Jump to content
 

Is my Zimo decoder faulty?


Torper
 Share

Recommended Posts

  • RMweb Premium

I've just built a kit of a CR 439 class 0-4-4T loco, powered by a Mashima motor.  Unfortunately, I cannot get it to run well using one of these Zimo Budget decoders with my NCE Powercab.  Basically, on applying power the loco builds up speed to about speed step 60, but then it remains at that speed (which is maybe about half speed, certainly nowhere near full speed) regardless of how much further I increase the speed on the controller (up to 128).  On the way down, the loco only begins to react and reduce speed at about speed step 60 again.  The same thing happens using 0-28 speed steps (gets to 14, then no more). and it also runs pretty roughly anyway.  So as I had another budget Zimo lying around (used, not new), I thought I'd try that.  Same thing happened.  So off comes the decoder, and revert to DC analogue control.  And it works perfectly, accelerating up through the entire range to full speed, and decelerating again, and running much better generally, eg starts at once whileas on DCC it needed a prod.  I should add that in an attempt to resolve the problem I reset both Zimos to factory default, but that didn't help.

 

Any ideas, anyone?  As things stand, i don't think I'm going to be buying any more of these budget Zimos, even if they were readily available.

 

DT

Edited by Torper
Link to post
Share on other sites

Surprised by that,  but without inspecting the loco for possible issues, its difficult to tell what's going on.   

 

I'd start by checking for stray current paths between motor and pickups.   That could be "no issue" on DC, but really mess up DCC (though usually doesn't work at all, so would be looking for a path with low resistance). 

 

Prodding to start indicates something is different when the decoder is installed compared to DC situation.  Could be balance of loco (0-4-4's being prone to un-balanced).  

 

I've fitted quite a few Zimo's to kit and scratchbuilt locos for many people, and not had problems with them.  

 

- Nigel

Link to post
Share on other sites

Zimo don’t make ‘budget’ decoders, they all have the full feature set available and the cheapest at around £20 provides absolutely superb control of locos.

 

there are a host of CVs that you should be looking at to maximise the throttle to 128 steps for the loco CV2 for the lowest speed, CV5 for max speed - leave CB6 at 1. You can also play with CV57 to reduce max voltage.

 

if you download the Zimo manual you will find all the CVs detailed and how to use them.

Link to post
Share on other sites

  • RMweb Gold

Have you got any kind of interference suppression components fitted in the motor circuit? They can affect a decoder's back-EMF measurement.

 

P.S. You need to test decoders from other manufacturers before singling out Zimos in particular. My guess is that they would all show the same symptoms.

 

Edited by Harlequin
  • Agree 1
Link to post
Share on other sites

  • RMweb Premium

Thanks everyone.

 

This si a kit built loco and as a result there are various bits where I've soldered wires to small lengths of copperclad circuit board within the chassis.  I have found that flux can seep into the areas where I have insulated these, causing not so much a short circuit but definitely a connection that makes the multimeter needles twitch.  So bearing in mind NigeI's comment about "stray paths" I unsoldered all these, cleaned them up, and resoldered them.  This time, not even a twitch from the multimeter.  And, gratifyingly, performance of the loco was very much improved - smooth running from the start, steady acceleration, just what you hope for from a Zimo.

 

That is until you get to speed step 60 or thereabouts when, as previously, there is no further acceleration, even when the controller is pushed up to 128.  So, using a rolling road on my Programming Track, I followed Iain's advice and had a look at the CVs - or at least, I looked at those that I was allowed to.  CV 29 was set by default to 14, but I reset it to 2 without any apparent effect, adverse or otherwise, on the running.  CV 2 it couldn't read.  CV 5 was 01, CV 6 it couldn't read.  Looking to set a long address I tried to access STD 1, but was rewarded with the message "Main Off", then "Can Not Read CV".  However, it was clear that the loco was receiving some message as it twitched when I tried to enter STD 1.

 

Phil, no interference suppression or indeed any other such components fitted.  As to other manufacturers' products, all my locos have Zimos but I have ordered one of the Lenz budget decoders to see if it performs any differently.

 

34"C", I tested them on the layout and they seemed to run normally.

 

Oh, and Iain, you'd best tell Coastal DCC about the "Budget" description as that's how they describe the MX600.  As "budget" can merely mean inexpensive, I personally don't see what's wrong with it.

 

DT

Link to post
Share on other sites

2 hours ago, Torper said:

Thanks everyone.

 

...

 

Oh, and Iain, you'd best tell Coastal DCC about the "Budget" description as that's how they describe the MX600.  As "budget" can merely mean inexpensive, I personally don't see what's wrong with it.

 

DT

 

As I have only ever paid £20-£25 for all of my Zimo decoders - all of which are fully featured I must be buying 'budget' decoders and as I am able to make every loco run beautifully at accurate scale speeds then I won't be changing from my 'budget' decoders.

 

Personally I don't consider them budget nor do I consider them cheap - I simply consider them to be reasonable value, especially when compared to the overpriced offerings from other manufacturers which don't perform as well as my 'budget' decoders from Zimo.

Link to post
Share on other sites

  • RMweb Premium

Zimo introduced the cheaper ‘budget’ range a while back. Certain decoders sold at the price of £20. Standard types to suit various scales, the ones sold in greater numbers I would guess. One each in 6-pin, 8-pin, 21-pin, next18 etc. This cost is much less than most of the others, which are mostly in the £32 upwards range and the normal price I’ve paid for them and the likes of CT & Lenz in the past. A marketing decision similar to the £99 across the board sound decoders, and which I welcome.

 

It’s true they are not budget in comparison with other makes for they use the same basic firmware across the decoder range as do other makes I.e. motor control, but they do have less features than the higher priced ones, less functions etc. The MX600 for example is a specifically made single sided design with ‘just’ four functions and no specific ground solder pad for stay-alive, so you have to bodge your own. But apart from CT they have by far the best motor control across the board.
 

Like many others I wouldn’t dream of using anything else.

 

Izzy

Link to post
Share on other sites

  • RMweb Premium

Latest progress report.  After a few more efforts I was eventually permitted to look at CVs 2, 5 and 6.  All appeared to be set at 0.  I changed them to 1, 160 and 250.  I also reduced the back EMF from 250 to 100.  All this increased CV29 from 2 to 34.  And to some extent it worked.  I did now have full speed control throughout all 128 steps.  The trouble is that it is very slow speed control and top speed at 128 is now the same as it previously was at 60, and about half what it is using analogue DC.  And on the track the loco was still a bit of a pig - it laboured its way along, stopping every now and again, at an even lower top speed.  This contrasts with analogue where the loco runs much faster and much more freely with seemingly little effort.  I doubt if its the motor as if that was faulty it surely wouldn't run so well on DC.

 

So I'm left a bit despondent.  I don't really want to take a decoder out of one of my other locos (all are Zimo, all are wired in).  I don't want to buy another Zimo MX600 in case the same thing happens (and please don't even ask me about the experiment with the Lenz decoder).  I can't think that both my MX600s can be faulty in exactly the same way.  Somewhere or other I've got an old Lais decoder that never worked well at the best of times and I might try that, just to see if the results are similar to those I'm getting with the Zimos.  After that  I'll probably partly dismantle the chassis, check all electrical connections very carefully, clean it up, and then try again.

 

DT

Link to post
Share on other sites

  • RMweb Premium

Thank you.  Changing CV6 to 1 did indeed improve matters a bit.  I haven't time just now to give the loco a run on the track (it's currently on a rolling road on my separate Programming Track) but will give it a go later in the evening to see how it works.

 

CV57 is set at 0, which the manual tells me is the default.  What the manual says about CV57 is that it concerns "Absolute motor drive voltage in tenths of volts at full speed (speed knob on top) should be present on the motor. #57 = 0: in this case is automatically adapts to the current rail voltage (Relative reference)".   I'm afraid that even after reading that several times I am none the wiser.

 

DT

Link to post
Share on other sites

I have used Zimo sound decoders for years with the two speed step CVs, 5 and 6 set to off (1 or 0), I can't remember. I then used CV 57 to increase or decrease the speed of my locos. The higher the number the faster and vice versa. 1 being really dead slow snail pace and 250 being warp factor 5. Not sure if non sound decoders behave differently. That may help or it may be total tosh but it has worked for me for around 15 years. Good luck!

Link to post
Share on other sites

  • RMweb Gold

CV57 is basically to manually tell the decoder your track voltage. I.E. if you know your DCC system puts 14V to the track then set CV57 to 140 instead of letting it try and guess. Not sure if it will make any real difference but can't do any harm.

Link to post
Share on other sites

58 minutes ago, Kaput said:

CV57 is basically to manually tell the decoder your track voltage. I.E. if you know your DCC system puts 14V to the track then set CV57 to 140 instead of letting it try and guess. Not sure if it will make any real difference but can't do any harm.

 

I read the manual a little differently. I think CV 57 sets the maximum voltage to be applied to the motor. This is a separate adjustment from CV 5.

 

From the Zimo manual:

 

CV 57 specifies the base voltage used for motor regulation. For example: if 14V is selected (CV value: 140), the decoder tries to send the exact fraction of this voltage, determined by the speed regulator position, to the motor, regardless of the voltage level at the track. As a result the speed remains constant even if the track voltage fluctuates, provided the track voltage (more precisely, the rectified and processed voltage inside the decoder, which is about 2V lower) doesn't fall below the absolute reference voltage.

 

So the maximum speed is limited in 3 ways, all simultaneously operating:

1. The voltage supplied by the DCC Control Unit governs the absolute maximum voltage available to the motor.

 

2. If CV 57 is set to a non-zero value, then the maximum APPLIED voltage across the motor is not allowed to exceed the value specified in CV 57 (in tenths of a volt).
If CV 57 is zero, then the full DCC Control Unit voltage is available.

 

3. If CV 5 has value 0 or 1, then top speed is determined by maximum APPLIED voltage.
If CV 5 has a value greater than 1, then this reduces the maximum speed to a fraction. 

If n is the value in CV 5, then the maximum speed is ( n / 255 ) times the speed available from the maximum APPLIED voltage.

 

Link to post
Share on other sites

  • RMweb Premium

As I see it, therefore, CV57 defaults to 0, and CV5 to 1.  If you want your loco to run to its maximum speed without being held back by the chip, you therefore use these values.  If, on the other hand, the prototype of your loco had, say, a maximum speed of 40mph whileas the model is capable of a scale 100mph, you'd use CV5 (and/or CV57?) to ensure that your model didn't exceed the prototype's maximum - is that right?

 

DT

Link to post
Share on other sites

15 hours ago, Torper said:

...please don't even ask me about the experiment with the Lenz decoder...

But we should, it is experimental data. In a good quality mechanism I should expect it to perform pretty much the same as the Zimo as far as maximum speed is concerned. (I use both decoder types regularly.)

 

If that is so, then I would suggest it is inadequate track voltage from the DCC system that is the problem, resulting in much less than 12V available at the motor terminals, whichever decoder is used.

 

It's tangential to this problem, but because there is no standardised relationship between voltage at the motor terminals and scale speed in RTR OO, I have set my DCC system for higher than standard track voltage, in order to make the more sluggish mechanisms achieve a scale maximum speed. This had the side effect of making my kit built mechanisms capable of galloping along at the rate they had achieved on 12V DC in the 'days of analogue past'.  (All are slow locos so this capability isn't required, and the CV's were set accordingly to produce a plod along rate of progress.)

Link to post
Share on other sites

8 hours ago, Mike Buckner said:

 

I read the manual a little differently. I think CV 57 sets the maximum voltage to be applied to the motor. This is a separate adjustment from CV 5.

 

From the Zimo manual:

 

CV 57 specifies the base voltage used for motor regulation. For example: if 14V is selected (CV value: 140), the decoder tries to send the exact fraction of this voltage, determined by the speed regulator position, to the motor, regardless of the voltage level at the track. As a result the speed remains constant even if the track voltage fluctuates, provided the track voltage (more precisely, the rectified and processed voltage inside the decoder, which is about 2V lower) doesn't fall below the absolute reference voltage.

 

 

 

The way I read that is I could fit a 3v motor and set CV to 30 - not that I would want to, though I suppose I might use such a motor to drive a scenic accessory like a windmill or maybe an exhaust fan on a diesel loco. Or have I missed the point ?

 

As our motors are all nominally 12v, surely that means we should set CV57 to 120?  Or are we accepting it as OK to exceed 12v even when flat out because of the PWM effect of DCC?

Link to post
Share on other sites

  • RMweb Premium
54 minutes ago, 34theletterbetweenB&D said:

But we should, it is experimental data. In a good quality mechanism I should expect it to perform pretty much the same as the Zimo as far as maximum speed is concerned. (I use both decoder types regularly.)

 

OK, but it's somewhat embarrassing.  I've only ever used Zimo decoders, all hard wired.  So I wired up the Lenz the same way I do with Zimos, and applied power.  The driving wheels did half a turn and then stopped.  Very shortly afterwards I noticed a smell of burning, and the Lenz decoder was too hot to touch (I was testing ths loco chassis without the body).  I think it's now dead.  Why this happened I still don't know - red and black to the pickups, orange and grey to the motor.  Without altering the set-up in any way I then refitted the Zimo and all was well, within the limitations I've already mentioned.  I'm having a week where everything I touch in the modelling world isn't turning out as hoped!  Although my multimeter doesn't discern any shorts, I think I'll maybe still go ahead and strip the loco of the elctric connections I've fitted, clean everything up, and replace them with new,

 

DT

Link to post
Share on other sites

2 hours ago, Torper said:

...The driving wheels did half a turn and then stopped.  Very shortly afterwards I noticed a smell of burning, and the Lenz decoder was too hot to touch (I was testing ths loco chassis without the body).  I think it's now dead.  Why this happened I still don't know - red and black to the pickups, orange and grey to the motor...

That's good information. Combined with all the other information: slow top speed, less than free running,  and some reluctance to start, this takes us back to the @Nigelcliffepost: there must be an intermittent and somewhat resistive connection path between rail and that rail's motor supply  somewhere on the chassis. No effect whatsoever on DC, but will constantly degrade DCC operation.

 

Take the worm off the motor shaft so that model rolls, and with the decoder still installed but no power on the track, test  for conductivity between rails and motor terminals as the model is rolled about.

 

(And at least you can get your money back on the Lenz warranty.)

Edited by 34theletterbetweenB&D
Link to post
Share on other sites

  • RMweb Premium
1 hour ago, 34theletterbetweenB&D said:

(And at least you can get your money back on the Lenz warranty.)

 

I'm not entirely sure about that.  According to the warranty terms, I have to contact Lenz in Germany and then return the decoder to them, presumably there.  They will then, at their discretion, either repair it or exchange it.  They won't do so if the problem resulted from "wear, normal use or improper use" (which seems to suggest any use at all), and the guarantee only applies if the product is "handled properly".  In this case, I bet they'll refuse liability on the basis that womething was wrong with the loco, and they may well be right.  Oh, and as I cut off the 8-pin plug so that I could hard wire it they'll no doubt say that I had altered it and voided any  warranty.   It only cast £19 and I'm not at all sure its worth the hassle, but I probably, and quite unreasonably, won't buy Lenz again.

 

I'll try the test you suggest this afternoon.

 

DT

Edited by Torper
Link to post
Share on other sites

  • RMweb Premium

Well here's a thing.  A long time ago I bought a Lais DCC decoder as I wanted to find out if a £7 decoder was any good.  Such tests as I did showed that it didn't really compare to Zimo, so I put it away.  Anyway, this afternoon I found it and thought I'd give it a go on my problem 439 - and it works!  Acceleration through all 128 steps up to a genuine full speed!  Now I'd be the first to concede that its controllability, especially at low speeds, won't compare to a Zimo, but why is it now working so much better than my two Zimo MMX600s?

 

DT

Link to post
Share on other sites

One key piece of information is not reading all CVs on programming track reliably.  That's indicative of a problem which needs fixing.  Testing by "does it smoke on the mainline" isn't a good approach; if something won't program reliably, it needs investigation.   As the pickups seem OK (DC running), it points to a problem inside the loco. 

 

Partial or intermittant shorts on metal kit-built locos are common.  Little things touch and make partial paths; rims onto brake gear, rims onto rods, motors moving inside the body and just catching, stray fibres, partial shorts through flux over PCB "gaps", are all places I've seen issues.   Different pickup methods lead to different types of issue - split frame has its own set, different to wipers/plungers.    There is also a possibility of a motor internal fault.   

 

A sprung/compensated loco has more scope for issues; the fault occurs, and the loco drops power momentarily, so the chassis relaxes slightly and the wheels (and motor + gearbox) move in their springs/compensation, the fault clears. 

 

Each time the loco is opened up, stuff moves, and can clear faults, at least temporarily. 

 

Some of these can be no-issue on DC, but disaster for DCC.  

 

 

Back on DC, what is the current consumption ?  My experience of Mashima motors suggests it should be around 100-200mA, much higher would indicate an issue on DC.   If testing with a cheap multi-meter start on the highest Amp scale (say 10A) and work down, otherwise risk blowing the fuse inside the multi-meter. 

 

 

- Nigel

  • Like 1
Link to post
Share on other sites

  • RMweb Premium
56 minutes ago, Nigelcliffe said:

Back on DC, what is the current consumption ?  My experience of Mashima motors suggests it should be around 100-200mA, much higher would indicate an issue on DC.   If testing with a cheap multi-meter start on the highest Amp scale (say 10A) and work down, otherwise risk blowing the fuse inside the multi-meter. 

 

Thanks, Nigel.  I don't think I can do the current consumption as my very basic Hilka analogue meter only seems to do voltages, not amps.

 

However, in the meantime, I tidied up the Lais installation, put the body back on the loco, gave it a long address, and put it on the layout.  And I'm astounded by the fact that it worked really quite well - smooth operation, much quieter than the Zimos had been, responsive throughout the 128 speed setting, and a very decent turn of speed. I was very tempted to leave it like that but I'm still going to check and probably renew all the electrical connection points on the loco just to ensure that so far as possible anynone of the problems you mention are cleared up.  Then I'll give the Zimo another go and if it still doesn't work properly I'll just use the Lais - maybe not the best, but quite adequate.  It's a pity that the Lenz failed as it did - it would have been interesting to see how the loco worked with it.

 

DT

  • Like 1
Link to post
Share on other sites

On 11/09/2020 at 14:23, Harlequin said:

Have you got any kind of interference suppression components fitted in the motor circuit? They can affect a decoder's back-EMF measurement.

 

P.S. You need to test decoders from other manufacturers before singling out Zimos in particular. My guess is that they would all show the same symptoms.

 

Youre absolutely right, these components aren't needed any more this isnt the 80s where locos interfere with analog  tv signals, Bachmann/Hornby should redesign their pcb's and remove all the unwanted components that cause problems with the decoder

Link to post
Share on other sites

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 account

Sign in

Already have an account? Sign in here.

Sign In Now
 Share

×
×
  • Create New...