Simond Posted August 12, 2018 Author Share Posted August 12, 2018 (edited) Addendum; there’s an easy fix. Put two cells in series. This will feed the step-up with around 7.4V (about 8.4 fully charged, down to about 5.6 at end of discharge), which will mean the step-up works much less hard, but... there’s always a but... the step-up shuts down if the input voltage gets too low, which protects the cell. If I double up, there’s no protection against over discharging the cells, and that wrecks them, and makes them dangerous to charge, too. I can be careful, whilst experimenting, but it’s not a real solution. But it is a decent way of checking my hypothesis, and, perhaps more importantly, eliminating most of the other possibilities. Best Simon Edited August 12, 2018 by Simond 4 Link to post Share on other sites More sharing options...
Giles Posted August 13, 2018 Share Posted August 13, 2018 Simon, I know little about electronics, but might there be another device/component that you could fit and act as a 'canary' that would activate or de-activate on a voltage drop? 1 Link to post Share on other sites More sharing options...
Simond Posted August 13, 2018 Author Share Posted August 13, 2018 Giles Indeed, yes, canaries of various temperaments are available... some can even talk! best Simon Link to post Share on other sites More sharing options...
CME and Bottlewasher Posted August 13, 2018 Share Posted August 13, 2018 Hi Simon, I'm liking your sense of adventure with is one! Kindest, CME Link to post Share on other sites More sharing options...
Simond Posted August 13, 2018 Author Share Posted August 13, 2018 (edited) I do love a good experiment... I’m sorry, these videos are really uninteresting, but the data they contain... Video 1, running forwards. Voltage very consistent, sound cuts out after a few moments, and can be restarted, but cuts out again. Video 2, running backwards. Voltage constant, sound doesn’t cut out at all. Then running forwards, sound cuts out. Voltage still 12.8 Video 3, running forwards, using my fingers as a brake And Video 4, running forwards, fingers as brake, but measuring current OK... So as noted below, the videos 1 - 3 are showing the voltage output of the step-up regulator, and it’s pretty much consistent at 12.8 +/- 0.02V unless I load up the loco, using my fingers as a brake. This was more or less exactly what I expected, with the exception of the decoder cutting out the sound when going forwards. That bears further examination, and I’ll do that perhaps tomorrow. I can easily connect the decoder to the pickups, and run it on the layout, or rolling road using my Lenz 5A booster, which won’t droop when loaded! Video 4 is really surprising. The scale of the multimeter is 0-2A and it’s showing the loco running at low speed, and with negligible load, at only 20mA which is very much less than I expected. The motor is a Canon, on a Premier Components helical gearbox. It’s obviously pretty efficient. As I wind the speed up, the current increases to around 60-70 mA, and then drops by around 10mA when the sound cuts out. You can see that the loco is still running at constant speed (it seems to have an odd speed curve, it’s very flat from about speed step 19 right up to 28, which is unexpected). As I then use my fingers to brake the loco, the current rises as expected, and cuts off very sharply at about 100mA suggesting the step-up is struggling to supply anything like the necessary current, and is simply cutting out. I think I need to look at my numbers again, as this doesn’t tally at all with what I posted a couple of days ago. Really very interesting. I have located a commercial BMS from Conrad, this will certainly manage the cells, and only £3 or so each. Using one per cell would allow me to use two or more cells in series, and thus reduce the demand on the step-up regulated supply. In turn, I’d expect that to provide considerably more current to the receiver, decoder and motor, and thus sort out the poor performance. The ultimate would be to use 4 in series, and do away with the step up entirely, but they would take up a lot of space, and need removal for charging, or a 5-pin charging lead. Not sure that’s very practical. Once I’ve understood the decoder issue a bit better, we’ll see how to sort that too. So some progress. More soon. Best Simon Edited August 13, 2018 by Simond 3 Link to post Share on other sites More sharing options...
Fastdax Posted August 13, 2018 Share Posted August 13, 2018 Video 2, running backwards. Voltage constant, sound doesn’t cut out at all. What's the DVM actually measuring? Link to post Share on other sites More sharing options...
Simond Posted August 13, 2018 Author Share Posted August 13, 2018 Hi Duncan, the DVM in videos 1, 2 & 3 is measuring the voltage output of the step-up converter, which is the input to the Tam Valley receiver. It’s regulated and adjusted to just under max output. It’s running on one 18650 Lithium Ion cell. Best Simon Link to post Share on other sites More sharing options...
Fastdax Posted August 13, 2018 Share Posted August 13, 2018 the DVM in videos 1, 2 & 3 is measuring the voltage output of the step-up converter, which is the input to the Tam Valley receiver. It’s regulated and adjusted to just under max output. It’s running on one 18650 Lithium Ion cell. I guess the next logical step is to put the DVM in series with the converter output and measure its current. Link to post Share on other sites More sharing options...
Simond Posted August 13, 2018 Author Share Posted August 13, 2018 (edited) Yup... As you see, I had, it just took two hours to load the videos on YouTube. I think our broadband needs more weetabix... Best Simon Edited August 13, 2018 by Simond 1 Link to post Share on other sites More sharing options...
Simond Posted August 14, 2018 Author Share Posted August 14, 2018 Next step, repeated the previous tests, but with two multimeters. The yellow one is reading Amps, the grey is reading Volts. It seems that at about 450mA the voltage is dropping to around 10.3V, and the system falls over. It won’t self reset, the battery must be turned off and on again, and the transmitter has to be reset to speed step zero. Presumably to prevent scary moments... Now I need to do some sums... But first, I think I’ll try two cells in parallel. Well, I’m bored of loading YouTube vids, so you’ll have to take my word for it, the voltage drop is almost negligible with two cells, and the trip still occurs at around 430mA. This isn’t proof positive, but is ver6 strong evidence that the step-up is the weakest link. Next step, connect the decoder up to the pick-ups, and try the sound out, and ensure it doesn’t trip out at all. Watch this space! Best Simon 1 Link to post Share on other sites More sharing options...
Simond Posted August 14, 2018 Author Share Posted August 14, 2018 (edited) And now, I can see the issue. With the loco running from the track DCC supply, the voltage was 13.67V on an Avo 410 true RMS meter(DCC is AC, remember), and the current was around 160mA, though I wouldn’t like to bet the DM25 is true RMS - so treat this with some scepticism. Once I started to apply the finger brake, the voltage did not drop much, as expected - it’s 12.5V in the photo. The current rose through. It didn’t stall even at 900mA, which is, of course more or less what I need to achieve from the battery supply. It’s difficult to photograph, this is as good as it gets! So, it appears that the current is hitting a limit, either in the step-up, or in the receiver. Next step, power the receiver from an external power supply.. EDIT... Curiously, seem to have an issue with this. It wouldn’t work off my normal Gaugemaster DC supply, either with the controller or just a bridge rectifier. Very odd. I therefore had to make a new pair of leads to use the main layout power supply, Which I couldn’t be bothered to do yesterday evening, but did before heading to the office this morning. On a DC supply to the receiver, the loco was drawing 130mA or so at 13V. Applying the finger brakes, the voltage drooped a little, but as the current rose past 350mA, the receiver dropped out. This seems to be the limiting factor, which is hugely disappointing. I know Tam Valley do a higher current receiver, but this is a significant blow, I’m really surprised at how low this current is. Most “HO sized” DCC decoders will handle at least one amp for short periods. At least I know that the step-up isn’t the main problem. Still need to sort out the decoder sound issue. Best Simon Edited August 15, 2018 by Simond 2 Link to post Share on other sites More sharing options...
davetheroad Posted August 24, 2018 Share Posted August 24, 2018 Just found the link to this topic on the Radio control page. I see you are having problems with brownouts, probably due to voltage sag. One way to reduce voltage sag is to use the biggest batteries you can fit as larger capacity cells sag less under high loads. All the components need to handle the highest loads and be rated for at least the highest voltage but I guess you know this I tested for voltage sag with my 00 scale locos by fitting a low voltage alarm. The results were interesting and alarming (sorry) There were a stream of alarms through out the run, they were random. my conclusion was that there are transitory spikes in the locos current draw demanding sufficient amps to make the voltage sag a long way!. The solution was 2 fold, first fit bigger batteries and second remove the alarm!. It did not brown out my Deltang receivers which would only shut down if the voltage dropped below the safety cut off levels for an appreciable length of time. My experience is with Deltang equipment in 00 scale so I am guessing for 0. If I was converting a large 0 scale Pacific I would buy a deltang Rx 65 with the piggy back board to handle 6 amps or the new Rx 66 which handles 4amps or 6 with a heat sink. they are both bulit with 20v components and are rated for 18V A large amp rated voltage regulator, for example the Pololu U3V70x has a version that delivers 15V at up to 5A at 95% efficiency from 3 lipo cells 2S or 3S lipo cells connected in series. Link to post Share on other sites More sharing options...
Simond Posted August 24, 2018 Author Share Posted August 24, 2018 (edited) Thanks Dave, My tests to date confirm that it is the receiver that cannot provide enough current for the motor & decoder. At stall the motor will pull upwards of 900mA and the receiver falls over at around 400. The cells I’m using can supply 10’s of Amps, and the step-up (which I initially suspected) can manage an amp and a bit, say 1250mA. The receiver is the weakest link. Tam Valley do offer a 3A version, but it’s £75 which seems a bit steep. I’ve had the current receiver on the ‘scope at work, and think I can add an H bridge to amplify its output, just want to check with a pal who knows more about this than I do. More news soon Best Simon Edited August 24, 2018 by Simond Link to post Share on other sites More sharing options...
Giles Posted August 25, 2018 Share Posted August 25, 2018 It's good to get to the bottom of these things at any rate.... good luck with it! I've only used Deltang without any sound of course, so it's a much simpler proposition! These things are always a learning curve.... We're at Telford on Saturday - are you around then? Link to post Share on other sites More sharing options...
RMweb Premium jamie92208 Posted August 25, 2018 RMweb Premium Share Posted August 25, 2018 It's good to get to the bottom of these things at any rate.... good luck with it! I've only used Deltang without any sound of course, so it's a much simpler proposition! These things are always a learning curve.... We're at Telford on Saturday - are you around then? If you do get to Telford there will almost certainly be people there who can help. Either on the Tech Committee stand or if not it would be worth talking to the guys running the test track are worth talking to. Jamie Link to post Share on other sites More sharing options...
Simond Posted August 25, 2018 Author Share Posted August 25, 2018 Giles, yes, will look out for you. I’ll probably wear the pink polo shirt! Jamie, thanks, it’s either the willingness to spend more money, or a bit of clever electronics, or an acceptance that it’s either-or that I need! I have no doubt the Deltang kit would do everything needed, EXCEPT the sound. I have a fleet of some 18 DCC fitted locos, most with sound, and I hoped that the Tam Valley solution would offer the opportunity to add batteries & R.C. to them, as that would then utilise the investment I’ve already made. Adding a few receivers, and batteries on a “plug it in when you want R/C and revert to DCC when you don't” seemed like a reasonable approach, and it’s certainly possible using the Tam Valley transmitter & receiver, but won’t give the performance I want, unless I spend £75, plus batteries on each loco, and the extra thirty quid per conversion does not appeal. I might well sell the Tam Valley kit and try the Duchess with Deltang - I’d lose the sound, but the r/c would work. The GW fleet can remain firmly in the DCC camp until a better option comes along. No decision made, but if someone makes a good offer for the Tam Valley Tx & Rx it might make up my mind! Best Simon Link to post Share on other sites More sharing options...
davetheroad Posted August 25, 2018 Share Posted August 25, 2018 Deltang Rx65 and Rx66 receivers can control a sound card. I have not tried it though Link to post Share on other sites More sharing options...
Simond Posted August 25, 2018 Author Share Posted August 25, 2018 (edited) Thanks Dave, My understanding is that a sound card is not the same beast as a sound-equipped decoder, if nothing else because it isn’t synchronised with the wheels. And, I’ve already got the decoders! Best Simon Edited August 25, 2018 by Simond 1 Link to post Share on other sites More sharing options...
Simond Posted March 3, 2019 Author Share Posted March 3, 2019 Latest news: I’ve been entertaining myself with Arduinos of late, but did solve the problem of the pultruded glass fibre rod. It’s available in black or white, and in 1, 1.5 & 2 mm sizes, in packs of 20 or so handy six-inch lengths, from fishing gear suppliers on eBay. It is electrically insulating. Prices were reasonable, and it surely has other uses in modelling, but I aim to fit the brakes in the near future. best Simon 1 1 1 Link to post Share on other sites More sharing options...
robertc Posted March 4, 2019 Share Posted March 4, 2019 Hi Simon , just had a casual glimpse over your recent posts. Not sure what types of lithium cells you are using but let the buyer beware when it comes to buying unknown brands online. I purchased several supposed A123 cells from a couple of different suppliers when Hobbyking ceased supplying them. None of the cells which looked genuine could deliver anywhere near the current of the genuine article and capacity was a manufacturers fantasy. I note you traced your issue down to the Tam Valley unit not capable of supplying the current but just be aware of this issue with batteries. My couple of BPRC models are just for use on non-DCC layouts and I live without the sound for the moment. The smooth and ever reliable power delivery of BPRC does make up somewhat for that loss. cheers Bob Link to post Share on other sites More sharing options...
Simond Posted March 5, 2019 Author Share Posted March 5, 2019 Thanks Bob, it’s certainly possible that they’re “nowhere near”, but I’d doubt they couldn’t make 5A, they’re 18650’s. I suppose a quick test might be in order. Now, where did I leave that old car headlight...? hoping to make some progress with the old girl over the next few weeks, th3 garden season is beckoning! cheers Simon Link to post Share on other sites More sharing options...
Phill Dyson (onslaught832) Posted March 10, 2019 Share Posted March 10, 2019 Hi Simon, I have found your experiment with Tam Valley & the Duchess most interesting, I do hope you resolve your issues with the Tam Valley system as at present it seems tge only option for synchronised true loco sound for us dead rail types. Cheers Phill 1 Link to post Share on other sites More sharing options...
Simond Posted March 10, 2019 Author Share Posted March 10, 2019 Thanks Phill, glad you found it, and that you found it interesting. I’m a bit discouraged tbh, as I wanted DCC sound and r/c, and it seems, as you say, the only way to have both. It does look like I could flash the cash & my dreams would come true (and I guess the smaller receiver might work fine in a smaller loco), but it’s far too damn expensive to think about conversion of the fleet. I might buy the higher power receiver just to prove the system, and to have the old girl available for track days and shows. Alternatively, I might sell off the Tam Valley stuff, go Deltang and see what sound cards are available. Might be interesting to see what alternatives could be made to work. For me, the chuff needs to be load related, so it gets stronger the harder the loco is working, and softer when idling or freewheeling, (nice to have snifters that sound right too!) and the whistle needs to be typical of the loco, so maybe there’s more than one way of skinning the cat. Currently very absorbed in designing layout control systems using Arduinos. Looking forward to some proper modelling sometime soon! best Simon 1 2 Link to post Share on other sites More sharing options...
Simond Posted March 30, 2019 Author Share Posted March 30, 2019 (edited) Sometimes you can’t win... Here we have an upside down Duchess chassis, with a pair of brake shoes. Now I know that Premier Components stuff has a well deserved reputation for dimensional accuracy, so I’ll assume that when I fitted the horns, I got them just too far forward, but the wheel flanges are horribly close to the pivots for the brake shoes. Indeed, having obtained some 1mm diameter pultruded glass fibre rod (eBay, fishing supplies) to use in plaice (sorry) of the normal brass wire to fit the brakes, I discovered that the brake shoes will not sit close to the wheels, as they cause the gf rod to flex. This means I need to redrill the holes in the frames. Happy days, not. The wheels and motion have to come off again to do it. otoh, The ash pan-mounted speaker is ideal. Pleased with that. I was able to solder the “triangular horseshoe” into the chassis using the RSU, which was helpful. The speaker is retained by two 10 BA screws. The decoder is taped to a bit of foam rubber, that’s taped to the top of the motor, thus keeping all the wiring within the chassis. Needs a bit of tidying. almost forgot; stripped the rear pony truck, and turned down the bosses of the wheels, deepened the countersink and shortened the axle to suit. This increases the maximum cross-throw of the axle, and hopefully prevents any derailments on 1750mm radius curves. I wonder if some kind of spring is needed too, but this is likely to limit haulage power, which isn’t ideal. It’s pretty much perfect going forwards, but the pony comes off occasionally going tender first. Maybe I’ll try some lead weight as an easy next step. still in two minds about the R/C. a bientot Simon Edited March 30, 2019 by Simond Forgot a bit! 4 1 Link to post Share on other sites More sharing options...
Simond Posted March 31, 2019 Author Share Posted March 31, 2019 (edited) Modifications made, touch of paint here & there... looking rather better. I don’t think I can get the wheels off with the brakes fitted, so hopefully won’t need to! This shot also shows my method of doing coupling rod bushes. They are standard Slaters top-hats, tapped 10BA and with two flats milled on - the brass thing to the lower right is a twiddling stick for tightening them, it’s a bit of 4.7mm (3/16”) rod with a hole up the middle and a flat bottomed slot milled across one end. A bit of k&s brass tube fits snugly over the end to stop it sliding off the nuts. I suspect I soldered the tumble on. It could be knurled, fitted with a screwdriver handle, or otherwise embellished, but it seems fine with just a plain spindle. Sandpipes would be nice, and they’re next, but first I’m going to ensure that it runs fine as it is! best Simon Edited March 31, 2019 by Simond 6 3 Link to post 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