Author Topic: By Jupiter  (Read 75279 times)

Online Vixen

  • Full Member
  • *****
  • Posts: 3075
  • Hampshire UK
Re: By Jupiter
« Reply #450 on: August 08, 2019, 10:00:07 AM »
Hello Per

You could be correct, time will tell.  :noidea: :noidea:

I have based my 5/16" plugs on a proven design by Graham Meek. So, I will not be making any predictions or design changes until I have completed my testing. Only then, will I have hard facts to work from.

Mike
It is the journey that matters, not the destination

Sometimes, it can be a long and winding road

Offline Roger B

  • Global Moderator
  • Full Member
  • *****
  • Posts: 6133
  • Switzerland
Re: By Jupiter
« Reply #451 on: August 08, 2019, 11:37:08 AM »
A key point with any spark plug is that the actual spark gap should be much (5 times at least) less than any other path for the voltage. The design you have looks feasible, the ends of the ceramic pieces are the biggest risk areas. If the PTFE seal is a close fit in the ceramic bead so there is no air path that should be ok. The tracking length on the combustion side of the ceramic may be a little short. What is the actual gap? Could it be made less?

If it is feasible to machine a sleeve to go from 5/16 to M10x1 (CM6 plug) I can test them in my horizontal engine. The sleeve would need to be screw cut or thread milled, normal taps and dies would destroy it.
Best regards

Roger

Online Vixen

  • Full Member
  • *****
  • Posts: 3075
  • Hampshire UK
Re: By Jupiter
« Reply #452 on: August 08, 2019, 12:13:56 PM »
Roger,

Both spark gaps are set during manufacture at 0.015". They could be increased or reduced by modifying or remaking the central conductor. The alternative path between the PTFE and the Corian is a respectable 0.087" which just exceeds your 5 times criteria.  The PTFE seal is tightly compressed ( no air gap) by the locking nut. It is compressed against both the ceramic and the corian faces, as well as sealing the body and central conductor.

Graham Meek successfully uses 4.0 mm ceramic fish spine beads on his 1/4" plugs, I have the luxury of being able to use 5.0 mm fish spines. The Fish spine ceramic beads are unglazed which may not help in wet running conditions.

Thanks for the generous offer to use your horizontal engine for the testing. However, I may have found a suitable 'mule' engine, in the form of a 35 cc Honda GX 35 four stroke industrial engine. It will soon be time for some experimenting. I always have the options of using one piece insulators in Corian or PTFE if the ceramic beads prove to be a problem.

Regards

Mike
It is the journey that matters, not the destination

Sometimes, it can be a long and winding road

Offline Admiral_dk

  • Full Member
  • *****
  • Posts: 3752
  • Søften - Denmark
Re: By Jupiter
« Reply #453 on: August 08, 2019, 09:58:13 PM »
Just a simple question Roger - do the 1:5 ratio also apply when there are substantial pressure difference between the two gaps ?
I ask because the required voltage to arc a certain distance increases with pressure.

The small Honda sounds like a very good test mule Mike - I really hope you can get a nice verification that your plugs are OK.

I forgot to add to the lawn mover story, the the problem where the plug even if it was a full size and not a modern small plug.

Offline Roger B

  • Global Moderator
  • Full Member
  • *****
  • Posts: 6133
  • Switzerland
Re: By Jupiter
« Reply #454 on: August 09, 2019, 09:55:00 AM »
Per, the 1-5 ratio is a very simple rule of thumb. If you are less than that you will generally have problems, in some circumstances you will need a lot more especially when comparing surface tracking to spark gaps.
Best regards

Roger

Online Vixen

  • Full Member
  • *****
  • Posts: 3075
  • Hampshire UK
Re: By Jupiter
« Reply #455 on: August 11, 2019, 07:07:47 PM »
It's been a busy weekend. Yesterday afternoon, I pulled the 35cc Honda GX 35 four-stroke engine out of one of my large, all metal, tank projects. The Honda GX is a known, good clean running engine but it had not been run for three years. I cleaned it up, mounted it on the running bench and went to the garage with my red can, to buy two litres of non leaded petrol (2 litres is the minimum they will dispense). I filled the tank and pulled the cord starter, the Honda burst into life on the second pull.

I took that to be a good sign.




I pulled the spark plug and found it was a 10mm x1 thread, I had expected the normal 14mm x 1.25 thread.for which I have already made an adapter. So, another adapter had to be made before I could start. As Roger pointed out earlier, there is not much wall thickness left when you cut a 5/16" thread inside a M10 thread. It looks just like a glued together Helicoil. I drilled and tapped the 5/16 " thread into the full diameter stock (EN8 steel) using a tap held in the tailstock. Then I reduced the stock to 10mm and machine cut the tread on the CNC lathe, in 0.002" increments. I half expected it to all fall apart but fortunately it turned out good.

Here is a family shot of the original Honda plug, the adapter and the model 5/16" plug. I had to use a 10mm red fibre washed to seal the adapter in the cylinder head, it was all I could find that fitted.

.


Next the acid test. Would my new design of spark plug actually work? There have been a number of suggestions that it may not and a number of explanations of why and how they would fail. The only way to know for certain is to do series of practical tests, examine the data and work from some the hard facts.

All thirty spark plugs were tested. The little Honda engine started and ran successfully on all thirty plugs. Four of the thirty plugs had a tendency to misfire and not run as clean as the rest. These have been quarantined awaiting a post mortem dismantling and examination. I believe a success rate of nearly 90% vindicates the design and it's one more item (actually a full set of 18 +spares) can be crossed off the ' still to do' list.




Sorry about the blurred photo, must have been a combination of excitement and a single cylinder engine buzzing away.

Happy happy happy

Mike  :cheers: :cheers:
« Last Edit: September 11, 2019, 03:17:53 PM by Vixen »
It is the journey that matters, not the destination

Sometimes, it can be a long and winding road

Online sco

  • Full Member
  • ****
  • Posts: 1693
  • Location: Northants UK
Re: By Jupiter
« Reply #456 on: August 11, 2019, 07:18:47 PM »
Good result Mike and I expect the Honda being a modern engine it will run lean with a high compression ratio so the plugs should get an easier time in the model engine.

Simon.
Ars longa, vita brevis.

Offline Roger B

  • Global Moderator
  • Full Member
  • *****
  • Posts: 6133
  • Switzerland
Re: By Jupiter
« Reply #457 on: August 11, 2019, 07:56:04 PM »
Splendid  :praise2:  :praise2: I wasn't sure if a M10 x 1 adaptor would be possible  ::)  :wine1:
Best regards

Roger

Online steamer

  • Global Moderator
  • Full Member
  • *****
  • Posts: 12697
  • Central Massachusetts, USA
Re: By Jupiter
« Reply #458 on: August 11, 2019, 08:28:06 PM »
Congratulations Mike!!!

Dave
"Mister M'Andrew, don't you think steam spoils romance at sea?"
Damned ijjit!

Offline Admiral_dk

  • Full Member
  • *****
  • Posts: 3752
  • Søften - Denmark
Re: By Jupiter
« Reply #459 on: August 11, 2019, 09:50:19 PM »
I'm so glad to hear that my worries after your first partly test, has been proven unjustified  :whoohoo:

I'm sorry if I sounded like a spoilsport or nay sayer - it just come down to, too many bad experiences with bad plugs that "works outside the engine, but not inside" over the last 15 years in my case.

One step closer to a running Jupiter  :LittleAngel:

Offline Art K

  • Full Member
  • ****
  • Posts: 1761
  • Madison, Wisconsin USA
Re: By Jupiter
« Reply #460 on: August 12, 2019, 03:14:33 AM »

Mike,
Great to see such a high good to bad ratio. Great job on the plugs! One more thing checked off on the to do list.
Art
"The beautiful thing about learning is that no one can take it away from you" B.B. King

Offline mike mott

  • Full Member
  • ****
  • Posts: 418
  • Alberta Canada
Re: By Jupiter
« Reply #461 on: August 12, 2019, 02:14:26 PM »
Mike, this is very encouraging news for those of us who want to try a new plug design (yours).

Mike
If you can imagine it you can build it

Online Vixen

  • Full Member
  • *****
  • Posts: 3075
  • Hampshire UK
Re: By Jupiter
« Reply #462 on: August 12, 2019, 02:36:26 PM »
The credit should go to Graham Meek. He pioneered the use of fish spine ceramic insulators.

My plugs 5/16" are based on his design for 1/4" spark plugs.

Regards

Mike
It is the journey that matters, not the destination

Sometimes, it can be a long and winding road

Online Vixen

  • Full Member
  • *****
  • Posts: 3075
  • Hampshire UK
Re: By Jupiter
« Reply #463 on: October 08, 2019, 06:53:02 PM »
After a short break updating my Bristol Mercury engine's display, it's back to working on the Bristol Jupiter.

I dismantled and  examined the five 5/16" spark plugs which had been misfiring during the previous test session. I found I needed to heat the stainless steel balls with a flame to soften the loctite joint, After that, it was an easy job to pull the plugs appart. In each case, I found the rim of the Corian electrical insulator had cracked around it's circumference by the force of the locking nut. A clear case of overtightening during assembly. I had not known the correct tightening torque to compress the PTFE cone to achieve a gas seal. In my enthusiasm I had used a 4 " socket handle and given it a good twist and then a bit more.

It was then a quick job to turn up a dozen new Corian insulators and a dozen new PTFE cone inserts, which provided a number of additional spares, should they be needed later. Foe each of the plugs I used a new fish spine thermal insulator, PTFE cone and Corian Insulator and discarded the previous items, This time I was more cautious when tightening the locking nut. I pressure tested each repaired plug to 8 Bar and discovered just how little force was actually required on the locking nut to achieve the gas seal. It was nice to find all the test adapters ready made and sitting in the plug box.

The final test was to run each of the repaired spark plugs in the little Honda test mule. All five spark plugs worked perfectly. I now have all 30 out of the batch of 30 spark plugs, all working , fully tested and available. So that's something else crossed off the list.

Mike
It is the journey that matters, not the destination

Sometimes, it can be a long and winding road

Offline Admiral_dk

  • Full Member
  • *****
  • Posts: 3752
  • Søften - Denmark
Re: By Jupiter
« Reply #464 on: October 08, 2019, 08:15:01 PM »
Great that you discovered the reason for the failure of those five plugs  :ThumbsUp:
As an added bonus, it will also prove valuable knowledge in the future  :)

It's also an example of why I love having a torque table for the job I have to assemble -> a lot less chance of bodging the job. Not quite so easy if you designed / build it yourself ....

 

SimplePortal 2.3.5 © 2008-2012, SimplePortal