Model Railway Forum banner
1 - 10 of 10 Posts

·
Registered
Joined
·
52 Posts
Discussion Starter · #1 ·
Hi all,

This post was originally made to the RMweb forums (no particular reason why I chose to post there first!) and following the very positive discussions and feedback from that I have spoken to a moderator on here and obtained approval for continuing the discussions over here. Apologies for its length!


I've updated the post from the one in RMweb simply to keep it current and correct following the discussions that have already taken place. If you want to follow that discussion as well then the link is here:

http://www.rmweb.co.uk/forum/viewtopic.php...p;sk=t&sd=a

I guess this is probably going to kick off a flamewar but I feel it has got to be covered for the reasons covered below. If Moderators feel i'm being just too "picky" about this then feel free to bin the post. It is NOT intended to be pro-Hornby, anti-anyone else or any other sway, the topic is NMRA Conformance and having spent quite a bit of time conversing with the right people within the NMRA in the states, the people that look after the standard, manage the conformance warrants and do the testing, I feel i've built up a firm understanding of it now and i'm shocked and amazed at how little this is reflected in the manufacturers and forum members.

I will also add one last caveat by saying that just because a system does not have a Conformance Warrant does NOT mean it's rubbish, it does NOT mean it's any less compatible than any other system and i'm not intending this to imply that - it just means you don't *KNOW* it is, it's not *CERTIFIED* to be, and my basis for this post is that as a consumer I feel we should all be in a position to *KNOW* this as a fact so that we do not have a repeat of the disaster with the first gen Hornby decoders and their Select system.

I have noticed that "NMRA Compatible" and "Fully NMRA Conformant" are phrases often bandied around by forum members and manufacturers alike and I can't help but think that your average Joe is going to get very confused by this and end up buying something that they thought was tested and certified but actually was not.

There are a number of problems with NMRA Conformance of DCC at the moment. The first is that not enough manufacturers actually seem to understand the confidence and value this offers to a customer that their purchase is going to be compatible going forwards, there are simply very few manufacturers actually participating in NMRA Conformance testing, though it is on the up, very very slowly.

The second problem is that there is a lack of understanding about how long a warrant is valid for or for what equipment - well, the simple fact is that it is valid, indefinitely, for the life of THAT VERSION of the equipment. Change the equipment and the new version does not have a conformance warrant. I posed two specific examples to the NMRA which I will briefly summarise now:

The Hornby Elite
At Firmware versions 1.0 and 1.1 it did not have a conformance warrant. However, for version 1.2 it was tested and certified and if you look on the C&I page at the NMRA you can verify for yourself that Hornby Elite Firmware 1.2 has warrant number 2007-1011. However, Hornby then updated the firmware again and issued version 1.3 and I asked the NMRA if this would still be certified - it is NOT. A change requires a re-certification. This is not my opinion, this is fact from the NMRA.

NCE PowerCab
The NCE PowerCab is based very heavily on the NCE PowerHouse Pro, essentially miniaturising the command station and putting it in to the same cab that you use with the PowerHousePro. The PowerHouse Pro has Conformance Warrant 2005-011. So, is the NCE PowerCab therefore certified? It is NOT. It is a different unit and the NMRA view is that it requires testing of its own to obtain its own warrant. This is not my opinion, this is fact from the NMRA.

In another communication with the NMRA they let me know that they have recently introduced a new self-certification process whereby manufacturers can actually now test their equipment themselves - they have always made the specs and hardware diagrams available anyway, now the manufacturers can actually run the tests in house, produce the documentation and the NMRA simply then review it and either request further testing or issue the warrant right away if it all looks good. What surprises me about this is that when i've spoken to Dave Nichols at ZTC at various exhibitions, he has told me that they built their own NMRA test kit from the specs and they test to the standards - they just didn't send the kit off for the NMRA to do the testing and therefore it didn't have the warrant. But with this new process, surely all the ZTC kit can now quite quickly gain full certification? One wonders why the effort has not been made, but this points back to my initial comment that the manufacturers do not seem to feel the value in NMRA conformance and therefore only cursory glance is made.

Look through the warrant list, look at the names missing - there is no ecos, no digitrax, no mrc/gaugemaster etc for example. Major brands with well respected gear, but none of it with a conformance warrant. Credit goes to NCE for having a warrant on their Power House Pro unit of course.

I strongly urge everyone looking at equipment to find the kit that best suits them and then verify it has a warrant, if it does not have a warrant then tell the manufacturer that you will only buy certified kit and you really want theirs to be certified. If enough people do this then I feel we can turn things around and "NMRA Conformant" can actually mean something and provide us end-users with the level of confidence that we actually want when we're spending money on DCC systems.

Very little of the above post is my opinion, it has all come out through discussion with the NMRA specifically with the people involved with the DCC certification process.

I'm happy to answer any questions, pass them on or you could even just write to them yourself - I didn't get any special handshake, I just wrote to their normal contact addresses to confirm that the state of certification was as dire as it seemed looking through the issued warrant list, and discussions have followed on over the last year or so since then.

Here is a link to the Conformance and Inspection Warrants Issued list, if it is not on this list it does NOT have a warrant:

http://www.nmra.org/standards/candi/warrants.html

It is also worth re-iterating the difference between being technically conformant and being certified conformant with a warrant. There are probably many pieces of kit out there that are technically conformant but are not certified as such and we should all be chasing up our preferred manufacturers and asking them WHY they aren't seeking certification from the NMRA. If there's a good reason why not then rather than just avoiding it they need to be working with the NMRA to resolve these issues - such as the former FCC requirement (see below).

Now, if you have an NCE PowerCab, contact NCE and find out what they're doing about getting a warrant for the kit and if there's a reason why not then let's work to resolve it. It looks like a fantastic bit of kit and is let down by not having that magic warrant number.

It's up to us as consumers to make sure that vendors know that certification equals money because right now I think that consumers are not pressing the issue enough - if getting certification will not sell them any more units (or continuing to not have it won't cost them sales) then there is no business case for the money and time required to get the certification.

FCC Certification Requirement

After speaking with Dave Nichols at ZTC at a show he told me about the bizarre and unreasonable requirement to have an FCC certification in order to even apply for DCC certification. Just to expand on this for those that aren't aware, all electronic goods must have the CE certification for sale in Europe or FCC certification for sale in the US, it doesn't matter where it was made just where it's being sold. So obviously a manufacturer in europe who is not selling to the US doesn't need (or want) to go through the expense and time of an FCC certification just to get their DCC warrant.

I raised this with the NMRA and they pledged to investigate and review it, next time I brought it up it was confirmed as removed when the guy emailed me back almost straight away:

QUOTE The NMRA Board of Directors agreed with my recommendation to remove this requirement from our compliance program. Therefore, an FCC report is not necessary to be issued an NMRA Warrant. This does not mean the product is exempt from having an FCC certification. It only means NMRA is not concerned about it.

Please be advised that the FCC does recognize the CE certificate from Europe.

Conclusion

As consumers I want to understand why we are not more demanding of our favourite manufacturers to get their equipment certified. The NMRA want to improve adoption but are struggling to see how to move it forwards - my answer is to get consumers fired up about it and have them demand it from their manufacturers.

I would be very interested to hear from people who simply don't care about the conformance warrant, why that is and their understandings of it - perhaps there is a genuine failing in the standard, perhaps there is a failing in understanding - I am not here to preach that it's perfect, but to say that standards are a GOOD thing and if we aren't adopting it because it's flawed or wrong, we should be FIXING it. If there's nothing really wrong with it then we should all be demanding compliance from our preferred manufacturers and asking them why they aren't - perhaps they have good reasons, let's find them out and try to improve things.

So why am I getting my knickers all twisted about this anyway? Because we go out to lots of model railway exhibitions all over the country talking as an independent about DCC. Since we're not a vendor or a manufacturer we can give completely impartial advice - and our advice is ALWAYS to check the Warrant list and ONLY to buy kit with the conformance warrant because it provides that level of confidence about the compatibility of the kit. The problem is, there's next to nothing on the list at the moment! We do talk about non-warranted kit but only really in terms of encouraging them to try it out and then talk to the manufacturer about getting it certified if they like it.

I hope the discussion will proceed positively and constructively with a view to improving the standardisation situation for DCC, thanks for your time reading and contributing.

Matt.
 

·
DT
Joined
·
4,794 Posts
We have talked about this a few times before. I think that it is interesting that warrants apply to a particular version of firmware.

It is obvious that some brands are using a warrant for a past product or similar to justify claiming conformance for current new products.

It brings into debate why do we need the NMRA - is it more trouble than it's worth... But if I may answer my own question with one point, Hornby for one would be quite happy not to have anything to do with the NMRA (they have said this openly) and as with the Scalextric digital system which is proprietary to them and not compatible to any other digital slot car brand, I'm sure that they would be just as happy doing their own version of DCC were it not for the masses of DCC users who demand some sort of conformity.

I'm amazed that Bachmann have said that ESU are able to self certify and that explains why there is no warrant listed under Bachmann for the Dynamis, but then they go on to say that they are working now to get one...
 

·
Registered
Joined
·
1,058 Posts
Why do we need the NMRA? Hornby Zero One and the incompatible Airfix Multiple Train Control System should be reason enough in my opinion.
 

·
Just another modeller
Joined
·
9,983 Posts
HI Matt, I enjoy your positive enthusiasm for the subject:

As one who is directly involved and who also worked damned hard sometimes quietly sometimes not to get Hornby off its high horse and onto the DCC bandwagon for several years I think that the horrible experience of Select and hornby decoder owners is proof positive that a global standard is very desirable.

quite sophisticated Analog multi-train control and more latterly DCC variants had several false starts over 30 years or so, and if a standards based approach had been followed in those early years, it'd be in a very advanced state right now.

Doug, I really do think that the answer to your question is moot: Digital control would be a fragmented and sad shadow of where it is today without it. Warrants by the way do not cover the firware, but are based on the nature of the physical outputs + the consistencey of a very small part of the the command and communication structure. every firmware change or revision does not invalidate a conformance warrant unless it changes these factors negatively.

To clarify:
Conformant: has been submitted for NMRA testing and passed the tests. a system cannot be called conformant unless its been tested by thenmra themselves.

Compliant: has been self tested and is warranted to comply. a product cannot use the term compliant and the dcc logo unless it meets the standards. if a Mfr does this then the NMRA will get on their case.

The CE and FCC issue.

The standards actually already allow for this and it has been acknowledged for quite some time that the system needs to comply with the local market regulation in which it is sold in order to be compliant or conformant. Removal of the term FCC is a good idea, but its presence has not stopepd any product being certified if it was not to be sold in USA.

FCC cerification is actually less rigid and draconian than CE testing by a country mile.

Why don't many companies go the whole way and do the conformance: basically NMRA is a volunteer organisation and NOT a regulatory body.

Its a volunteer manned organisation so its very structure makes it very difficult.

Timing has been slow and manufacturers who fully understand the specification have peferred to make damned sure it is compliant before release rather than accept an added significant delay due to conformance testing.

Basically the move to create a self test platform by the NMRA was an acknowedgement of this, and when completed will be a pragmatic and practical answer to much of the problem.

Its still under very delayed development but when finalised it will provide a complete test and analysis platform nad the process will require a set of data to be created utilising the self test equipment which would then be duly notarised and registered.

Manufacturers are keen for this to happen and are supporting the move BUT the contractor engaged by the NMRA is a light year behind on delivery. I should add its not going to be a cheap bit of kit either, and will be a stretch for the smaller businesses responsible for much of the positive evolution of DCC accessories!

You should not underestimate the ability of the market and the NMRA, volunteers or not, to act if a Mfr misuses the conformance claim ability - NMRA does not want to be and cannot be a "DCC police" but there are caring and responsible people in every part of the world that can and do act with good success where problems arise.

The problem is only where the arrogance of the manufacturer tries to take advantage of the term DCC and then openly flaunts its specification. This is what Hornby did with its initial releases until eventually, publicity and market forces - largely YOU as users, let it be known that it wasn't going to be accepted.

Yes... there are many who either use both or just prefer compliance only because it doesn't delay them but have you noticed it is also they who often lead the pack in design and feature innovation. NCE, Digitrax, ESU(ecos), almost all EU suppliers except lenz (who is currently at the back of the pack software-wise anyway).

They do it not because they couldn't be conformant (because universally they can...) but because they cannot afford to wait to bring a product to market and lose the timing edge that is critical to all marketing.

Be aware too that you need to be careful what you wish for :).

Standards are actually evolving and WILL change constantly. - what complied yesterday may not tomorrow. good example - the beloved 8 pin plug will be phased out and no longer be a standard.... in favour of a whole suite of new ones - good for the Mfrs, a bitch for DCC modellers for a decade to come I predict!

NMRA is currently working to try to bring the reluctant ones to heel and would love a more complete set of brands who are conformant but is actually comfortable with compliance as long as its properly done

final comments: Compliance and conformance are actually dead easy to achieve.

For decoders there is only the very basic things such as CV29 and a couple of others such as DC/DCCcompatiblity and minor compulsory CV use... interestingly guess which decoder brand fails even this?

Standards are also very narrow and minor in command station terms in that they really only cover the quailty of the waveform and primary issues like voltages at track etc.

The amazing thing is that product so far off spec when released they are not even entitled to be called DCC can EVER be created- and without naming names, both still are techincally well off being "fit for market".

Standards should be throughout: What is truly needed is:

(1) that we vote with our feet and do NOT accept partial solutions such as H's or attitudes such as SK's. - where one product is now conformant because it was in real terms forced to be by market attitude but others are still unfit for sale.

(2) that the standards continue to evolve - Current DCC still revolves around & uses 1980's technology at its core and its way past time for the whole of the standards to be revisited and adopt initiatives that have already been possible for a long time. As a comaprison take a look at the bus specification for ESU's MFX (developed for marklin) and traditional DCC.

(3) Things like long address and aliasing must become part of the standards, as should things like both regulated methods for programming on the main and programming track, unrush currents for sound decoders, absolutely rigid use of higher functions (MRC gaugemaster for example fail to follow standards there) and most of the current "core RPs" (recommended practices) should now become standards, not recommendations.

many of these things are under active consideration, as is the very nature of the DCC bus.

Matthew: You should apply to join the NMRA working Group if you have not already done so. Standards can only work if the energy is applied properly at the roots. It may be frustrating making progress but it can only come from energy applied within. Such things are better done from within than from a soapbox.

Richard
DCCconcepts

PS: I've said my piece on this and similar subjects many times, and the above sums up my feelings. I'm happy to respond to issues where there is value in it, but l don't want to play tennis with the subject for the sake of it.
 

·
Just another modeller
Joined
·
9,983 Posts
QUOTE (Matthew Peddlesden @ 10 Apr 2008, 19:59) <{POST_SNAPBACK}>- and our advice is ALWAYS to check the Warrant list and ONLY to buy kit with the conformance warrant because it provides that level of confidence about the compatibility of the kit. The problem is, there's next to nothing on the list at the moment! We do talk about non-warranted kit but only really in terms of encouraging them to try it out and then talk to the manufacturer about getting it certified if they like it.

I hope the discussion will proceed positively and constructively with a view to improving the standardisation situation for DCC, thanks for your time reading and contributing.

Matt.

***One small point there Matt:

I totally agree with the posititive intent of the concept but beware - some conformant product is markedly inferior to compliant product of other brands, so using the existence of the warrant as the sole arbiter of choice at this point in time is not necessarily doing the punters a positive service.

I DO most sincerely look forward to a time when that will not be the case, however I think some things have to change before it becomes correct - the warrants need a use by date added, they need to be far more universally applied for and the standards themselves need to be updated to adopt todays DCC possibilities as compulsory rather than optional features or abilities.

I strongly agree with lobbying at every opportunity to encourage Mfrs to apply for conformance though - that goes without saying.

Richard
 

·
Registered
Joined
·
52 Posts
Discussion Starter · #6 ·
QUOTE I'm amazed that Bachmann have said that ESU are able to self certify and that explains why there is no warrant listed under Bachmann for the Dynamis, but then they go on to say that they are working now to get one...

I think perhaps there is some confusion about self certification...

Self Certification is a process that lets the manufacturer do all the testing themselves, they produce the reports and so forth - and then it must all be submitted to the NMRA who will then issue the conformance warrant once they've reviewed the documentation and are satisfied that it is complete, so the NMRA are still the ones issuing the conformance warrants and this means that your equipment is still not able to display the logo until it's received one from the NMRA.

Richard - apologies for picking your post apart, I just want to reply in-context to save me waffling more than necessary


QUOTE every firmware change or revision does not invalidate a conformance warrant unless it changes these factors negatively.

Hmm, this is at odds with the reply I got from the NMRA, I posed the following question:

QUOTE Lastly, I wonder if you could clarify the continuation of warrants across versions. Presumably equipment at a particular version is tested but I don't have a way to know whether the equipment has been significantly changed since a warrant was issued - e.g. the NCE warrant was back in 2005, presumably three years later there's been some hardware modifications so do these require re-testing and how do I know this has taken place? I note that the Hornby Elite is listed at "version 1.2" - what happens now that they have released 1.3 firmware, is this version not covered by the warrant or is it assumed that once euqipment has received a warrant then all future versions are covered?

and the reply was simply as follows:

QUOTE Changing a version number requires retesting.

QUOTE Conformant: has been submitted for NMRA testing and passed the tests. a system cannot be called conformant unless its been tested by thenmra themselves.

Compliant: has been self tested and is warranted to comply. a product cannot use the term compliant and the dcc logo unless it meets the standards. if a Mfr does this then the NMRA will get on their case.

I guess this is part of my reason for wanting to understand things more simply - word-games like this are what confuse the general consumer, even more so when you throw in "compatible" with a further different meaning, the poor consumer now doesn't really know whether it's warranted or not.

Bottom line from what I have understood from NMRA is that it's very simple:

The NMRA issue conformance warrants. Conformance warrants can only be issued by the NMRA.
Manufacturers can complete the testing and submit the results, but cannot issue their own conformance warrant.
You can see if a piece of equipment has a conformance warrant by looking through the issued warrants list at the URL I gave earlier.

QUOTE For decoders there is only the very basic things such as CV29 and a couple of others such as DC/DCCcompatiblity and minor compulsory CV use... interestingly guess which decoder brand fails even this?

Well, there are also a slew of electrical tests to pass as well but yes decoders should be relatively easy to put through testing.

QUOTE They do it not because they couldn't be conformant (because universally they can...) but because they cannot afford to wait to bring a product to market and lose the timing edge that is critical to all marketing.

Then what I would ask the manufacturers is why they can't go ahead and bring it to market after their own testing is completed, and once it's out there then seek the NMRA warrant (which should just be submitting paperwork) which can be appended to their sales literature once it has been obtained - a couple of months delayed release definitely can hurt sales, but a couple of months after release adding the warrant number shouldn't cause any problems at all?

Thanks for the input Richard, I was hoping you'd chip in with your valuable experience!

Regards
Matt.
 

·
Registered
Joined
·
52 Posts
Discussion Starter · #7 ·
QUOTE I totally agree with the posititive intent of the concept but beware - some conformant product is markedly inferior to compliant product of other brands, so using the existence of the warrant as the sole arbiter of choice at this point in time is not necessarily doing the punters a positive service.

Well we certainly point that out and I have no problem at all in recommending that people wander around a show and talk to the manufacturers and vendors, try the equipment on the stands out and find what's good for them. There's so very little that actually DOES have a warrant that limiting to that is almost impractical unless it's a small layout suitable for an Elite or Dynamis really. But the recommendation is, if you find you like Digitrax (for example) then get in touch with them and tell them you want their kit conformant - more people start doing this and this will raise the "value" of the conformance warrant and we'll see more manufacturers considering it for their new kit; so chances are it isn't going to help them for the system they buy today but it might make it much more likely that their next system has a warrant.

Cheers,
Matt.
 

·
Registered
Joined
·
2,151 Posts
QUOTE (Richard Johnson @ 10 Apr 2008, 14:45) <{POST_SNAPBACK}>Why don't many companies go the whole way and do the conformance: basically NMRA is a volunteer organisation and NOT a regulatory body.

Its a volunteer manned organisation so its very structure makes it very difficult.

Timing has been slow and manufacturers who fully understand the specification have peferred to make damned sure it is compliant before release rather than accept an added significant delay due to conformance testing.........

........ NCE, Digitrax, ESU(ecos), almost all EU suppliers except lenz (who is currently at the back of the pack software-wise anyway). They do it not because they couldn't be conformant (because universally they can...) but because they cannot afford to wait to bring a product to market and lose the timing edge that is critical to all marketing.

Richard
DCCconcepts

Thanks Matthew and Richard for enlighting the likes of us.

The above I believe pretty much sums it all up.

Baykal
 

·
Just another modeller
Joined
·
9,983 Posts
Matt - I am an NMRA member, a registered DCC Mfr and part of two of the DCC working groups - so I have no confusion or grayness about this issue at all.... I agree its a bit "jelly like" in interpretation (as are some of the RP's in fact) but the intention is to certify wihout creating an unusual burden on the Mfr - so there IS and there MUST be a degree of acceptance of integrity on behalf of the Mfrs.

If it was perfect, the terms wouldn't actually matter as we'd not even need to discuss it!

I had almost the same conversation as you with senior members of the NMRA quite some time ago, and again with the NMRA president (touching largely about the difficulties of a volunteer organisation creating mandates for Mfrs) by email 12 months ago.

Software versions change in decoders sometimes ten times a year and in "online updateable" DCC systems aspects of the code will change and be refined many many times per year... Not in substance, but in detail. In the PROM based systems some code will almost certainly change in detail change batch by batch. each will end up with a slight version number change...and this has never been considered to require a re-test. Bear in mind we aren't talking wholesale differences or anything that affects the primary DCC signal at all in 99.9% of cases.

Therefore there will always be changes from the precise unit submitted to the NMRA, however these changes will usually be to enhance and improve, not to take the unit away from the standard... All are positives and no sane Mfr who really wants to do well will play games with the standards, and those products which prove less than compatible in the real world are very soon reamed well and truly by the consumers as they should be.

There DOES need to be a mechanism to cope with it comfortably now more are moving from PROM to purely software based or flash upgradeable devices. THAT is why I said that I think that certification should have a date base or "use by date"

- so re-testing could be done routinely and to a schedule that both sides could cope with without adding too much of a burden anywhere.

Incidentally, the "family product" concept also exists in the quite draconian CE regulations - for example an audio amplifier range or CD player range only requires ONE product to be fully tested and the others can be covered by self declaration up to a point. This was added as a pragmatic way to prevent cost blowout for compliance and encourage Mfrs. A good idea!

The BIG advantage that CE has is that in case of naughtyness, the fines are huge - I wish the NMRA had such power (cause then they'd also have the same budgets and be able to test quicker)

In the end I wonder how it will pan out.

Ideally the self test which is done by the "in process of being created" NMRA created test gear with dedicated software embedded is a great answer, because results can then only be acurate so self testing would have a built-in "judgement" and Mfrs could use it for pre-evaluation too, which would make the Mfrs job so much easier, increase participation in the programme and leep time to market much quicker!

Re the conformance and compliant thing. I actually agree with you but the NMRA created the two terms in the first place, during a period where they were looking for cooperation but feeling toothless..

NMRA permit and in some ways have in the past encouraged the use of the term DCC compatible but the NMRA DCC football has only ever been permitted to adorn packaging if the product has passed NMRA testing.

THIS issue is also being (or is supposed to be being) revisited in an attempt to take firmer hold of the situation right now, but will still take a while to come into effect. Very few Mfrs ever abused this issue.

Hopefully, a fresh global DCC logo with a much more significant meaning will evolve!

Interestingly Hornby was the ONLY major brand in memory apart from MRC to misuse the DCC concept and branding so badly and miss the mark by so far - so getting them back in line was important... So while I am still of the opinion they have to fix the other products properly, I'm very very positive about their new found approach to elite.

I smile at their option of "classic and Standard DCC operation" though... cheeky buggers, how can a cockup be "classic" :) :)

Richard
 
1 - 10 of 10 Posts
Top