Skip to main content

Recently, there have been a few discussions about running A-A configurations with two powered units and four motors with the Cruise Commander.  It is alleged by no less than the man that designed the Cruise Commander, Jon Z., that the major issue that might compromise this issue is overheating of the bridge rectifier. Apparently, this degrades the rectifier and increases the leakage currents through it enough that it compromises the FET drivers and subsequently kills them.  Obviously, we want to prevent this!

I also note that the MTH boards actually heatsink the rectifier and not the driver FET's, that seems to make a lot of sense based on this information.  Also, a number of larger Lionel locomotives have a DCDS that also has a tethered bridge that is bolted to the frame for heatsinking.  It would have been nice if the Cruise Commander had mounted the bridge under the PCB and secured it to the heatsink as well, but we can't have everything.

Given this large body of evidence about keeping the bridge cool, I decided to modify the Cruise Commander that I'm installing in an A-A configuration with both units powered to give it a tethered bridge.

ERR Cruise Commander with Tethered Bridge

The hardest part of the job was getting the bridge off the board without damaging the board.  The bridge actually manages to dissipate quite a bit of heat from the leads, apparently by design.  Given that the Rectron BR84 8A bridge is only 99 cents, I decided that it wasn't worth trying to save the old bridge rectifier for the piddling amount of money it costs for a new one.  I ended up removing the driver FET heatsink and folding the driver FET's out flat.  Then using the Dremel tool I chopped up the bridge to remove it so I didn't damage the PCB.  It was then easy to remove the remaining leads and solder my jumpers from the replacement bridge to the PCB to complete the job.

The bridge will be bolted to the chassis with heatsink compound and a flat washer on top to maximize heat transfer.

Attachments

Images (1)
  • ERR Cruise Commander with Tethered Bridge
Last edited by gunrunnerjohn
Original Post

Replies sorted oldest to newest

I'm driving all the motors with one CC, they will all be wired in parallel.  I've never bothered to do this for one or two motors.  This came out of a conversation about this topic where Jon Z. chimed in and explained that the bridge was the limiting factor and when it gets hot it degrades enough that the leakage currents affect the driver FET's and can kill them.  Since I have had several CC units returned to me for repair with bad FET's, that was interesting news.  This will become my standard fix if this all works out as I expect it to.

John,

Will it require a 4 pin or 10 pin tether?  As with MTH B units, would it be a pass through tether?  If one CC can do this, can you also run a RS by tether in the B unit by adding data wires?

I can see where cooling on the FET's is beneficial, especially with lack of cooling inside a locomotive.

Ron

Last edited by Ron_S

Well, since I'm going to be sending lighting and couplers back to the rear unit, it'll require the 10-pin tether.  I have eight of them planned for use, who knows what else might come up.

  • 2 - power
  • 2 - motor
  • 1 - front light
  • 1 - rear light
  • 1 - cab light
  • 1 - coupler

You could run serial data back to have another RS board.  But since they'd be getting the same data at the same time, I'd just use a second speaker.

Last edited by gunrunnerjohn

Ok, I have come to an emergency stop on my ABA conversions, if the tether works so only one CC is needed and tether between the 3 units, I will retrofit tethers and remove the CC's in the trailing units.

I feel like many here, overwhelmed and undereducated. I had a good handle on MU with CC in every unit, but once I watched Pete's video on his K-Line ABA E-8 with one Commander, it is definitely the cost effective way to go, the added benefit, I now have 6 extra CC and can convert more of my conventional fleet.

Ron

John I am driving four motors using one DCDR and also with a TAS SAM board with good results I did beef up the sinks for the triacs and the bridge on both.  The DCDR is in a Lionel 18953 Alco PA-1 which has Mabuchi 545 motors. Yeah I'm driving four of them with one DCDR. My brother has my layout in his basement and I left the Alco AA set there. He is not so restrained with the throttle as I am and has been pulling 12 Lionel NYC 72' cars with the AA Alcos for almost three years now so if anything was going to fry he would have found it. The TAS SAW board is in a MTH UP Veranda turbine he is just as heavy handed with it though he's only had it for about five months. So I know driving four motors on at least two different non cruise driver boards works.  Now the question.  I have four Cruise Commanders and have been going back and forth with whether or not to try what you have done on a MTH centipede AA but I was contemplating going to a 12A Bridge. Much of my hesitancy has been whether or not the curves, leakage, PIV etc of the new bridge must be an exact match, close match, or not to worry, with the original 8A bridge for the board to function properly.  I am wondering if you gave a larger bridge some consideration when you broke it out and what factors swayed  your decision to stay with the 8A bridge. My experience with MOSFET PWM drivers is limited but my impression is they are fussier than triacs .        j

Last edited by JohnActon
@Ron_S posted:

Ok, I have come to an emergency stop on my ABA conversions, if the tether works so only one CC is needed and tether between the 3 units, I will retrofit tethers and remove the CC's in the trailing units.

I feel like many here, overwhelmed and undereducated. I had a good handle on MU with CC in every unit, but once I watched Pete's video on his K-Line ABA E-8 with one Commander, it is definitely the cost effective way to go, the added benefit, I now have 6 extra CC and can convert more of my conventional fleet.

Ron

Ron, I hope to try out my UP ABAs next week on the club layout pulling a load. I will have more info a week from now. I would hold off until I know if this experiment will actually work in practice.

Pete

@JohnActon posted:

I am wondering if you gave a larger bridge some consideration when you broke it out and what factors swayed  your decision to stay with the 8A bridge. My experience with MOSFET PWM drivers is limited but my impression is they are fussier than triacs .        j

I think keeping it cool is the key difference, I not only bolted it to the chassis, I did actually use real heatsink paste to help with the heat transfer.  I happened to have the 8A bridges on hand, that's why I used them.  Like I said, I initially thought I'd just unsolder the existing bridge and use it, but when it resisted removal I decided that it wasn't worth the risk of wrecking the board to try to save it.

I agree about the FET vs triac, I suspect the FET's are touchier as a rule.  OTOH, I use them in a host of switching applications and have had minimal issues with them dying.

I also suspect that it's only in extreme cases that the rectifier gets hot enough to be an issue, normally a four motored A-A would not be drawing more than a couple amps at most under load.

@Ron_S posted:

@gunrunnerjohn   John, Could you post a few pics of a modified commander in a unit?  Also, where is the best source for the paste for the heatsinks?

Ron, look at the first post of this thread!

As for the heatsink compound, it's available all over.  Here are a couple of choices at Digikey.

CAIG HSC67-6G

Gelid Solutions TC-GC-03-D

You can find tons more on other sites like eBay, AliExpress, etc.

@Ron_S posted:

For the novice like myself, how durable are the pins on an FET? Can they be bent several times as do the removal of the BR and complete the tether?

You can bend them a few times, but it only takes two.  Bent them close to flat to chop up the bridge and install the tether wires, then bend them back.

I'll also short circuit the inevitable comments about "why didn't you remove the original bridge and use it".  I did briefly try a desoldering on the bridge, but when it became apparent that I'd need more heat than I'm comfortable using on a PCB to fully loosen it for removal, I decided to not risk damaging the PCB for a cheap bridge rectifier.  YMMV, but I like my method to minimize the possibility of wrecking a $100 board.

@Ron_S posted:

@gunrunnerjohn  Thanks John, is it better to use a thicker washer about the size of the BR top?

The washer has to be smaller than the bridge for obvious reasons, the connections are on the corners!  The washer is really to provide an even pressure to the bottom mounting hole boss instead of relying on the nut that might put pressure on one side and crack the plastic.  I think I used about a 3/8" diameter washer, it was "big enough".

@Lou N posted:

Norm,

I have one of these in the queue for upgrade.  What did you use for the antenna?

Regards,

Lou N

Whatever TAS used to supply with their EOB upgrades. My set had EOB for many years before I swapped it over to ERR. I was fighting what turned out to be a grounding issue and found the Cruise Commander was less picky about the Centipede set having poor grounding. I added some ground jumpers to the trucks at some point even after converting over. I have long insulated block triggers for my signals and some diesels hate it. The bad grounding clue is if they run better when the train they are pulling enters the block and bridges the ground.

For what it’s worth I could never pull a load hard enough to annoy the standard rectifier and mosfet/heat sink setup with EOB or ERR. YMMV though.

For what it’s worth I could never pull a load hard enough to annoy the standard rectifier and mosfet/heat sink setup with EOB or ERR. YMMV though.

According to Jon Z. it's not that you truly overload it, but just get it hot enough to increase the leakage current, that's what does the FET's in.  You're not actually overloading the FET's max current.

I think the old Lionel DCDRs used the same FETs as the ERR stuff?  I’d have to go look though. I have a few in the stash. Anyone remember how those DCDRs used to slam motors to a reversed voltage stop at high speed/direction change? I have a project on the bench right now with one. Brought back memories!

Have you guys ever tried getting those small extruded aluminum heat sinks for the rectifiers?

I have a few heat sinks I have pulled off old computer CPUs and video driver chips. I didn't use one on the Cruise M rectifier but have used them on an E unit used in a conventional Lionel GG1 that was prone to overheating.

Lot of useful parts in old desktop computers. Good idea to strip them out before recycling.

Pete

@Norton posted:

Just a reminder, My K-Line ABA had six motors driven by a single non cruise DCDR.

I would think that board is just as suseptable to overheating and failure as the ERR Cruise boards if that was an issue. Then again maybe not. Maybe Jon Z knows the difference or how K-Line got away with it.

The DCDR uses triacs, a totally different animal.  The FET's used on the Cruise boards would be a lot more susceptible to the reverse leakage current and voltage from the rectifier.  I'm sure that Jon does indeed know the difference.

I’m also of the mindset that having more than 10amps available at the track is in arc welding territory.

A tale told to me many years ago by a Youngstown Model Railroad member (2 rail DC powered by car batteries)...One night someone dropped a nail and it went across the tracks and it did weld.  Available instantaneous current will get you every time.

I designed my own O scale DCC decoder and used 40A MOSFETS.  Purely a matter of convenience with no expectations of needing that amount of current.

Lou N

@Lou N posted:

Yes.  Bergquist makes paste type heat sink grease but also makes film type.  Some sheets are adhesive on both sides allowing the attachment of small heat sinks.  Copper is a better choice of sink material.

Lou N

Lou, that is what I thought for years.   About  3-4  years ago I bought some finned copper stock and similar aluminum finned stock and made some sinks for the triacs and bridge on a pair of DCDRs  for which I was planning on driving four 545 motors with one DCDR in a pair of Lionel 18952 Alco PA-1s. I wanted to rob one of the DCDRs for use in a Williams Trainmaster with twin Pittman motors. The copper sinks are still working well however I am not sure there is a clear advantage, as copper will absorb heat faster aluminum will dissipate it faster. I cut the finned copper and aluminum into roughly equal shapes and installed them on two identical DCDRs and A-B running the four motored AA pair 15 minutes at as close to the same speed as I could on a circle of O-72. Running in conventional with the track voltage set on 14v.  After the fifteen minute run I stopped and as fast as I could point my digital IR thermometer at the case of the triacs  I read the temps first with the copper finned DCDR then with the aluminum version they were less than one degree apart.  I figured that was the limits of my test setup. Then I ran across an article on the web where the writer was discussing this very topic and his conclusion was what I had decided from my experiment.  I was trying to find that article tonight but before I did I ran across a related discussion about automobile radiators Aluminum vs Copper.   https://www.researchgate.net/p..._aluminium_or_copper  Read about  1/3 down the page and look for the name Ammar Eqbal, Indian Institute of Technology Patna where he says the best heatsinks are a composite with a copper base and aluminum fins.  I am glad I did the test but to my way of thinking it was a draw.          j

Last edited by JohnActon

Keeping it cool to minimize the leakage current is far more important than the actual rating of the bridge.  8A is plenty with the heatsink, why make it larger for no apparent gain?

A 10 amp bridge will generate less heat than a 8 amp, won't it?  At a given current.  If these things are getting that hot that they are breaking down. even heat sinked to the chassis, you'd figure I higher current rated bridge would be less likely to breakdown.

Last edited by superwarp1

Ever try wiring the motors in series in each unit? I've always wanted to try that with an ERR CC but never have for some reason.

All that will do is greatly reduce the pulling power as each motor will not be able to generate maximum torque at the lower voltages available.  I have actually tried it, but that was the only effect I saw.  I think the right way it to manage the cooling of the rectifier in addition to the FET's.

@superwarp1 posted:

A 10 amp bridge will generate less heat than a 8 amp, won't it?  At a given current.  If these things are getting that hot that they are breaking down. even heat sinked to the chassis, you'd figure I higher current rated bridge would be less likely to breakdown.

Actually, given similar voltage drop specifications across the diodes, it'll generate exactly the same amount of heat, think about it.

Most modern dual-canned engines seem a bit overpowered anyway. Combined with relatively steep gear ratios, the real idea behind wiring motors in series (in each unit) would be for smoother control with some level of acceptable loss of speed pulling power. I never tried this with CEMF control but it might be worth experimenting with next time I open up my Centipedes.

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.
×
×