Loksound install with Caps and Inductor

[Steveruger45]

Just installed a 73100 decoder in a Kato Dash 9 instead of a 73800 like my last one of these locos.
I put 2 x 220uF x 20v caps and a 100uH x 340mA inductor wired in series with the caps positive to see if this will solve the read/write issues of the lokprogrammer on a loco with caps installed.
In short it does.  I have not had any issues since with the lokprogrammer.
Here are some photos of the install including the details of the inductor I got from Digi-Key.
The inductor is only about 3mm cubed and only about 25c each.  I got 40 of these so I see a lot of soldering in my near future.
Now I know this has posted ok, sans photos of my computer screen shots of the lokprogrammer in action, I’d like to thank Pete for mentioning this possibility and that this is what Zimo recommends to do on their decoders.
Thanks Pete, this works a dream on the loksound decoder too.

BTW, in case it’s not clear in the photo, the speaker enclosure for the 9x16mm speaker is lengthened to increase cubic capacity by another 180mm3 under the rear of the decoder for a total of about 930mm3

[RBrodzinsky]

Steve, very cool. But, I have never had a problem with the LokProgrammer if I have only 440uF, or even 660uF of capacitors. I only start to see problems above that, and once I hit about 1mF, it is a complete no-go.

Can you try adding extra capacitors (with shell off, of course) as a test of the inductor?

[Steveruger45]

Hi Rick
I too only got issues on occasions without the inductor.

I can try as you suggest with more caps to find out.

But, I am noticing now some interference on the speaker on this install whereas my other similar loco with an 8x12 speaker and no inductor sounds cleaner.

 Not sure yet if it’s the speaker itself or the inductor.  It is an unshielded inductor.  There are shielded ones to be had.
So everyone, watch this space while I figure it out further

[Steveruger45]

Ok. I just now disconnected the caps and inductor and the speaker still sounds scratchy.  So I see I’m going to have to redo the speaker. Anyhow, it’s not the inductor so that’s a plus.

Rick, on the caps front,adding more to try, after thinking a bit more about that and RTFM a bit more I’m a bit concerned to do that due to a much larger inrush. Big capacitance should have the more complex three wire circuit so I’m going to bow out of that test.

[peteski]

Let me again recommend that you guys read the ZIMO decoder manual.  Towards the end of the manual it has some good examples of circuits for adding large keep-alive caps.
Here is the download: http://www.zimo.at/web2010/documents/MX-KleineDecoder_E.pdf

[Steveruger45]

/>
Ok, all is good I re-downloaded the sound file this time over a hard wire connection and re-uploaded it to the decoder in the loco fully installed with caps and inductor in circuit with absolutely no problems. 
It sounds good now.
I may turn the volume down a bit more.

Yes Pete I have read that manual a few times which is why I backed out of Rick’s suggestion to test with a larger cap group.

Anyhow I happy it works well so no more disconnecting the caps to install a new sound file in a loco.
Yea.

[peteski]

Excellent, excellent! 

[Steveruger45]

Here is a cut and paste excerpt from page 57 of the Zimo decoder manual specific to decoders without direct cap hook-up.  Which I felt was probably the most applicable to the loksound 73100 in this experiment.
Quote
Decoders not prepared for direct capacitor hook-up:
Here, a simple connection of a capacitor between the decoder’s ground and common positive is not advisable, since it could lead to undesired side effects: Software updates, sound uploads, programming in service mode (on the programming track) and the ZIMO loco number identification will be more difficult or even impossible. This applies to the decoders MX620, MX621 and MX630 as well as the sound decoders MX640, MX646 and MX647, which require additional external components to be added by the user in order to prevent above mentioned side effects.
Capacitor sizes of up to 220uF (possibly even up to 470uF) only need a choke coil (100mH / 100mA, also available form ZIMO) to enable decoder firmware updates with the ZIMO update module and ZIMO’s loco number identification pulses.
Unquote

Note, when you look at this section in this manual the drawings show the inductor (choke) as 100uH so I believe they have Misstated in the text as quoted above as mH meaning  milliHenry.

PDF of manual attached for reference


[ Guests cannot view attachments ]

[peteski]

Note, when you look at this section in this manual the drawings show the inductor (choke) as 100uH so I believe they have Misstated in the text as quoted above as mH meaning  milliHenry.

PDF of manual attached for reference

Yes, the inductance of small chokes/coils/inductors is most often stated in micro-Henrys ( μH or uH).  A 100 mH coil would have to be fairly large physically (since it would need many turns of wire). So mH is likely a typo.