Paul_Rimmer Posted January 2, 2018 Share Posted January 2, 2018 I'm new to DCC and have a Hornby Select controller. I have one DCC locomotive and two DC locomotives at the moment so not trying to run more than three as suggested. The problem I have is irrespective of the number of locomotives running. It appears to be a random event but every so often, the system overloads and the controller displays OL. It doesn't appear to depend on a locomotive being at a specific place on the track. If, when the system overloads, I pull the power cable out from the back of the controller and then plug it back in, the locomotives just start up again as if nothing had happened. If the overload was caused by a short, I would have expected the overload to have happened again.I would welcome any suggestions as to how to identify the fault. I don't particulary want to buy the 4v transformer for the Select unless this is the only option. Link to comment Share on other sites More sharing options...
96RAF Posted January 2, 2018 Share Posted January 2, 2018 If Select overloads as in too many amps being drawn when using the 1 amp power unit then self protection is by way of a voltage drop at the wall plug device and the Select will reboot over and over (start screen sequence will show time and again). A short is the usual overload condition but too many locos will also pull up the OL flag then reboot scenario. If there is definitely no short and its only one loco causing the problem then I suggest taking a stall current reading of that one using a DC supply and a multi-meter that will read more than 2 amps DC - many meters only have 200mA capability and are not suitable for the test.Rob Link to comment Share on other sites More sharing options...
Chrissaf Posted January 2, 2018 Share Posted January 2, 2018 Are you trying the run the two DC locomotives on the DCC track using the Select. Either together or one at a time..Regardless of what the Select manual might say regarding support of a DC Analogue loco on a DCC controlled track.......don't do it..If you continue to do so, you really do risk overheating and burning out the armature coils of your DC Analogue locos..Also the DC Analogue support feature is only designed to attempt** to operate a single DC Analogue loco at a time, NOT two..You really must fit a DCC decoder to the two DC Analogue locos before running them via your Select..Note** I used the word "attempt" on purpose. DC support on DCC is a legacy left over from the early days of DCC. The track voltage on DCC is an alternating square wave with a peak to peak voltage of about 28 volts. When you select address Zero (for DC support) all you are doing is 'stretching' the Zero bits in the digital waveform to create a positive or negative voltage bias in the alternating voltage waveform. The DC loco reacts to these bias voltages as best it can to go forward or backwards. Much of the power in the DCC waveform is left to alternate back and forth in the motor coils, generating heat but without movement. This heat build up can be audibly detected by the motor giving off a high pitched whine or buzzing (I assume you can hear this on your DC locos when you place them on your DCC track)..As I have just said above, DCC Address Zero is a legacy support feature and not fit for purpose on modern locomotive motors. Please, please, please fit a DCC decoder to your locos, without one, they will fail, maybe not today, maybe not tomorrow, but they will fail eventually if you continue to run them natively on DCC without decoder modification..The extra current they are drawing (Select showing OL) may be an early sign that they are beginning to fail already. Link to comment Share on other sites More sharing options...
96RAF Posted January 2, 2018 Share Posted January 2, 2018 I missed the 2 x DC locos. No doubt that would cause the problem.Rob Link to comment Share on other sites More sharing options...
Paul_Rimmer Posted January 2, 2018 Author Share Posted January 2, 2018 Thank you both for your comments. I note what you say Chris about the use of DC locos on a DCC layout. This seems to be diametrically opposed to what the Select manual says which is that the high pitched noise a DC loco emits is normal. I will though fit a decoder if possible. One of the locos is the West Coast Highlander 0-4-0 locomotive which is a great little loco but I don't know if there is room to fit a decoder. The other is a Thomas and I have seen the instructions on the Hornby website for fitting a decoder. I was going to use the standard decoder from my DCC loco (R3086 Flying Scotsman) and then replace that one with a sound decoder. I think R8106 TTS Sound Decoder - A1 & A3 Class is the right one.Nevertheless, I have tried running just the DCC loco (R3086 Flying Scotsman) on its own with the two DC locos isolated and I still get the same problem - the loco stopping for no apparent reason. Rob, I don't know much about the electronics so I'm not too sure how to take a stall current reading. Is there more information on this somewhere? Link to comment Share on other sites More sharing options...
Chrissaf Posted January 2, 2018 Share Posted January 2, 2018 When the Select manual says that the high pitched noise is normal. That is a little bit misleading. What it is really saying is that one should expect to hear a high pitched noise. The fact that you can hear it, means that the motor is being stressed. The noise is generated by the coils in the motor armature vibrating because a high frequency alternating current is passing through them. If they are vibrating, but the motor isn't turning, then there is current flowing in the windings that generates heat that can not be dissipated via air flow generated by the armature spinning. Also, the vibrations themselves generate heat. It is like rubbing two sticks together to start a fire. The loco motor itself is designed to be operated on DC and not an alternating current. As an analogy, a stationary car with the engine running with no cooling fan..If you were to poll the experienced members on this forum, they will almost certainly concur that it is not a good idea to use the 'Address Zero' feature even if supported by the controller and documented in the manual..In life, just because you can do something.....doesn't necessarily mean that it is a good idea..With regard to your supplementary comment regarding 'stall test'......see this link.http://www.brian-lambert.co.uk/DCC.html#NORMAL Link to comment Share on other sites More sharing options...
96RAF Posted January 2, 2018 Share Posted January 2, 2018 Brian Lamberts site tells all about a stall test, but I think the problem is running 2 x DC locos on a DCC track, so probably not necessary at this time. Heres a picture to help with the 0-4-0 conversion which is also covered on the main Hornby site DCC conversions page. You can see I installed a socket to make installing a decoder easier, whereas the Hornby conversion is hard wired. Rob /media/tinymce_upload/2678c2bdf595d9db2ba43feef2622960.jpg Link to comment Share on other sites More sharing options...
Fishmanoz Posted January 2, 2018 Share Posted January 2, 2018 To add to what Rob and Chris have said - on analogue control, the Select manual starts with the words “It is possible but not recommended ....” which isn’t exactly a “ringing” endorsement of the practice. I agree you run a high risk of blowing the motor. If the Select goes into OL running only FS, this is not normal, you are almost certainly get a fault condition from a temporary short circuit. You say there is no pattern to it but it wouldn’t happen to occur on occasion when crossing points or a crossover? If that is the case, the chances are the back-to-back measurement of at least one wheel set on FS is incorrect and shorting as it crosses the points. While looking at the Brian Lambert site as recommended above, look at his coverage of this and how to adjust back-to-back. Note it may not happen on every traverse, only sometimes. And it may be an issue with one or both of your DC locos too. Link to comment Share on other sites More sharing options...
Paul_Rimmer Posted January 2, 2018 Author Share Posted January 2, 2018 Thanks for now. I'll see how I get on with just running a DCC loco. Link to comment Share on other sites More sharing options...
Chrissaf Posted January 2, 2018 Share Posted January 2, 2018 Based upon Fishy's post above, I went looking myself for the relevant statement he quoted.......see below:./media/tinymce_upload/6565494697ac894e3a75d77900390a45.jpg.Paul,This manual extract was copied from the current 1.5 Select manual. If your copy of the manual doesn't have this sentence in it (yellow highlight) then you may have an earlier manual version..EDIT: Just checked the earlier Select manuals, they all have the quoted sentence in them. Link to comment Share on other sites More sharing options...
Fishmanoz Posted January 2, 2018 Share Posted January 2, 2018 Paul please bear with me while I have a short discussion with Chris and Rob about analog operation. Chris made this statement in his first post-“Much of the power in the DCC waveform is left to alternate back and forth in the motor coils, generating heat but without movement.” I don’t think that the much power part of this can be supported. A motor is a highly inductive load and so has impedance increasing with frequency. In fact the value is 2*pi*f*L where f is the frequency and L the inductance in henries. Given the high impedance at DCC frequencies, very little current is drawn and very little power dissipated. All the motor reacts to is the average DC value of the signal. Now let’s consider the “stretched zero” DCC signal applied to an analog motor. In fact, we are talking a PWM signal here, so how is it different to the PWM used to drive the motor running DCC, or a PWM analog controller? I can hypothesise a couple of differences. First, the voltage is higher on the DCC track than is the DCC motor drive. Second, the condition is worst when the loco is stationery as only one or two motor coils will be receiving current continuously and with no cooling from rotating, as opposed to all coils alternatively while moving. Is a third difference that DCC motor drive is switched off at speed step zero or stationery? Even if it is, what about speed step 1 which will likely not move the loco? any thoughts on other differences which might explain damage with an analog loco on a DCC track but no damage in the straight DC or “proper” DCC application? Link to comment Share on other sites More sharing options...
Fishmanoz Posted January 2, 2018 Share Posted January 2, 2018 My Select manual is a download from at least 12 months ago. Link to comment Share on other sites More sharing options...
Chrissaf Posted January 2, 2018 Share Posted January 2, 2018 Chris made this statement in his first post-“Much of the power in the DCC waveform is left to alternate back and forth in the motor coils, generating heat but without movement.”.I was just trying to keep it simple...... Link to comment Share on other sites More sharing options...
Fishmanoz Posted January 2, 2018 Share Posted January 2, 2018 Sorry Chris, I don’t think simple can hack it here. Link to comment Share on other sites More sharing options...
96RAF Posted January 2, 2018 Share Posted January 2, 2018 FishyI bought a book when in uk Electronics for Model Railways and it explained the DC motor versus DCC PWM output very well, including how frequency and magnet strength can affect the noise a motor makes and if the armature will be liable to fore/aft chatter. The number of poles was also a feature along with armature core lamination materials. Unfortunately the book is sitting on a table in UK and I am back in Cyprus. Nor can I remember the details well enough to quote chapter and verse, but the end game reached about DC locos on DCC track was ‘don't do it’.Rob Link to comment Share on other sites More sharing options...
Chrissaf Posted January 2, 2018 Share Posted January 2, 2018 Although neither of these resources adequately explain the scientific physics of exactly why the DC motor in a loco generates heat when put on a DCC track, they do however both concur with the theory..The first images are scans from my "Digital Command Control" book by Stan Ames ISBN 91-85496-49-9/media/tinymce_upload/8f3d650dfdfa1053e909c88fdb7045b1.jpg/media/tinymce_upload/9c566e94d611d442abceb2e1f3887a40.jpg .The second image is an extract from the following webpage:https://sites.google.com/site/markgurries/home/technical-discussions/dc-loco-on-dcc/dc-locomotive-damage-risk./media/tinymce_upload/63119207117de6f36fa95ae88b2af845.jpg.Mark Gurries (who wrote the text above) is a member of the NMRA, it is possible that he too is simplifying the theory (like I did), but he does state that the heat is generated by DCC current flowing in the motor.. Link to comment Share on other sites More sharing options...
Fishmanoz Posted January 2, 2018 Share Posted January 2, 2018 I agree with all of that. My question is how is the DCC motor drive signal different when the motor is stationary. PS. Oops, previously I have been saying motors not moving are pens and paper! Link to comment Share on other sites More sharing options...
Chrissaf Posted January 2, 2018 Share Posted January 2, 2018 Afraid I can't offer an answer on that...... Link to comment Share on other sites More sharing options...
2e0dtoeric Posted January 2, 2018 Share Posted January 2, 2018 I may be entirely on the wrong track (sorry!) but as I see it, the pulsed square-wave voltage is (say) positive at full voltage, then crosses the zero volts, and is then negative at full voltage. (Chrissaf's drawing is a bit misleading).So with a loco moving in one direction (talking about a dc motor), it sees a long positive 0, and short 1's, a loco moving backwards sees a long negative 0 and short 1's. (pseudo unbalanced AC).A stopped loco will see long 0's and short 1's of alternatively pos and neg of the same duration, (balanced pseudo AC) so each pos 0 will rotate the armature a small amount, but then a neg 0 will rotate it back again, so one pole of the armature will be energised alternately pos and neg, making it rock back and to. Due to the frequency of the square-wave, the motor will appear to squeal, and the one almost static energised pole will heat up, as with no true rotation and 'rest period' it cannot dissipate the heat generated. Additional current loading will occur, because the armature will constantly be stopping and restarting in the other direction, and as you are aware, starting/accelerating consumes more power (and creates more heat) than a constant 'cruise'. - And, if the armature has a mounted flywheel, there is even more mass to keep reversing!edited for a typo Link to comment Share on other sites More sharing options...
96RAF Posted January 3, 2018 Share Posted January 3, 2018 The above explanation by Eric is valid because a DC loco has its motor connected directly to the track, even if DCC Ready (by way of the shorting plug), so it feels the full DCC waveform even at zero throttle, whereas a DCC motor is isolated by the decoder until throttle is commanded.Rob Link to comment Share on other sites More sharing options...
96RAF Posted January 3, 2018 Share Posted January 3, 2018 Going back to the original snag where it is confirmed the FS did cause an OL indication using the 1amp PSU. This should not happen as I explained earlier - the self protection is by way of the 1amp dropping voltage and thenSelect rebooting until the overload is sorted. To get an OL indication you are asking the Select to carry out the protection and this can only happen if the 4amp PSU is used. It follows that either the 1amp PSU is not dropping voltage and/or the Select has a fault causing the OL to kick in at much lower amps. I would suggest a call to HCC.Rob Link to comment Share on other sites More sharing options...
Fishmanoz Posted January 3, 2018 Share Posted January 3, 2018 Again agreed. But again tell me how this is different to then applying DCC throttle at speed step 4 when Vstart is speed step 5. So full DCC motor volts are being applied and the motor is not rotating. Link to comment Share on other sites More sharing options...
96RAF Posted January 3, 2018 Share Posted January 3, 2018 Fishy if we are talking about a decoder equipped loco then I don’t think the motor will see any voltage until Vstart kicks in, then it will get speed step 5 voltage and go. I don’t know about the practicality of hanging a multimeter off the motor terminals to prove there is no voltage supplied until Vstart provides it. Rob Link to comment Share on other sites More sharing options...
Fishmanoz Posted January 3, 2018 Share Posted January 3, 2018 Let’s assume an 8249 which doesn’t allow for setting Vstart but, if able to set it, it is 5, and the throttle is at 4. The decoder must be supplying volts at this stage as it knows no better. The only saving grace I can see is that the throttle is unlikely to be left like this for long, unlike a DC loco sitting in a live DCC siding. Link to comment Share on other sites More sharing options...
2e0dtoeric Posted January 3, 2018 Share Posted January 3, 2018 Who is going to volunteer to 'take the lid off', put the loco on a rolling road, hang a meter on, and see?(Phut the magic dragon - ) Link to comment Share on other sites More sharing options...
Recommended Posts
Archived
This topic is now archived and is closed to further replies.