RMweb Gold Donw Posted August 25, 2014 RMweb Gold Share Posted August 25, 2014 I would imagine that one would want to turn the sound off once the loco is on the cassette one would really want locos on cassettes chuffing away or idling diesels. The next question is does the capacitor charge from the motor current or from the DCC supply. If it charges from the motor current then the question is how long will it hold the charge or does it just discharge straight away through the motor once the decoder switches off motor power. If the latter then the problem will apply to any form of fy where locos are turned off via the decoder. If however the charging is done from the DCC supply putting a loco onto a cassette for a minute and applying the DCC before it sets off should assist. A simple test would be to take a loco place it on the DCC track but not drive it remove from the dcc supply and instantly drive it forward if the wheels turn somewhat charging has occurred from the DCC. So the answer would be to have the loco ready on the cassette with DCC power for a short period before it enters onto the layout. I must try this out a some stage. Don Link to comment Share on other sites More sharing options...
Simond Posted August 25, 2014 Share Posted August 25, 2014 Chaz I bought my ESU decoders from Howes - (actually MrsD bought them for my birthday) - got their sound projects for the 28xx and the King. There is an interesting part of the manual that explains how to "auto-tune" the chip settings to the motor and gearbox - I did this, and then adjusted the settings to synchronise the "chuff". As I said before, tried a 2200 cap and it was pretty much useless, my guesstimation/calculation gave me a "stay alive" duration of 0.02 seconds, so I went away and did what I should have done in the first place, fixed the pick-ups. From February to here, finished and painted the 28xx, and put a little time into the King, but I have not followed up on the stay-alive. Now I have got the Fowler back from Paul, and seen first-hand how very impressive it is, I will put some priority on doing so. I will order some caps today and report further when I get a chance. The relevant bit of the ESU manual is at section 11.1.3, on page 48. Don't know if this applies only to V4. It is exciting as your pride and joy will accelerate like a startled fawn, and equally suddenly lurch to a stop. With your Peckett, this is probably not warp-speed, but it was terrifying with the King which has a Maxon motor and an ABC gearbox as the Greater Windowledge is not really very long! HTH Simon Link to comment Share on other sites More sharing options...
chaz Posted August 25, 2014 Author Share Posted August 25, 2014 I would imagine that one would want to turn the sound off once the loco is on the cassette one would really want locos on cassettes chuffing away or idling diesels. The next question is does the capacitor charge from the motor current or from the DCC supply. If it charges from the motor current then the question is how long will it hold the charge or does it just discharge straight away through the motor once the decoder switches off motor power. If the latter then the problem will apply to any form of fy where locos are turned off via the decoder. If however the charging is done from the DCC supply putting a loco onto a cassette for a minute and applying the DCC before it sets off should assist. A simple test would be to take a loco place it on the DCC track but not drive it remove from the dcc supply and instantly drive it forward if the wheels turn somewhat charging has occurred from the DCC. So the answer would be to have the loco ready on the cassette with DCC power for a short period before it enters onto the layout. I must try this out a some stage. Don looking at where the wires from the PowerPack are soldered on the decoder PCB I think the charging is from the DCC supply side. It can't possibly discharge directly through the motor can it? This would give the decoder no control over speed or direction and would produce no supply to the sound. At the moment I am not too concerned about the charging rate - from my experimenting it appears to be a few seconds not two minutes. What is far more worrying is that... The performance is very poor - there is lots of motor rattle - and pull away and stops are jerky with the loco sometimes leaping forward. The keep-alive doesn't seem to do what it's supposed to, although the poor performance may be masking the effect. Just to explain again... The keep-alive is an ESU PowerPack - not a home-made affair - so I have no control over component values it appears to work to some extent in that... if the loco is lifted from the track when stationary the sound continues for about 4 seconds if the loco is lifted when moving the wheels revolve for what I estimate to be something between 0.5 and a second If ESU can't help me to get the setup to work acceptably I will be seeking a refund. At the moment I am rather frustrated - my Ixion Hudswell Clarke has Zimo decoder, without a keep-alive, and works so much better - it looks as though I will be be relying on this loco at Telford. Chaz Link to comment Share on other sites More sharing options...
chaz Posted August 25, 2014 Author Share Posted August 25, 2014 (edited) Chaz I bought my ESU decoders from Howes - (actually MrsD bought them for my birthday) - got their sound projects for the 28xx and the King. There is an interesting part of the manual that explains how to "auto-tune" the chip settings to the motor and gearbox - I did this, and then adjusted the settings to synchronise the "chuff". As I said before, tried a 2200 cap and it was pretty much useless, my guesstimation/calculation gave me a "stay alive" duration of 0.02 seconds, so I went away and did what I should have done in the first place, fixed the pick-ups. From February to here, finished and painted the 28xx, and put a little time into the King, but I have not followed up on the stay-alive. Now I have got the Fowler back from Paul, and seen first-hand how very impressive it is, I will put some priority on doing so. I will order some caps today and report further when I get a chance. The relevant bit of the ESU manual is at section 11.1.3, on page 48. Don't know if this applies only to V4. It is exciting as your pride and joy will accelerate like a startled fawn, and equally suddenly lurch to a stop. With your Peckett, this is probably not warp-speed, but it was terrifying with the King which has a Maxon motor and an ABC gearbox as the Greater Windowledge is not really very long! HTH Simon Thanks Simon, the auto tune facility looks like it might do the trick. Is the manual you refer to a on-line download? I got nothing in the box with the decoder and the keep-alive instructions are all in German - although the wiring details, being diagrams, are universal. Chaz PS - I assume from you mentioning ordering capacitors that you are doing a home-brew keep-alive rather then ESU's PowerPack. This is very small and fits into the Peckett's frames. Big capacitors wouldn't. The decoder fits diagonally inside the space in the middle of the frames - the keep-alive is in the triangular space nearest the camera. Edited August 25, 2014 by chaz Link to comment Share on other sites More sharing options...
RMweb Gold Donw Posted August 25, 2014 RMweb Gold Share Posted August 25, 2014 Chaz I think you are right and it is on the dcc supply side which I suppose is necessary to keep the sound going. I assume it charges up with a retified supply from the dcc signal and then the decoder runs as though on a dc supply when current is interrupted. I wonder if the autotune idea may be right it could be that the output from the decoder has some feedback and is misreading what the loco is doing. It does sound rather like a learner's kangaroo starts in a car. Don Link to comment Share on other sites More sharing options...
daifly Posted August 25, 2014 Share Posted August 25, 2014 If ESU can't help me to get the setup to work acceptably I will be seeking a refund. At the moment I am rather frustrated - my Ixion Hudswell Clarke has Zimo decoder, without a keep-alive, and works so much better - it looks as though I will be be relying on this loco at Telford. Chaz It will be worth taking it with you to Guildex and dragging Steph Dale off the Technical Committee stand - he's a real boffin on all things DCC and ESU in particular. It might well be worth contacting him directly in advance - he's a very helpful chap. Contact via http://www.stephdale.co.uk/ or http://www.euram-online.co.uk/contents.htm HTH Dave Link to comment Share on other sites More sharing options...
chaz Posted August 25, 2014 Author Share Posted August 25, 2014 Chaz I think you are right and it is on the dcc supply side which I suppose is necessary to keep the sound going. I assume it charges up with a retified supply from the dcc signal and then the decoder runs as though on a dc supply when current is interrupted. I wonder if the autotune idea may be right it could be that the output from the decoder has some feedback and is misreading what the loco is doing. It does sound rather like a learner's kangaroo starts in a car. Don Yup, exactly so. Also the motor rattles (grumbles, growls - take your pick of suitable words) throughout the speed range. Not good. Link to comment Share on other sites More sharing options...
chaz Posted August 25, 2014 Author Share Posted August 25, 2014 Chaz It will be worth taking it with you to Guildex and dragging Steph Dale off the Technical Committee stand - he's a real boffin on all things DCC and ESU in particular. It might well be worth contacting him directly in advance - he's a very helpful chap. Contact via http://www.stephdale.co.uk/ or http://www.euram-online.co.uk/contents.htm HTH Dave Thanks for that useful suggestion Dave. And yes it does help. I will certainly take the Peckett to Telford even if its duties are covered by the Hudswell Clarke. But before I contact Steph Dale I will see what a bit of auto tuning will do. Link to comment Share on other sites More sharing options...
Simond Posted August 25, 2014 Share Posted August 25, 2014 hi Chaz sorry if I misunderstood. You can, of course, put your own KA capacitors in your loco in place of the ESU part, but I doubt it is worth it. The KA are supplied from the diodes that supply the electronics in the decoder; as I understand it there are + & - rails within the decoder, powered by a bridge rectifier from the track and the KA keeps these at a potential for a period of time dependent on the capacitor values, and the loads (typically motor is biggest, then sound, then lights). The decoder uses further switching components to drive the various loads, from these rails. There are a number of people on RMWeb who know far more than I do and who would, I'm sure help, if my thoughts don't sort it. You might try to contact Suzie and (Andrew) Crosland. The KA should not make any difference to your motor performance per se, and from what you are saying above it is absolutely doing what it's there to do, it is keeping the motor (and/or sound) running for a short period when the loco is not getting any supply. So I believe the issue is with performance - and the rattle / jerky behaviour suggests that the pwm and load compensation is not working correctly. Assuming the decoder is not broken (a safe assumption I believe) I think you should start by looking at these; I don't have a manual for the V3.5, but in the V4 manual, it starts at page 46, section 11. Just to confirm my thoughts, you might try to turn off compensation and check that the locos run at least smoothly. Read CV56, make a note of it, and then set it to zero. this will disable your load compensation and it should then not jerk or rattle at all. If so, then you can try some of the settings (or the auto setting method that I already suggested) that they give in the manual. HTH Simon Link to comment Share on other sites More sharing options...
Simond Posted August 25, 2014 Share Posted August 25, 2014 ESU download website http://www.esu.eu/en/downloads/instruction-manuals/digital-decoders/ br Simon Link to comment Share on other sites More sharing options...
chaz Posted August 25, 2014 Author Share Posted August 25, 2014 The plot thickens (in fact it might have congealed....) Full marks to you, Simon, for the info that moved me on a bit. I downloaded the LokSound V4 manual and referring to section 11.1.3 - as Simon suggested - I came across the following.... "LokSound V4.0 decoders offer a feature for automatic calibration of the motor. I most cases this procedure leads to excellent load compensation characteristics. However, due to the infinite number of possible combinations of motors and gear boxes a good result cannot be guaranteed in all cases. But it is certainly worth a try.Proceed as follows:Set the vehicle onto a piece of straight and preferably level track. • This track must be at least as long as to allow the locomotive torun at full speed for about two seconds. This movement takes place automatically during calibration. Provide buffer stops or something similar to prevent the locomotive from running off the track!Call up the locomotive on your throttle and make sure that F1 is • switched off and the throttle is set to speed step 0. Set the direction of travel as desired.Write the value 0 into CV 54 (either on the programming track • or with POM).Now press the “F1” button. The locomotive will start running at • full speed automatically. During about 1.5 seconds you will not be able to control this locomotive!The locomotive will automatically stop and the identified load • compensation parameters will be saved in the CVs 51, 52, 53, 54 and 55.If you wish you can now further optimise the load control parameters starting from the calibrated settings." So I did that, using a spare bit of Peco track as a programming track to set CV54 to 0. I have never liked POM and avoid using it if possible. As I have two of Dock Green's boards set up for checking/adjustment/repair I was able to place the loco at one end and, with some trepidation, press F1. The loco did take off at warp factor 9 and ran about four feet (the length of one of the boards) before stopping as abruptly as it started. The result? Well the loco does run much more smoothly and the motor growl is reduced (but not eliminated). The sound, as it did before the auto tune, runs on for about 4 seconds if the loco is lifted from the track BUT the motor only runs briefly for a fraction of a second before stopping. No risk of the loco running amok off the end of a cassette, say. This suggests that there is still something wrong. Incidentally I can now confirm that it takes quite a time to fully charge the keep-alive capacitors - about two minutes seems right. This obviously doesn't stop the loco being used before the time is up. The fact that the run-on when power is removed is so short suggests that something is not right in the set-up. So efforts to improve matters will continue. WTS Chaz Link to comment Share on other sites More sharing options...
Simond Posted August 25, 2014 Share Posted August 25, 2014 Chaz check CV 113... When you connect an external capacitor or PowerPack, you are able to make the decoder switch off after a certain time. CV113 is responsible for that, since you are able to determine at what time the decoder is to switch off (as a multiple of 0.0164 seconds). You should set a time between 0.3 and 1.0 seconds to prevent your locos from driving too far during an emergency br Simon Link to comment Share on other sites More sharing options...
chaz Posted August 25, 2014 Author Share Posted August 25, 2014 hi Chaz sorry if I misunderstood. You can, of course, put your own KA capacitors in your loco in place of the ESU part, but I doubt it is worth it. The KA are supplied from the diodes that supply the electronics in the decoder; as I understand it there are + & - rails within the decoder, powered by a bridge rectifier from the track and the KA keeps these at a potential for a period of time dependent on the capacitor values, and the loads (typically motor is biggest, then sound, then lights). The decoder uses further switching components to drive the various loads, from these rails. There are a number of people on RMWeb who know far more than I do and who would, I'm sure help, if my thoughts don't sort it. You might try to contact Suzie and (Andrew) Crosland. The KA should not make any difference to your motor performance per se, and from what you are saying above it is absolutely doing what it's there to do, it is keeping the motor (and/or sound) running for a short period when the loco is not getting any supply. So I believe the issue is with performance - and the rattle / jerky behaviour suggests that the pwm and load compensation is not working correctly. Assuming the decoder is not broken (a safe assumption I believe) I think you should start by looking at these; I don't have a manual for the V3.5, but in the V4 manual, it starts at page 46, section 11. Just to confirm my thoughts, you might try to turn off compensation and check that the locos run at least smoothly. Read CV56, make a note of it, and then set it to zero. this will disable your load compensation and it should then not jerk or rattle at all. If so, then you can try some of the settings (or the auto setting method that I already suggested) that they give in the manual. HTH Simon Our postings passed in the ether Simon. You will see from my last posting that I have improved the performance with the auto-tune procedure BUT I am still not satisfied with the keep-alive which has a small effect on running but a much larger effect on the sound. As you say its purpose is to keep the loco moving when deprived of the supply. IMHO the effect is just too brief to be of much use and certainly doesn't justify the cost of the upgrade. My main motive in upgrading to V4 was to enable use of the ESU keep-alive which, being small, fits nicely into the frames of the Peckett. This is needed to improve the operation of an 0-4-0, albeit compensated, at shows when the dust storm provoked by visitors produces more stalls than I was used to suffering on my home layout. I am not there yet but am now left wondering what to try next. Link to comment Share on other sites More sharing options...
chaz Posted August 25, 2014 Author Share Posted August 25, 2014 (edited) Chaz check CV 113... When you connect an external capacitor or PowerPack, you are able to make the decoder switch off after a certain time. CV113 is responsible for that, since you are able to determine at what time the decoder is to switch off (as a multiple of 0.0164 seconds). You should set a time between 0.3 and 1.0 seconds to prevent your locos from driving too far during an emergency br Simon Have checked CV113 as you suggest and it is set to 255, which should give the maximum (about 4 seconds, according to the manual). At present the set-up is not achieving anything like that with the motor (see posting above). If and when I get a satisfactory result from the keep-alive I may well reduce the value to prevent the loco running off the end of a cassette - embarrassing, but not disastrous. If I may quote you... "from what you are saying above [the KA] is absolutely doing what it's there to do, it is keeping the motor (and/or sound) running for a short period when the loco is not getting any supply." Yes, but for the briefest of times. I was expecting more. It is implied in the manual that 4 seconds is possible.... Chaz Edited August 25, 2014 by chaz Link to comment Share on other sites More sharing options...
Simond Posted August 25, 2014 Share Posted August 25, 2014 (edited) Chaz I'm pleased that the auto tune seemed to help - some minor tweaking may improve matters further My results with 2200uF KA were similarly disappointing, but the Fowler is a revelation. If you have space, 5 or 6 of the 1F 2.7v supercaps in series will provide a much bigger capacitance. You need a current limiting resistor (eg 100 ohm, maybe a bit less) in parallel with a 1A diode, to ensure that they charge reasonably slowly, but can freely discharge. http://uk.rs-online.com/web/p/electric-double-layer-capacitors/7637750/ I've ordered 25 today for my experiments. Will advise progress. Good luck Simon Edited August 25, 2014 by Simond Link to comment Share on other sites More sharing options...
chaz Posted August 25, 2014 Author Share Posted August 25, 2014 (edited) Thanks Simon for your continued interest and help. There isn't enough room in my loco for any large capacitors - you can see from my photo how tightly packed the frames are, the firebox is full of motor and the saddle tank has the speaker. It's for that reason that I decided to use the ESU PowerPack having seen the size of it. I will be disappointed if it will not solve the stalling problem. What I really need is to hear from other people who have used one. it may well be that I am expecting too much - I hope not. In the ESU V4 manual that I downloaded earlier today it says... "You can solder a powerful energy buffer to all LokSound V4.0 H0 / micro decoders....... This „PowerPack“ allows your locomotive to keep running for 2 seconds without power." I would be quite happy with 2 seconds! Chaz PS I have just downloaded the manual for the PowerPack and quote the relevant passage... "The bypass time can be adjusted by means of CV 113. The standard value of 50 results in approx. 0.8 seconds. If you operate block sections, with which a signal stop is caused by switching the operating voltage off, and would like the delayed shut off time to be reduced in order to grant a point-exact deceleration. Reduce the value in CV 113.Values smaller than 10 should be avoided, otherwise the effect of the PowerPack is hardly measurable.If you stop the engine for a very long time, it can occur that the energy stored in the PowerPack is not sufficient. In this case the decoder, before reaching the adjusted time, will naturally switch off.The attainable stored energy depends strongly on the current consumption and can reach up to 3 seconds. The Maximum Value of CV113 is 255. A recommended value of 150 will give about 2.45 sec of stored energy, and will charge properly on most layouts." That does suggest that I should be getting better results than I am, does it not? Tomorrow I will try reducing the value in CV113 from the present max' of 255 and see what difference it makes - starting with the recommended 150. Edited August 25, 2014 by chaz Link to comment Share on other sites More sharing options...
Simond Posted August 26, 2014 Share Posted August 26, 2014 Chaz I agree with your interpretation of about 2.45 sec of stored energy & I should be getting better results than I am! It is probably translation, but the energy stored does not depend on the current consumption - it depends on the capacitance and voltage. The rate at which it is used, and therefore the time to deplete it, does, of course, depend on the current consumption, ie, how fast you use it. Energy stored = 1/2 C V^2 - in your case, 0.5 x 0.0033 x 15 x 15 or about 0.37J your motor is likely to take 0.5A @ 15V (I'm guessing values but the principle is good) which is 7.5W 0.37 / 7.5 is not a lot... got to dash Simon Link to comment Share on other sites More sharing options...
chaz Posted August 26, 2014 Author Share Posted August 26, 2014 (edited) Chaz I agree with your interpretation of about 2.45 sec of stored energy & I should be getting better results than I am! It is probably translation, but the energy stored does not depend on the current consumption - it depends on the capacitance and voltage. The rate at which it is used, and therefore the time to deplete it, does, of course, depend on the current consumption, ie, how fast you use it. Energy stored = 1/2 C V^2 - in your case, 0.5 x 0.0033 x 15 x 15 or about 0.37J your motor is likely to take 0.5A @ 15V (I'm guessing values but the principle is good) which is 7.5W 0.37 / 7.5 is not a lot... got to dash Simon I agree with all of that except (and I may be clutching at straws) the Mashima motor is unlikely to be drawing as much as 0.5A. If the consumption is more like 250mA then the time will be twice (yes?) - although twice not very much is still not a lot! Where to go from here? Well, I will take the loco to Telford. I will give the wheels a thorough clean - it's obvious to me that, as things stand, the keep-alive is giving some help in keeping the wheels turning, but not much. The Peckett can do a turn on Dock Green with the Ixion Hudswell Clarke sharing the duties, and ready to take over should "Susan" misbehave. If the opportunity arises I will talk to any DCC experts I can find at the show, with a view to any further tweaks I might try. Again, thanks Simon for your useful suggestions and interest shown. Much appreciated. Chaz PS Of course there are three consumers of the energy stored... the motor - by far the largest drain the sound amplifier the decoder The Peckett doesn't have working lights (thank goodness!) Would running the loco on a DC bench PSU with an ammeter give some indication of the total current consumption? Edited August 26, 2014 by chaz Link to comment Share on other sites More sharing options...
RMweb Gold Donw Posted August 26, 2014 RMweb Gold Share Posted August 26, 2014 HAve you checked the motor current draw Chaz. I remember two locos both built from the same kit where one motor was drawing 0.3A the other 0.8 the obvious assumption was binding somewhere but it wasn't that just one motor not running as well. It could only be assumed that the magnetic field on one was much weaker. Could it be your motor is a rogue one? Secondly I assumed the Keep alive pack is quoting times based on modern 00 motors rather than 0 gauge ones. BTW I am finding this a useful topic, this sort of thing where you are relating experienced problems with informed comments from others is invaluable. Don Link to comment Share on other sites More sharing options...
Simond Posted August 26, 2014 Share Posted August 26, 2014 Running the loco on a dc supply would give some indication of motor consumption, but I think Don is right to suggest that the times given are probably based on very much lower consumptions than your Mashima (and I was guessing anyway!) Yes it's pretty much linear, so 250mA draw should run twice as long (but the sound & decoder consumptions are going on too so they are a constant drain) as 500mA draw But even so it's difficult to send how they expect to get 4 seconds from 3300uF Imagine a low consumption motor 100mA @ 10V so 1W If my calculated energy of 0.37J is correct, you'll get 0.37 seconds, less whatever is needed for sound & decoder - which I think you suggested was "a few seconds", which implies they are maybe a tenth of this hypothetical motor consumption - so maybe 0.3 seconds? So I conclude that more capacity is required - there are 5 caps in the Fowler, mounted between the axles roughly where your decoder is - from Memory they are 18 long and 8 dia and you can see photos in Paul's posts on the Ixion Fowler thread - can you squeeze the decoder up into the tank along with the speaker? Good luck! Simon Link to comment Share on other sites More sharing options...
chaz Posted August 26, 2014 Author Share Posted August 26, 2014 HAve you checked the motor current draw Chaz. I remember two locos both built from the same kit where one motor was drawing 0.3A the other 0.8 the obvious assumption was binding somewhere but it wasn't that just one motor not running as well. It could only be assumed that the magnetic field on one was much weaker. Could it be your motor is a rogue one? Secondly I assumed the Keep alive pack is quoting times based on modern 00 motors rather than 0 gauge ones. BTW I am finding this a useful topic, this sort of thing where you are relating experienced problems with informed comments from others is invaluable. Don I disconnected the grey/orange leads from the motor and ran it from my DC bench PSU. This has an ammeter and voltmeter. I don't know how accurate the meters are but assuming they are reasonably accurate here are my results. Figures for current are necessarily estimates given the size of the meter scale. at 6V - 125mA at 10V - 150mA at 12V - 175mA at 15V - 200mA I put a finger on a wheel flange to represent a load - probably a much heavier load than the motor will experience on Dock Green - The current never rose above 250mA. Very satisfactory results which suggest that the motor is not a rogue and the loco is free-running. Chaz 1 Link to comment Share on other sites More sharing options...
chaz Posted August 26, 2014 Author Share Posted August 26, 2014 Running the loco on a dc supply would give some indication of motor consumption, but I think Don is right to suggest that the times given are probably based on very much lower consumptions than your Mashima (and I was guessing anyway!) Yes it's pretty much linear, so 250mA draw should run twice as long (but the sound & decoder consumptions are going on too so they are a constant drain) as 500mA draw But even so it's difficult to send how they expect to get 4 seconds from 3300uF Imagine a low consumption motor 100mA @ 10V so 1W If my calculated energy of 0.37J is correct, you'll get 0.37 seconds, less whatever is needed for sound & decoder - which I think you suggested was "a few seconds", which implies they are maybe a tenth of this hypothetical motor consumption - so maybe 0.3 seconds? So I conclude that more capacity is required - there are 5 caps in the Fowler, mounted between the axles roughly where your decoder is - from Memory they are 18 long and 8 dia and you can see photos in Paul's posts on the Ixion Fowler thread - can you squeeze the decoder up into the tank along with the speaker? Good luck! Simon Simon, where does the 3300uF figure come from? I thought there was some clever voltage conversion going on on the PowerPack that enabled a much higher value, but physically smaller as its operating voltage could be lower. Ah HA - is the 3300uf an equivalent, allowing for the voltage difference? In answer to your question - yes, I probably could push the decoder into the tank but have been reluctant to do so as it means far more plug and socketed wiring. I will check Paul's photos but don't know how many capacitors I will go into the space I would have if I moved the decoder - I will measure the space later. Chaz Link to comment Share on other sites More sharing options...
Simond Posted August 26, 2014 Share Posted August 26, 2014 Chaz I have no idea. I thought I had read it in this thread, but I can't find it. The Loksound v4 manual mentions 2200uF on page 26. I had a look at the powerpack manual too, and it doesn't mention it either. Possibly dreaming... Do take into account that typical supercaps are rated at 2.7 or 5.5 volts. The DC rails in your decoder are likely to be at about 15v so you need to put several in series, which disappointingly reduces the capacitance. The ESU manual suggests a 25V rating, but this seems a bit "generous" to me. (That said, apparently they do go off with a bang if they get annoyed...) My supercaps are due for delivery tomorrow. Simon Link to comment Share on other sites More sharing options...
Simond Posted August 26, 2014 Share Posted August 26, 2014 Some interesting bits & pieces http://www.railpage.com.au/f-p1838210.htm http://www.members.optusnet.com.au/mainnorth/alive.htm http://www.rmweb.co.uk/community/index.php?/topic/54342-making-your-own-keepstay-alive-units/ Best Simon Link to comment Share on other sites More sharing options...
chaz Posted August 26, 2014 Author Share Posted August 26, 2014 Simon, thanks for the links which I will follow up later. I have measured up the space between the frames which is currently occupied by the decoder and the PowerPack - it's 24 x 25.3 x 21mm. Not a big space for big capacitors! You could do me one more favour - once you have your supercaps could you measure up one please and post the dimensions here? My current thoughts are to take the Peckett to Telford, with its understudy, and try it out on Dock Green in a full operating session. If its performance satisfies me then I will leave well alone. Should it fall short then I will look for a different solution. Chaz Link to comment Share on other sites More sharing options...
Recommended Posts
Create an account or sign in to comment
You need to be a member in order to leave a comment
Create an account
Sign up for a new account in our community. It's easy!
Register a new accountSign in
Already have an account? Sign in here.
Sign In Now