2

I recently replaced my thermostat flange which is made of plastic. Unfortunately, I didn't find any torque specs for its bolts in the manual. Since, plastic parts break easily, I'm afraid of over-tightening the bolts. How can I make a safe guess about the acceptable amount of torque for such a plastic part? How tight do you usually tighten a plastic flange?

Peugeot 406 1.8 16V XU7JP4 LFY MY 2000

enter image description here

6
  • 1
    How is the housing sealed? Is it via o-ring or gasket? (I'm talking about the part which you're bolting down.) Commented Jun 10, 2021 at 0:13
  • 1
    Make, model and engine? Then we may read the specs in the manual if it can be found.
    – Solar Mike
    Commented Jun 10, 2021 at 5:00
  • @Paulster2 - It is sealed via thermostat o-ring.
    – LFY MP7.3
    Commented Jun 10, 2021 at 6:28
  • @Solar Mike - Peugeot 406 1.8 16V XU7JP4 LFY 2000
    – LFY MP7.3
    Commented Jun 10, 2021 at 6:29
  • Please add that detail to the original question, better than leaving it in comments.
    – Solar Mike
    Commented Jun 10, 2021 at 10:40

2 Answers 2

3

You have three possibilities:

  1. You somehow get the official workshop manual (I have no idea how) and use the official torque (If given)
  2. You ask around until you get the "internet quote" of your least mistrust.
  3. You do it like most of the people out there, and develop "a feeling" for the needed torque. Yes, with some experience it is possible to do it by feeling, with an acceptable success rate. A possible approach would be to torque down the old piece onto another sacrificial piece, on your workbench, until destruction. So you might get a feeling for the needed torque. Also, have a backup plan in case something breaks.

If you are worried about the bolt coming loose, you can apply some low-strength threadlocker.

5
  • Thanks a million for your comprehensive answer dear Martin. I also have an idea. My theory is to start with tightening the bolts to 5 nm (3.7 lb-ft) and start the engine. If it leaked, I will try 6 nm. If again leaked, I add another 1 nm to the tightening torque. I continue doing this until the leakage stops. For example, if leakage stopped at 9 nm, I will tighten the bolts to 10 - 11 nm. What do you think?
    – LFY MP7.3
    Commented Jun 10, 2021 at 11:47
  • 2
    I doubt that this works. If the torque is too low, the seal might be tight, for some time, perhaps due to the cold engine. Later the bolt gets loose and then you have a massive coolant leak, directly on the thermostat, ideally while going full throttle on the highway. Do it right the first time
    – Martin
    Commented Jun 10, 2021 at 11:56
  • Agree with Martin's answer here (#3). I would bet this is what the engineers assumed when creating this part. Most of the time when you can't find a torque number, there really isn't one. You have to get a feel for it. An o-ring will seal (in most cases) regardless of the torque on the bolts, as long as the bolts are "tight enough" to hold the o-ring in place it will seal. Commented Jun 10, 2021 at 14:10
  • 1
    @Martin - Today, I tightened the bolts with my hand and stopped when I felt they are tight enough. I trusted my feeling! Then, I filled and bled the system and took it for a drive in different conditions (mid to full throttle). I returned home and checked everything. There was no sign of leakage. Success!
    – LFY MP7.3
    Commented Jun 11, 2021 at 18:53
  • Good to hear. Congratulations
    – Martin
    Commented Jun 11, 2021 at 19:50
-1

Should come with an o-ring, tighten to 10 ft-lb or 14nm.

2
  • 14 nm seems a lot to start with. I think it should be between 5 - 8 nm but I'm not sure.
    – LFY MP7.3
    Commented Jun 10, 2021 at 7:16
  • 14nm is not a lot, clearly you're not sure though so why would you think it is? If you're too afraid to use 10lb-ft or 14nm, then do your 8nm but use low strength thread locker as Martin suggested.
    – NetServOps
    Commented Jun 11, 2021 at 0:17

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .