Removing capacitors - original question

[GEfan]

When fitting decoders is it necessary to remove the capacitor? Am fitting to a Heljan class 15 but have seen some advice that suggests the capacitor may need to be removed to improve slow running.

 

 

 

[smokebox]

I don't automatically remove capacitors. I fit the decoder and see how it runs. If it doesn't run well then I would remove the capacitors.

There are some makes of decoder which just won't work properly if the capacitors are left in, (Zimo?)

 

I have Hornby, Bachmann, TCS and Lenz decoders and none of them have needed to have the capacitors removed.

[mezzoman253]

I don't automatically remove capacitors. I fit the decoder and see how it runs. If it doesn't run well then I would remove the capacitors.

There are some makes of decoder which just won't work properly if the capacitors are left in, (Zimo?)

 

I have Hornby, Bachmann, TCS and Lenz decoders and none of them have needed to have the capacitors removed.

 

Wise words, as this subject has a lot of urban myths attached to it.

 

In theory you should not remove capacitors at all as they are there to comply with electro magnetic interference suppression legislation. However some decoders have alternative suppression built in, which in theory makes them ok. I'm personally not convinced about that, but some decoder/capacitor combinations just don't seem to work, hence Smokebox's comment.

 

Rob

[Mr.S.corn78]

HI,

 

i have always removed the suppressors/capacitors when fitting decoders. Its not that much more effort to do it.

 

The new loksound V4 says you must remove them in the instructions so theres one that you have to do it.

 

cheers

Simon

[Lady_Ava_Hay]

This argument has raged for........well since the advent of commercially available DCC.

 

In general it seems best to remove them particularly from modern stock and even more particularly if cheaper decoders are being fitted.

 

It will invalidate the warranty to a certain extent and you will need to refit them if you take the decoder out.

 

Basically, they are not needed if a decoder is fitted and with modern can motors of low power consumption that don't spark as much as the old stuff, they are less necessary now anyway.

 

It does seem though that some models and some combinations of decoder and command station benefit most. Another reason to remove them anyway but don't expect improvement automatically or at all.

[James]

you will need to refit them if you take the decoder out.

Why would you?

 

I've removed them from a number of RTR locos which run on DC (should allow for quicker conversion later) and I've never seen anyone fit capacitors to kit built locos...

[hollywoodfoundry]

The output driver of a DCC decoder consists usually of a FET high frequency pulse width modulated stage. That circuitry really does not like to be connected to a capacitor, but it is possible to design it such that it can cope with an existing capacitor, but only if you know the value of that capacitor. As the radio frequency interference suppression network fitted to a model locomotive can consist of capacitors, chokes or a combination, then it is very unlikely that the decoder has been designed to cope with all of these types and values.

 

For this reason, most decoder manufacturers ask you to remove the suppression components, so the decoder is working in a known environment. From an electronics point of view, I would prefer to know that all my motor/decoder combinations were operating with an optimum load on them, so I remove suppression components as a matter of course.

[Dagworth]

four pages of serious discussion of the issue here http://www.rmweb.co....hp?f=10&t=39953 and also two more here http://www.rmweb.co....php?f=10&t=5675

 

I'll quote from the second topic linked to

Running will never be worse without them than with, but it will almost always be better.

Snip em out.

 

Andi

 

Andi

[Dagworth]

Andi, No ones questioning that that's true. It's the consequences of that action, and the potential affect on someone else, that I'm trying to address.

 

Rob

 

Ok, a different quote from me from one of the two topics I linked to

Any coil moving in a magnetic field will act as a generator, this applies very strongly to the DC motors in our models. The electricity generated by the motor tries to flow in the opposite direction to the electricity that is powering the motor, hence it is known as Back Electro Motive Force (back EMF). The decoder sends pulses of full voltage power to the motor, the length of the pulses being on relative to the gaps where the power is off being varied to alter the average voltage received by the motor and the speed it turns. The decoder looks at the back EMF generated by the motor to determine exactly how fast it is turning and adjusts the pulse length to try to turn the motor at the target speed being asked for by the command station. As the loco starts to climb a hill it slows down, the back EMF falls, the decoder compensates by making the pulses longer and the loco speeds up to the same speed it was travelling at before it came to the hill. Fine back EMF control can overcome sticky points in geartrains and chassis' allowing much slower control than would otherwise be possible. This relies on the decoder being able to tell exactly what the motor is doing at any given time.

The poor control that can be experienced with decoder equipped locos where the capacitors are left in situ is less to do with the capacitor acting as a short circuit or low impedance path around the motor and more to do with the capacitor affecting the ability of the decoder to accurately read the back EMF generated by the motor. The decoder is designed to sample this in the gaps between the pulses of power that it sends to the motor. If the back EMF is swallowed by capacitors the decoder sees the lack of back EMF as an indication that the motor is not turning and so supplies it with more power. This leads inevitably to loss of slow speed control. It can also lead to some very odd behaviour, I've seen locos that appear to have a kangaroo in the fire box purely because there is a capacitor in circuit, removal of the offending cap produced a totally different behaviour.

 

This is largely irrelevant as to the main point of this thread however which has been to try to establish how much of a problem the radio frequency noise generated by the motor causes.

In a DC loco the motor is directly electrically connected to the rails, which can act as a quite effective aerial to broadcast the motor noise. The caps short away the high freqeuncy noise, and then often there are two inductor coils between the loco pick ups and the motor as well. Coils act the opposite way to capacitors, they pass DC but block high frequency signals (the unwanted motor noise) so that any noise that does escape the capacitor is prevented from reaching the rails.

In a DCC decoder fitted loco there is no direct path from the motor to the rails, so motor noise cannot reach the rails as the decoder is in the way. The issue then is how much of the noise is radiated from the wires linking the motor to the decoder if we remove the caps and so stop the noise being shorted away in the first place.

 

Hope this makes sense...

 

Andi

[Dagworth]

Someone please answer:

 

Why did no Mainline Locos (at least the ones I bought) have any interference suppression components?

Made in Hong Kong not UK? Although my Mainline 56 I'm pretty sure did have a cap

Why did Triang/Triang Hornby/Hornby Dublo all have interfernce suppression components? (Even my 1950s items!)

see above

Why in current Hornby/Bachmann products that are DCC ready do they recommend you do not remove the components?

arse covering? Or to stop people snipping the wrong bits then complaining that the loco doesn't work...

 

Andi

[melmerby]

Here is TCS's take on it, Re a Bachmann HO loco:

 

"There is a capacitor across the motor circuit that drastically affects the speed curve when using a Quiet Drive decoder. It causes an over-speed problem. It must be removed from the circuit. Do this by simply cutting the capacitor out . The capacitor is typically yellow and has a marking of 105.

 

Technical explanation. The capacitors are placed in the circuit by Bachmann for the purpose of reducing the possibility that electrical noise generated by the motor when running would interfere with a TV or radio or cell phone or possibly some other electronic device. All TCS decoders have this capacitor built into the decoder and there for external capacitors are not needed. The TCS decoder is tuned to work with its own internal capacitor in operation. With external capacitors applied in the circuit the result is a de tuning which results in poor speed control. That is why the Bachmann applied capacitors need to be removed. The internal decoder capacitor reduces the possibility of electrical interference so there is no harm in removing the external capacitors."

 

Keith

[Nigelcliffe]

In the same vein as Keith's reply, this is Zimo's take on it. Zimo being European, so likely to have an interest in CE compliance:

 

General notes on installation:

Do noise suppression components on a locomotive motor have a negative affect on motor regulation?

Yes, sometimes . . .

Explanation: Motors of model railroad locomotives are often equipped with choke coils and capacitors, which are supposed to suppress or filter out noise (causing poor TV reception etc.), caused by the sparks arcing across the motor‟s brushes.

Such components impair the motor regulation. Compared to others, ZIMO decoders manage quite well and there is hardly a difference in performance with or without those components in place. However, in recent years larger choke coils are being installed in many locomotives than was the case earlier – and these can noticeably compromise drivability.

The potentially “harmful†choke coils are often recognizable by their shape, as they look like a resistor with color bands (in contrast to a wire wound ferrite bar). That doesn‟t mean though that these choke coils have a negative effect in all cases.

Lessons learned and accompanying measures…

ROCO, BRAWA, Hornby – so far present no problems, no action necessary.

FLEISCHMANN H0 with “Round motors†– choke coils are no problem; capacitors should be replaced if needed, especially the ones between frame and motor (may destroy the decoder if left in place)!

Newer Bühler motors present no problems so far.

TRIX H0 – choke coil between track and decoder plug should be removed!

MINITRIX, FLEISCHMANN PICCOLO – very inconsistent; removing of capacitors is often advantageous; choke coils on the other hand presented no problems so far.

Indications of an actual negative effect of such components, besides a general unsatisfactory motor control (jerking…), are:

- weak BEMF compensation: as confirmation, set the decoder for testing purposes to low frequency – CV #9 = 200 – and see whether the control compensation becomes stronger. If that‟s the case, the choke coils are most likely to blame for the weak compensation in the high frequency range.

- if a difference in compensation is noticeable between 20 and 40 kHz (selectable in CV #112, Bit 5); if so, it is very possible that the choke coils or capacitors are the cause.

Remedy: Bypass choke coils (or replace with wire strap)! Remove capacitors! Capacitors are less likely to interfere with motor regulations but cannot be ruled out (see “Round motor†above).

 

Discussing a particular Fleischmann motor:

Furthermore … especially on these motors it is recommended to remove the capacitors and bypass the choke coils at the motor. Attention: often the “worst†capacitors are the ones that are least visible and/or accessible, especially between the motor brushes and rail or frame.

[melmerby]

Interestingly the TCS installation mentioned was using a DP2X-UK, quite popular in the UK, although this was a US outline steam loco.

 

The FCC regs might also be something that needs to be complied with in US.

 

Keith

[spamcan61]

 

The FCC regs might also be something that needs to be complied with in US.

 

Keith

 

yep, FCC CFR47 part15 is the nearest equivalent to the European EMC directive.

[Dagworth]

This is the DC blanking plug from a Kato N gauge class 66.

 

The suppressor cap is the square blob in the middle. When you swap the blanking plug for a decoder you remove the suppression caps in the process, and have no choice about it. I wish other manufacturers would take this step too...

 

Andi

 

 

[Ron Ron Ron]

The suppressor cap is the square blob in the middle. When you swap the blanking plug for a decoder you remove the suppression caps in the process, and have no choice about it. I wish other manufacturers would take this step too...

If only.

Why in current Hornby/Bachmann products that are DCC ready do they recommend you do not remove the components?

In fact Bachmann recommend removing capacitors. From their own website..........

Remove the interference capacitor from the board circuit if it is fitted. The AC nature of the DCC signal on the track side of the decoder will mean that your controller will detect the capacitor as a short circuit and a programming error will result.

The DCC decoder contains suppression circuitry.