The first couple of years of the monoshock DT175 CDI definitely had a high speed and low speed coil. From the 1978 DT175 Service Manual in the Tech Library:



For the 1980 DT175 in the US, they went to a 6 wire CDI box (from 7 wire) with corresponding stator that now had a single source coil. From what I've seen over the years, the UK got the same "upgrade". I've found the later version to be far less trouble and... knock wood... have never had a bad one yet. Earlier version is less reliable for some reason.

You can swap to the six wire version... both Flywheel/stator assembly and CDI box... you just end up with an extra unused wire in the loom between the stator and the CDI box.

I would guess the dual coil version is less reliable, as the CDI unit itself will have additional circuitry?

It would be interesting to see if a Ducati type coil/CDI unit works on Yamaha engines. They are cheap and easily available.

Thanks for the clarification Mark. 
So we have a confirmed possible third culprit.  When I get back on my feet ( hurty ankle) I will try RTs brilliant idea of heating up the high speed coil with a heat gun.  I never thought of that.

I already changed the plug and I have a new ignition coil on the way.  Not an expensive OEM one, just a cheapie to try out.

​​​​​ At the time I thought it was throttle opening related and a weak HT system failing to overcome the higher resistance across the spark gap at more open throttle settings and hence higher compression and higher resistance.
When I did the ignition, I did the spark gap test and it was jumping 6mm with ease.

However, thanks to certain comments on this thread, I remember that around half throttle also coincided with around 4000 rpm on my test run and no amount of coil resistance testing and spark jump testing at room temperature and at kickstart speeds can possibly replicate what is happening under the mag rotor at operating temp and high rpm.

If my paranoia over this bike gets any worse,    I will also be blaming the rotor magnets, but it charges ok and runs the lights, indicators ect just fine.
 

Yamfan wrote: " I would guess the dual coil version is less reliable, as the CDI unit itself will have additional circuitry?"

Logical guess...  except that the issues for me always have something to do with the stator.  The maddening part is the source coils will often read close to or even within factory spec but either the high speed or low speed won't work.  (Of the 3 I've had, all have been the high speed coil) Swapping to a "known good" CDI box from a perfectly running bike has never fixed the issue for me.  And the CDI box from the problem bike works fine on the good running bike.  Swapping out the high speed coil has worked the one time I tried it...  found a NOS coil on Ebay. 

Now all my 175's have the later 6 wire CDI, a VAPE, points, or a later CDI system from a more modern Yamaha.  I love the points systems.  Reliable as a rock, just hard to find.

 

It seems as though the dual source coil system, is less reliable than the single coil ones?

I have quite a few TY175 points stator plates, and they are very easy to find in the UK. If any use to you, can see exactly what I have?

The issue with the dual coil systems failing, might have to do with the high speed coil, running far hotter than the low speed one, and damaging the insulation?

The dual coil systems were used on some TZs, and apparently they were never very reliable.

Electrex manufacture and supply the high speed coil for £42.  

Also fits 1980s Kawasaki KX400s and 500s as well as the early 80s Yam YZ465 and 490.

Unfortunately the house accountant has put her foot down and just announced a do not use the credit card again this month policy. 


 

Well, hobbled into the conservatory to have another look. First rechecked all the coil readings. All still in spec.

I removed the rotor nut and before I pulled it to warm up the coils with a heat gun as RT suggested, thought I would recheck the timing.. Bang on at 1.8mm btdc.

On the mark.


At 1.8mm BTDC after pulling past TDC, going backwards, zeroing gauge at TDC, going back a further 90 degrees then going forwards again till the gauge hit 1.8mm.


Coils heated to around 70C, hot enough to barely touch them and readings taken again.

All the readings increased, it is a close call as usually Yamaha specify coil readings at 20C and all were fine at that.
I would tentatively say the coils are fine, however more knowledgeable comments from the forum experts gratefully accepted.

HT cap was checked. Fine at 5.1 ohms.
HT coil was checked,- primary fine at 1.1 ohms, secondary fine at 5.9 ohms.

Compared with new coil. Agreed with the readings of the original within a couple of points. I fitted the new coil as the old one was showing signs of rust around where the iron core enters the case. The laminations had not swollen, but it was a matter of time.

Low speed source coil - Brown wire connected across the meter to black stator ground wire. Spec is 420 ohms (plus or minus 10%)


Trigger coil - Red wire with white tracer connected to black stator ground wire. spec is 12.4 ohms, again ten percent either way.



High Speed source coil - red wire connected to stator ground wire. Spec is 13.6 ohms plus or minus the 10 percent.
Out of spec hot, but fine when cold.

Try putting a strobe light on it, when it's running badly, and see what the spark is like.

Shall we look at the possibility that its not a spark issue. Then again i wonder--just to prove a point-- if you give the plug a 'big gap' like 40 thou then see if it accentuates the erratic running when hot. Some carplug i think are like --11 on the end which i think means 44 thou ==1.1mm. Mind you they would be running a battery powered TCI or something. It's easy to head off in the wrong dirrection & keep going further then end up like me--Lost lol.
Just a note, when i heat tested suspect coils i had the meter on while heating. Think it was on my YZ465 at the time & it would just go open circuit, then come back when cooled.

Also, possibly what YAMFAN was getting at, marks can be spot on at 1.8 & nothing will ever change for the marks unless the plate is moved in the slots. But what it's doing while running might not match up to the marks. Just be an interesting experiment.
Actualy my old genuine yamaha test meter would be good for the wide gap test as has an adjustable gap in a window. Never done that running but tested lots of plug coils off the bike.