paraphrase someone here...

If there is such study that provides proof - not opinion - that over-current situations more often cause damage than do power surges situations, I'd appreciate it if someone could show that exact language in such a study.

That door swings both ways.

Doesn't it makes sense to protect against both types of occurrences? 

This is also what I have read in a couple of articles on the subject of protection. It may not be the last spike you get, but and accumulation of them over time. So the more you can prevent the better off you will be.

Wish I would have saved these articles to link to here, but I did not. I am pretty sure one of them was the QSI article being mentioned here. I am also thinking (possibly incorrectly) I read somewhere that some former QSI people or their findings led to the development of the PSX-AC's that I put a plug in for every once in a while. I will try and find some more info on this. Regardless, this is a very interesting discussion.

It seems that some here are mixing up the failures caused by a short (overcurrent) condition, and a voltage spike condition. Also it seems that the protections against both are not understood by all. Electronics are protected from an overvoltage spike  by a TVS and not a circuit breaker or fuse. The breaker or fuse will not act directly in an overvoltage scenario. Now hypothetically, if a spike does burn out an unprotected  board, and it shorts out, it may possibly draw sufficient current to trip the breaker or blow a fuse. A derailment may damage a spring or something else in the direct line of the derailed wheels, which will trip the breaker. In a loco with a properly installed TVS, the board(s) should be protected from any voltage spike caused by a derailment. I am not aware of any overcurrent failures of electronics in normal use. I am not sure, but I do not think a stalled motor will burn out a motor driver board, or for that mater a shorted speaker and a sound board. Correct me if I am wrong.

Larry

Done. Here is a link to the old QSI website advertising their product, and here is a link to their report. Enjoy.

Larry

Actually, Lionel did NOT put one in the TMCC Direct Lockon, an omission that somewhat surprised me.  I opened one up when this discussion came up previously, and there isn't one in there.

I would be looking for better breakers if that happened to me. Don't know what you have for transformers or breakers, but the few breaker specs I have looked at, the trip times are pretty slow without a lot of extra current. They have to get to 200%-400% of their rating to get down into the 1 to 15 second trip range. At 135% of their rating they trip in an hour. I have never tested any of these personally, just reading the specs. 

GRJ,

Do you know what Lionel uses for the fast breakers in the TMCC lockon's? Those are pretty fast from what I have heard. I have not seen one in person.

John,

can you point us to the study that says a majority of failures are due to circuit breaker/fuse deficiencies?  To paraphrase someone here...

I don't have to. I'm stating an opinion, not quoting facts.

There have been several done. Two of which I posted about and a third posted by someone else. Why not read them and then comment.

I read the QSI document that was posted.I did not see anything stating more boards died due to lack of TVS protection than due to overcurrent.

Do not misconstrue what I'm saying. I have an opinion and, until I see an authoritative statement from a respected source with facts to back it up, I see no reason to alter my opinion:

More electronics are damaged due to overcurrent situations than due to transient voltage surges.

That could be. However, I believe more electronics are damaged due to high voltage spikes(caused by the collapse of continuity in a circuit) and their cumulative effect than due to overcurrent situations.

I'm not sure what you mean by a surge.

I'm sure we'll never settle which is the majority cause but it would seem sensible to use TVS's to protect against spikes, even if it is much less frequent than other causes as it is easy to do and has no down side.

Scotie

We tried that once on our modular layout when we had issues with multiple large passenger trains.  After several derailments during a show that didn't trip the breakers, but rather was doing some welding, I took the second brick off.  I refuse to run my equipment on the layout with those in place now.

The breakers are fast in the TMCC Direct Lockon, but they're similar in design to the breakers in the PH180.  Neither has a TVS installed.  Since I have the schematic of the PH180, I attached it to this post. 

The TMCC Direct Lockon also kill DCS signals dead, so I've stopped using them.  I feel the PH180 and a TVS in the TIU are sufficient for track protection.  Some, but not nearly all of my locomotives have TVS protection installed.  I have a couple that are in the backshop now for work, they'll get them before they leave.

 You are most certainly 100% correct IMO.  However, after having worked many years in an industry that regularly tests a piece of equipment to death, I can tell you that transient protection is very important.  Admittedly, we were testing for a very harsh environment, but then some aspects of the O-gauge environment are a pretty harsh environment as well.

I plan on continuing to use and recommend TVS protection, both at the track power level and in electronically operated locomotives and rolling stock.  Obviously, others are free to believe and do what they think best.

I still have a PowerGuard on each of my railpower districts[photo below] but anymore they are basically just for observing the flashing voltage spike indicator lights. I have TVS installed at the secondary of the 180 PoHo transformers and downstream on the layout at Railpower Distribution T-strips in each power district.[operating TMCC and Conventional]. I am of an age that produces hand tremors so opening locomotives is no longer an option. 

After dismantling a larger layout I donated one of of my surplus QSI PowerGuards to EE Dale Manquen for performing an autopsy of its voltage spike clamping capability versus Transient Voltage Suppressors. After his investigation Dale opined that the PowerGuards had comparatively insufficient voltage spike clamping capability---did not measure up to that of a simple inexpensive TVS.

Several years ago, I first learned of TVS from Bob Nelson, EE while posting on another Magazine's Forum. He was a big advocate of utilizing TVS on post war transformers , including the pw ZW. I was defending the purchase of my 5 QSI PowerGuards [$79 each] when he advised that I could accomplish effective voltage spike protection with the TVS and save about $78 per unit. It is difficult to argue based on ignorance so I eventually followed Nelson's advice regarding the use of TVS.

I cannot speak for EEs experienced in the characteristics of model railroad electronics such as Gunrunner John, Dale Manquen or Bob Nelson. But I have relied variously on their evaluation and comment regarding the nature of voltage spikes and overcurrent surges. Some opine that unlike an overcurrent surge, a voltage spike will not trip a breaker or blow a fuse. Thereby being one of the reasons for installing TVS to clamp the spike---the breaker/fuse doesn't interrupt it.

Also, unless you are using the 180 PoHo, with its very quick-acting electronic breaker, I believe fast blow fuses are better for reacting to control overcurrent surges than the typical 0-15 second trip time range of inexpensive thermal breakers such as the $4-5 Potter-Brumfield [which I used on two throttles of three pw ZWs for most of the 1990s]. Of course in my opinion the answer as regards external breaker use is purchase of more expensive magnetic units having rapid trip time.

My 2 cents.

The TMCC Direct Lockon also kill DCS signals dead, so I've stopped using them.  I feel the PH180 and a TVS in the TIU are sufficient for track protection.  Some, but not nearly all of my locomotives have TVS protection installed.  I have a couple that are in the backshop now for work, they'll get them before they leave.

 

It took me quite a while to figure out why one would even need a TMCC lockon. A lot of the Lionel stuff seems either complicated or redundant and explanations weren't always clear from the catalogs. They had so much stuff compared to MTH and DCS. I am learning though, now that it's all changing to LCS.

Some of the recent posts with the rubbing and pinched wires have me thinking about starting to look at the innards of my engines. Some may be getting a TVS soon.

Thanks for the schematic, I might even be able to cipher some of it. Fun to look at anyway.

Whether it is over current or voltage spikes that  kill digital control boards in trains, one cure for over current from track shorts is a fast acting circuit breaker. The fastest one I could find was the AirPax T-21-61 series circuit breakers. They actually have a chart on the Sensata web page for all the circuit breakers and the one that appears to give us the most protection would be the "fast" curve. It can trip in 100 milliseconds or up to one second. But you need to review the curves to get a less general spec that this.

http://airpax.sensata.com/pdfs/snapak.pdf

The Circuit Breaker in my MTH 100 watt power brick was so slow that the TIU fuses would blow before  the power brick CB would break. Those are 20 amp fuses by the way and a bugger to get out as you have to take the sheet metal box cover off to get to them. This CB is as fast or faster than the Lionel180 watt power brick CB.

LDBennett

No. You'll have to research it. I think the link is for a DC CB. You have to use a FAST trip AC magnetically tripped CB.