Guys,
I’d really appreciate some advice please.  I have a track that’s nearing completion and I’m moving on to mixing and mastering.  The problem I have is that due to living circumstances I have to use headphones 90% of the time and the rest of the time the only other option is the hi-fi in my living room.  I had thought I had a pretty good mix on my headphones but when I listened on the hi-fi it was absolutely terrible!
A word on kit – the headphones are Audio Technica M50’s, so they should be pretty good and they’re coming out of my Roland UA-25EX audio interface.  The hi-fi is very good and I always thought natural to warm sounding.  Speakers Spendor S8e floorstanders (£2k), Musical Fidelity amps (pre + monoblocks - £2K) i.e. it’s fairly high-end British stuff, not exotic.  Connected to the UA-25EX via a (not ideal!) long unshielded interconnect.  Obviously my living room is totally untreated acoustically.
The song started out on Maschine and all tracks are now imported into Sonar Plat and I have available Izotope Neutron and Ozone 7 that I just bought.
I should also say that I know that my ears are not well trained.  Sometimes when I listen to YouTube tutorials on mixing/mastering, I have difficulty hearing the difference!
Where to go from here?

1. Should I trust the headphones or the hi-fi?
2. I saw a tutorial on mixing with pink noise.  You solo each track in turn against a constant pink noise and reduce the volume until the track has just disappeared.  Finally turn off the pink noise and enable all tracks and the mix is virtually done.  Anyone tried this?  Also comes back to question (1) – should I try this on headphones or hi-fi?
3. Would you put Neutron on every single track and how much should I trust the track assistant?
4. Mastering is some way off at the moment, I’ve only used modest eq and boost11 before.  Any advice on using Ozone 7?  I do plan to export the whole mix to a simple stereo track before feeding it into Ozone, but haven’t planned any further than that.

Sorry if this is RTFM territory, I know it’s a huge subject, but the track is for a competition and I’d really like to do well!

Do lots of comparisons with professionally mixed tracks in a similar style (instrument setup). The point is not to get (for instance) the guitars to sound the same but it will give you an idea of what a good instrument balance is and how your lows/highs relate. I often find I have too much mid in my mix when I get to that phase.

I would say the answer here depends on your true focus. From what I'm understanding, you have two goals:
1. Learn mixing, learning mixing in Sonar, sort out the best monitoring option using what you have available.
2. Submit this particular song to a competition.
 
Honestly, if #2 is the most important goal right now - and you have the limitations of inaccurate monitoring tools and are new to mixing - I'd bounce the tracks and have someone else mix. Then send out for pro mastering. This isn't to doubt your ability to learn mixing, just trying to be practical. I've mixed for many years now, and if I ran into the monitoring situation you have while entering a song competition, I would find someone else to mix.
 
That said, if the competition is secondary, or you really are determined to mix for the competition, I would probably use the the hi-fi system to mix. There are inherent issues with this! This system sounds too good, probably deals with lower frequencies much better than the average listener's stereo, and does other processing that will make a mix sound better. As Sanderxpander said, I'd listen to a ton of similar music through those speakers to get your ears tuned to them, and then reference back to those songs frequently while you're mixing. Then I'd use your headphones as a comparison. They won't support the same frequencies, but you'll most likely hear if horrible mids or highs are too dominant in the mix. Work slow, compare, make small adjustments, repeat. If you can use your car stereo, it's usually a great reference as well because lows and low-mids will generally punch hard and sound pretty bad if they're too boomy. Plus - you've probably listened to a lot of music in that stereo and will be most familiar for comparison.
 
For mastering - again my first recommendation since you have the competition is to have someone else master it. Ideally, find someone who will master from stems instead of a stereo track so they have a little extra control to help correct anything really weird in the mix. I should stress that mastering shouldn't generally be used for mix correction, but nonetheless. Find someone who will be cool about it, and let them help you by giving them stems and make some adjustments.
 
If you really want to do it yourself, here are my recommendations.

• Get your mix as close to the right sound as possible before mastering. Don't rely on mastering to finish your mix.
• Because of the monitoring limitations, I'd be careful applying presets that give a large color change to the song. In Ozone, "Mix master" seems to generally work well for some leveling things off and level boost without crazy color. The 4 band presets toward the top of the list can be okay too, but add more color.
• Due to the monitoring limitations, I'd recommend not pushing the mix too loud. It'll be nearly impossible to get a commercial volume track with bad monitoring without squashing the song. I'd just accept it will be quieter, people will need to turn it up a little, but you'll have a better sound overall.

I hope that helps!

As other have said, use reference tracks to make them similar in stereo imaging and frequency.

I myself am a Neutron user and love it. I don't normally use track assistant because I gee rally apply my own compressions and EQs to what I like, granted I'm in a treated control room. I use Neutrons Eq sidechain aloot. Especially because I mostly deal with hip-hop artists that don't have tracked out beats can make it difficult to create space for the vocal. So basically I only use Neutron on vocal tracks I want to do EQ sidechaining with. Other tracks I use other EQs/compressors.

As far as your monitoring situation (which iv been in similar situations). Mix it, do the infamous car test, play it on your Hi-fi, listen in other headphones and notice the differences/similarities. Mentally note or write down the changes you wanna make and change them. It's like Kentucky Windage... but with Audio.

Some good advice here, of course.

I mix on headphones. Have done for many years. I do have a pair of nearfield speakers but I only use them as a final check. I also check, when not in lazy mode, on my TV sound system.

One thing I always do now, though, is put SPAN on my master buss to watch out for troublesome frequencies.

I gave up "mastering" a couple of years ago. My ears are not up to the task... And neither are my skills, I hasten to add.

Mastering is a very subtle art, if you don't have a great mix, you may have to do too much to get a good mastered version of the mix, compromising the outcome if you go too far.
 
Though, you're always in control of the recordings you make, and how the sounds are mixed, it's really about what outcome you want. For example: an acoustic guitar can be mixed very bright, or very subdued, depending on how important it is in the song idea . . . and especially, how loud it needs to be in relation to everything else.
 
As far as mixing and mastering "90% on headphones"  . . . the most difficult challenge would be to get a proper tonal balance between the bass and the treble. 100 hz sounds "different" on headphones than it does in an ambient room.
 
Headphones are great for fine tuning, but you can be fooled by the closeness and ambience in your head. As others have stated, it can't be done quickly if you are new to it all, you ears gain experience the more you do it . . . and the most important thing is to listen in as many sonic environments as you can, since it sounds like you don't have access to a "tuned" studio room . . . car, truck, living room stereo, earbuds, headphones, small speakers, big speakers, indoors, outdoors, listen to it loud, listen to it quiet, early in the morning, late at night.
 

One plugin that can help is WAves Virtual mix room, its actually really good if your purely confined to mixing on headphones. It helps make the sound in the headphone replicate a room and monitors. It's not too pricey. I believe it's on sale now for like $70. Not sure what your budget is like but it's a good investment if your working on headphones a lot.

also toneboosters morphit can do a good job, and cheaper at 30 euros, tonebooster make pretty good plugs
 
http://www.toneboosters.com/tb-morphit/
 
and arguably one of the best sonarworks reference 3 is excellent, a bit more pricey at 99 euro, but well worth it
 
http://www.sonarworks.com/headphones
 
i have the waves nx, toneboosters morphit and sonarworks ref 3, i'd give the edge to the sonarworks, but they all do the job well, and at the price of toneboosters morphit, it's certainly worth a try. you can demo all of them for yourself
 
there is also a little thing which i think should be in sonar, made by craig anderton which can be used for this type of thing, i have tried it, and it's ok, not up to the others mentioned, but maybe worth a try, can't recall what it's called though
 
[edit] you can also use toneboosters isone in conjunction with morphit, or either or on its own, a lot use them in conjunction with eachother, some use one or the other, it's relatively cheap at 19.95 euro
 
http://www.toneboosters.com/tb-isone/

There have already been posted some very good advices above.
Basicely every way is possible, even with a hifi system and headphones,
but it will probably take longer time
Good studio monitor speakers can accelerate the quality of your mixing skills,
otherwise there are young "producer" able to make very good mixes with
cheap desktop speakers.It takes longer and it´s more effort
to manage that.
Mastering is imo another theme.
 

A little advice, not because it's great but because you're going to run into it by home mastering - if your mastering preset greatly improves tonal balance (more highs or lows or whatever), try to fix it in your mix instead of leaving it to the mastering preset. This is because the mastering processor can only deal with the stereo signal and if you fix it in your mix you have a choice between, for instance, adding more bass guitar or boosting the low end of the guitars.