Kiwi_Ed Posted October 16, 2014 Share Posted October 16, 2014 Hello All, After spending some time speed matching my newly acquired Graham Farish Class 108s I finally set up a successful consist last night. I am using a Roco Multimaus controller and changed CV19 for both trains (including CV22 to control the lights from the consist). I'm fairly new to DCC, so it's a lot of trial and error with getting the speed table and other CVs right.. The train's speeds match each other lovely and I'm very happy with the way the train slowly increases speed until it reaches its maximum speed. However, and herein lies the problem, when running in consist the trains only reach about half of their maximum speed.. I have no idea if I forgot to change any CV values or if this is simply a case of my controller not giving enough juice to run 2 trains at full speed.. I hope you have any idea on this. Many thanks, Edwin Link to post Share on other sites More sharing options...
Kiwi_Ed Posted October 16, 2014 Author Share Posted October 16, 2014 Hello Dutch_Master, Even though my decoder have this shunting speed option (F6), that's not causing this problem. This was one of the things I was thinking of (in case I did inadvertently activated in one of the trains before setting the consist) but this was not the case. As a matter of fact, the shunting speed still worked in the consist and cut the speed in half again.. I'm kinda guessing that the Multimaus isn't really capable of recognising these trains running in consist and therefore only giving enough power to run one train at max speed. Even though this doesn't make sense (not a electrical mega brain here..) but am I wrong to think that the motors basically run parallel on the power supply so should get the same amount of power (I hope I explain this right, but you'll probably get the gist of it). If this is caused by the Multimaus the easiest solution (besides getting a new controller when the budget allows) would be to be running two different speed tables. This means that setting up a consist takes a bit more time but it would get past the problem. Obviously this is not desirable, so if anybody knows of any other causes or cures for the problem I'd be delighted to hear! Have a nice day! Edwin Link to post Share on other sites More sharing options...
Adrian Wintle Posted October 16, 2014 Share Posted October 16, 2014 Depending on the decoder, there might be a difference in Back-EMF settings in consist (Digitrax disables Back-EMF by default in an advanced consist), so your carefully speed matched units may actually not be speed-matched in the consist. There also might be some issues with Back-EMF fighting between the two units, or the effects of marginally different starting speeds causing the Back-EMF to absorb more voltage starting the consist, leaving a smaller range to work with (and hence a lower top speed) The Multimaus will supply power as demanded (up to the maximum it can supply) - it doesn't 'allocate' power to an addressed unit. It is no different in power supply than running the two units on different addresses. Adrian Link to post Share on other sites More sharing options...
Grovenor Posted October 16, 2014 Share Posted October 16, 2014 Consisting should not affect the maximum speed which depends on the speed step sent to the decoder and the voltage on the track. Measure your track voltage with one train running and again with two trains running. Are you using the Roco transformer? Keith Link to post Share on other sites More sharing options...
Kiwi_Ed Posted October 16, 2014 Author Share Posted October 16, 2014 Adrian, Keith, thank you very much for your reply. I ran the consist 'uncoupled' with a few inches of space between the two DMUs. Just to be sure they were actually running at the same speed and reacted the same. So there shouldn't really be an issue with the Back-EMF, should there be? The decoders are Digitrax (I'll believe, should double-check..) so like you said Back-EMF should be disabled. I am using the Roco transformer with a few string of diodes between the controller and the track to lower the voltage (as described elsewhere on this forum a few times). This is to prevent frying my decoders. I'll have a look at the voltage this evening when I go home. I don't suspect there to be a difference because then I should notice a different as well when running multiple (non-consist) trains on the track. Unless I'm missing something here. Link to post Share on other sites More sharing options...
RMweb Gold Tony_S Posted October 16, 2014 RMweb Gold Share Posted October 16, 2014 Have you accidentally set the Multimaus to 14 speed steps? Try turning the light on and off (doesn't matter whether or not one is fitted!) and see if the speed alters on your consist. Link to post Share on other sites More sharing options...
Kiwi_Ed Posted October 16, 2014 Author Share Posted October 16, 2014 Hi Tony, Train runs the same with or without the lights.. They also run at normal maximum speed when not in the consist (with or without lights). Judging by the wonderful slow and gentile increase and decrease of speed I'm pretty sure I have everything set to 28/128 speed steps. As far as voltage is concerned. My (el cheapo) multimeter measures 15V without any trains on the track and 12.5V with trains. Regardless if I'm running a consist or not. I'd suspect this not to be the problem as the trains react normally when not running in a consist (even when running 3 trains, the max I can fit on my temporary test track) there doesn't seems to be a problem. I also opened up my DMUs to check what chip was inside and I'm not quite sure.. I don't see any brand or numbers or anything on the sleeve around the chip. They are just 6 pin chips with an unmarked light blue sleeve. Could it be that these are cheap decoders that don't support advanced consist? I've emailed the guy that I bought the trains from on Ebay, hoping to get an answer there. I'll let you know! Link to post Share on other sites More sharing options...
RMweb Gold RFS Posted October 16, 2014 RMweb Gold Share Posted October 16, 2014 To find the make and model of a decoder, you can put the loco on the programming track and read back CV 8 (manufacturer id) and CV 7 (model id). The NMRA has a list of manufacturer ids here - http://www.nmra.org/sites/default/files/appendix_a2c_s-9.2.2.pdf Link to post Share on other sites More sharing options...
RMweb Gold Tony_S Posted October 16, 2014 RMweb Gold Share Posted October 16, 2014 A 6pin decoder in a blue sleeve sounds like the Bachmann decoder. All the ones I have are the original one that doesn't work on DC. The new Bachmann 6 pin ones do and I think they may have a clear sleeve. I'll see if I can find the details for an original blue one. I think though the data sheet supplied wasn't correct anyway! Possibly this one? It does appear to support advanced consisting and even shows how to set a speed table with a limited top speed! Tony Link to post Share on other sites More sharing options...
Kiwi_Ed Posted October 17, 2014 Author Share Posted October 17, 2014 Just got an email back from the fellow I bought the trains from and the decoders are indeed Bachmann Branchline 36-558 decoder (see Tony's link) and should support advanced consisting. I'll probably be trying Dutch_Master's suggestion of removing the diodes and see what happens. I'm a bit weary if this, as blowing the decoders would mean setting me back about €80/£65 which would completely decimate the great deal I got on Ebay buying these two units... But I am made to believe what's causing the decoders to blow is overheating them, which shouldn't really happen when I'm running these two trains for a very short while (hopefully). I need to have a good think about this... Are there any other N gauge modellers out there that use the Multimaus without any voltage regulating measures?? Thanks everybody for trying to solve this problem with me, anyway! Link to post Share on other sites More sharing options...
Crosland Posted October 17, 2014 Share Posted October 17, 2014 I would keep the diodes. They are there to protect the motor, as much as the decoder, from the high track voltage. Abetter solution is to replace the power supply with a lower voltage one. The multiMaus is notorious for having a higher than neccessary track voltage for OO, let alone N gauge. The only thing I can think of is that the decoder is switching on (or off) the speed table when set to use a consist address but I can't think why that would be. Not being able to read back the CVs is a hindrance to debugging this. Andrew Link to post Share on other sites More sharing options...
Kiwi_Ed Posted October 17, 2014 Author Share Posted October 17, 2014 Andrew, thanks for your reaction. I have an adapter for the DC bachman controller that came with a N-gauge starter set. If I remember correct it puts out 16V. Do you think that one is suitable (if I remember correctly it even has the same type of plug)? It's a bloody nuisance not to be able to read the CVs. Especially as I paid nearly as much for my Multimaus as a Prodigy Express, which does read CVs as far as I know. At the time I thought I was convinced that the Multimaus was just what I needed (considering price, functionality and the option to add a fairly cheap second handset in the form of a Lokmaus). I must say I got a bit carried away bidding on Ebay and after loosing a few auctions at just a bit under the maximum bit, but at the highest price I was willing to pay, I decided to bit a bit more than I initially intended. This did make me win an auction but after receiving the multimaus I found out that there wasn't a connector cable between the controller and the track.. So I ended up spending another tenner or so more... Finally it all worked out only a tiny bit less than the Prodigy Express, which I would have prefered to have, was it not that I had decided it was a bit over my budget. Anyway, that's just me ranting.. It's still a great budget controller and I know a lot of people are being very happy with them. For now I'm stuck with it, but if the Christmas bonus allows I might be putting this one on Ebay and trade it in for a Prodigy Express. The other disadvantage of not being able to read CVs is that I have no idea what the values are of these two 108s. One of them drives really nice, great slow acceleration, deceleration and great speed. The second drives nicely as well but was behaving completely different than the first one. With a lot of trial and error I managed to more or less match the speed and acceleration but so far haven't been able to match the deceleration. I have CV04 set to the max (31) but the second one is still stopping/slowing faster than the first. I probably have to play with CV24 a bit to change the formula, but it would have saved me a lot of time if I just could have read the CV in the second unit, transfer the values to the second and then only fine-tune them. blablabla.. I should really do something about always typing so much. B) Link to post Share on other sites More sharing options...
Adrian Wintle Posted October 17, 2014 Share Posted October 17, 2014 I don't believe that the Prodigy Express has readback - my MRC version certainly doesn't. I think you need to go to the Advance for that. Adrian Link to post Share on other sites More sharing options...
RMweb Gold Tony_S Posted October 17, 2014 RMweb Gold Share Posted October 17, 2014 I quite like the Multimaus, though I mainly use the handsets now as extra ones on my Lenz system. Have you considered getting a Sprog and use JMRI, and then using that for your programming? If mine broke I wouldn't cope ( I probably would, I'd buy another one!) Tony Link to post Share on other sites More sharing options...
Kiwi_Ed Posted October 17, 2014 Author Share Posted October 17, 2014 Maybe I shouldn't be too fussed about the Multimaus. It's doing its job so far and beside a few things it can't do it does a fine job. I've been thinking of tying an Arduino into my control system. Haven't really looked into the DCC possibilities of the Arduino properly but I might be able to read out CVs with that and a bit of hardware/programming. The Sprog sounds interesting also. And Adrian, thank you for telling me the Express doesn't read CV. I find the manual at that particular point a bit vague. But anyway, back to the topic: I think I'm getting there! Feeling bold tonight I took my cheapest engine with the cheapest chip and removed the diodes from my setup. No major differences there in running the engine in 'normal' mode. But after changing CV19 this little engine wouldn't budge an fraction of an inch, just sat there very softly humming away at itself.. So, I figured, it wasn't the voltage causing this problem. After this I put the diodes back in and started playing around (changed CV19 back to 0). As I firstly wanted to know if this problem had anything to do with the Back EMF, I disabled it by changing CV57 to 0. What happened next was surprising: the engine wouldn't budge an fraction of an inch, just sat there very softly humming away at itself.. Sounds familiar?! Indeed, exactly what happens when I changed CV 19. Wondering what would happen if I changed CV19 this time, I did. And still the same, so I changed it back to 0. Looking at the table (see below) in the Digitrax manual (this was a digitrax decoder) I started playing with the values. Firstly I changed the value to 15 which made my engine fly off to ridiculous speeds, obviously wrong value.. Anyway to make a long story short I discovered that by finding the value that would run the engine nicely when not in a consist and then adding the value in the right column it would make the train move at about the same speed when it is programmed to run in a consist. At least that's what happened with this engine. Haven't had the time yet to check this out with my two 108s, but I will do that over the weekend and report back to you! We might have cracked the case!! Thanks a lot! Ed Link to post Share on other sites More sharing options...
Adrian Wintle Posted October 17, 2014 Share Posted October 17, 2014 I think you'll find that is a Digitrax-only feature - Back-EMF implementation is not standard between manufacturers. The whole sitting there and humming thing is probably because you don't have CV2 (start voltage) set high enough. Back-EMF can make up for a low CV2 setting so you may find that when you disable it the loco needs more throttle before it moves. Adrian Link to post Share on other sites More sharing options...
Crosland Posted October 18, 2014 Share Posted October 18, 2014 Adrian was entirely correct in "implementation is not standard between manufactures". Each manufacturer has their own secret sauce and there are no standard CVs for adjustment. To be fully effective the use of BEMF requires a closed loop control algorithm which needs to be tuned to the motor. Andrew Link to post Share on other sites More sharing options...
Adrian Wintle Posted October 18, 2014 Share Posted October 18, 2014 CV10 is not implemented by all manufacturers. It isn't a supported CV on my Lenz, Digitrax, TCS, or ESU decoders. It is supported by SoundTraxx. Adrian Link to post Share on other sites More sharing options...
Kiwi_Ed Posted October 18, 2014 Author Share Posted October 18, 2014 I'll still haven't had a chance to play with the 108s and the Bachman Decoders in there. But if I Do I'll let you know once I've tried! Does anybody know if Bachman has an extensive manual on their decoders, including recommended values for BackEMF? All I can find on their website is a 2 page very basic manual.. Link to post Share on other sites More sharing options...
Crosland Posted October 18, 2014 Share Posted October 18, 2014 CV10 simply sets the speed above which BEMF is disabled. There are no standard CVs for adjusting the feedback, such as the sampling rate or the P, I and D parameters for the control loop. If you don't understand what I am talking about read a Zimo decoder manual. Andrew Link to post Share on other sites More sharing options...
Kiwi_Ed Posted October 18, 2014 Author Share Posted October 18, 2014 Hi Andrew, I started messing before I read your post so didn't read the manual before that. Anyway. Changing the values of CV57 fixed my problem. It took a bit of tinkering and I also had to adjust my top speed a bit after it. But both trains run the same speed in the consist and this speed is somewhat close the the max speed they are running when not in a consist. I'm pretty sure that with a bit of fine tuning I can get them to run even closer to the speed they run separated. At the moment this is now about a second of a difference on an eight feet long oval, so not to bad in my opinion.. I'll go and have a read through the Zimo manual to find out if this CV10 can be of any use in this case! Have a nice day! Edwin Link to post Share on other sites More sharing options...
RMweb Gold Tony_S Posted October 18, 2014 RMweb Gold Share Posted October 18, 2014 Does anybody know if Bachman has an extensive manual on their decoders, including recommended values for BackEMF? All I can find on their website is a 2 page very basic manual.. As far as I know that is it! Link to post Share on other sites More sharing options...
Crosland Posted October 19, 2014 Share Posted October 19, 2014 I'll go and have a read through the Zimo manual to find out if this CV10 can be of any use in this case! I thought you were using Digitrax decoders? In which case the Zimo manual will not be relevant. Andrew Link to post Share on other sites More sharing options...
Kiwi_Ed Posted October 19, 2014 Author Share Posted October 19, 2014 Hello lads, I'd usually would use Digitrax decoders but these two 108s came with Bachman decoders. The Bachman manual is appallingly basic, not even worth the read if you'd ask me.. So there is no mention about CV10 in the Bachman manual and I believe also not in the Digitrax one (not sure about that). Even though the Zimo manual is not entirely relevant to my decoders it does explain all the CVs quite well including CV10. All I can say is that there is so much more to learn.. And with the right manual I would be able to do so much more fine tuning. Perhaps I should email Bachman and see if they can provide me with something a bit more elaborate. I'd love to play around with some of the things mentioned in the Zimo manual, but as I am not quite sure if these CVs are the same or indeed have any function in the Bachman decoders I'm not going to mess around with it too much. For now my 108s run brilliantly smooth and I'm very happy with that! Thanks again for all your help. Link to post Share on other sites More sharing options...
Phil S Posted October 19, 2014 Share Posted October 19, 2014 Phil: Point number 1: I would recommend replacing the Roco Transformer with a Switched Mode Power Supply - such as supplied for 'laptops': 18V 3.5A or 4A is what you require. This will then give a stabelised poutput voltage to the track - regrardless of hwether any trains are running ! Measurements with the Standard Transformer (which is no longer supplied onece stocks are exhausted, due to EC regulations on energy efficiency) show, in the UK a poosble off-load track voltage (accurately measured with an oscilloscope or RRampmeter) reaching 22V .... and dropping as more current is taken: This is correct and normal for a transformer, as it is unregulated! It will, witj 230Vac input, give the quoted 16V ac rms outpt when the full current is taken [ or higher in the UK!]..... Coach lighting when no train is moving gets bright! On the other hand, with the SMPS the `18V DC is maintained, and becomes a stable 16Vdcc on track (which is in spec .... and will give about 14V dc for controlled lighting, and 12V max for the controlled motor, via a decoder, [assuming 2V drops on the diode bridges in the decoders, and not allowing for any track impedance losses] It is highly likely that the extra current consumed by multiple motors, with your additional circuitry, is causing your voltage drop problem - have you compared against simple runnign both motors around the track at the same time - even using one handset - simply by programming them as different numbers, and starting on, then the other ??? .... it will clearly demonstrate that it is nothing to do with it being a consist !! Link to post Share on other sites More sharing options...
Recommended Posts
Archived
This topic is now archived and is closed to further replies.