ESU 5 decoders, which one?

[GM50 4164]

I know the 58813 is the multi-protocol version while the 58820 is the DCC version. What are the pros and cons of each and which one are you guys using?

[reinhardtjh]

Dave Heap on either the ESU-Loksound or JMRI message list explained the differences very well and if I can find his post, I'll copy it.  But basically, if you want to consist your locos with other USA make decoders (Digitrax, TCS, NCE, BLI, Bachmann), it's much easier now with the DCC version of the LokSound 5.  This is because ESU in the V4, Select and the LokSound 5 Multi-protocol decoders did not follow NMRA guidelines and the rate of momentum change is several times slower (or is it faster, I can't remember) than decoders which follow NMRA DCC guidelines. 

So I found one mention:

There are several significant differences between the 5 DCC and the 5 Multiprotocol (DCC/MM/SX/M4) decoders:

1) The Multiprotocol decoders use the Eurocentric 0.25 momentum multiplier while the 5 DCC decoders use the NMRA-specified 0.896 multiplier.

2) The Multiprotocol decoders do not support the 3-point speed table (CV6 is not available, 28-point speed table is always enabled) while the 5 DCC decoders support both 3-point and 28 point speed tables.

Item (1) breaks conformity with the V4.0 range but restores conformity with NMRA-compliant brands and allows for the longer acceleration/deceleration times we prefer in the USA/Aussie market, reflecting our different layout style from the UK/European market.

So if you have a lot of USA made decoders you may want the DCC version, whereas if you have a lot of ESU V4 and V4 Select or European sourced decoders, then you may want the Multiprotocol version.  On the other hand the fact that the street price of the DCC is about $15 cheaper might sway you that way.  But the "loose wires" of the 58813 can be a benefit also.  Supposedly ESU will be coming out with a "loose wire" version of the DCC decoder in the near future.

[jdcolombo]

I don't fully agree with John's conclusions.

The major difference as far as I'm concerned is that the Multi-Protocol decoder does NOT implement the 3-point speed curve (there is no CV6).  I use the 3-point curve all the time, so this is a deal-killer for me when it comes to the differences.  The 5 DCC version does have the 3-point speed curve.

While the 5 DCC version does, in fact, change the momentum "scale," you can easily compensate for that by changing the value you use in CV3 and CV4.  What should work is to take the value you use for the Select, divide it by 3.5 and you'll be close.  For example, if you use a value of 20 in your Selects for momentum, a value of 6 in the new 5 DCC should be pretty close (20 divided by 3.5 = 5.7).

What you can't compensate for is the complete lack of CV6 in the multi-protocol version, IF you are fond of using the 3-point speed curve.

Plus, the multi-protocol version is several dollars more expensive than the DCC version.

John C.

[reinhardtjh]

John C makes good point.  It really comes down to economics (the DCC only version is cheaper) and which fudge you are more comfortable with.  You can make the momentum rise/decay rates nearly identical by the calculations he gives, but if you want to go the other way, you can also compensate for the missing CV6 by programming the speed table appropriately.  By hand that would be tedious, but with JMRI or the LokSound Programmer software, it's fairly easy.

Personally, I'm going with the DCC only decoders.  Mostly for the price difference. Most of the LokSound 5's I have now are targeted for steam locomotives so the flat ribbon cable is less of a hindrance.  Hopefully by the time I have a diesel hood unit they will have the loose wire version available.  Or else I'll cut the flat ribbon off the connector and get better at soldering to the tiny pads.

[jdcolombo]

Yes, John is right: I should have said that you CAN compensate for CV6 by using the 28-point speed table, but I find that more of a pain than dealing with the momentum conversion.

I’m using the DCC versions, too.  Like John, right now all I have left to convert are a few streamers.  But also praying for a hard wire version, like they have for the regular-size 5DCC, for potential future diesel applications.

John C.

[RBrodzinsky]

At least the new pads are larger than the ones on the V4/Selects!  And all pads are on the same side of the decoder.

And just a comment on the "calculation": if the fractional value in accel/decel rates is an issue for matching the 5 DCC with V4/Selects/LokPilots, one can always do a minor adjust in the rates of the older decoders, so that the division results in an integer value. (BTW - the factor is 3.56, or 3.6 if you want to round to nearest tenth)

On the ESU LokSound Yahoo group, Matt Herman did say the alternate wiring options are coming, soon, for the DCC model

[GM50 4164]

So the 58820 has the ribbon cable, I get that but there is also the pad with additional solder pads there where the speaker wires are as well. What are these pads used for? Are these duplicate soldering pads that the ribbon cable uses? How are you all attaching wires to the NEM-652 plug? Are you cutting them off?

[RBrodzinsky]

The pads duplicate all the wires in the ribbon. You can cut the ribbon off right at the edge of the connector and solder wires to it, for a hardwire installation.