High temperature issue at double helical gearbox after revamping.

Dear All,

I have a case of high temperature issue at double helical gearbox which journal bearing temp of HSS increased to 103oC after 3 months and then fluctuated from 83 to 102oC.

What  done are: major maintenance for GB and revamping for Ammonia unit. It means the load increased 107-110% than normal.

I summarized the information and my solution in slide.

Hope to receive your discussion and any advice.

Thanks & Regards,

N.V.Đ

Attachments

Original Post

I see a comment reading "What done in turnaround 2017? 1. Replace new set of opinion and gear. 2. Replace new journal and thrust bearing."

Why was the gearset replaced?  When you replace bearings, are they varnished?  Is the goal in bearing replacement a clearance change or are the bearings damaged?

Hi John,

The gear and pinion is replaced due to 15 years of using and it appear little bit worn on teeth surface.

Bearing temp before turnaround is around 80oC and after two years of operating, when inspecting, it also record vanish on loading bearing surface (dark yellow color). The change of new bearing because the journal of pinion and gear is not different between old and new set. The journal of HSS of new set is smaller than 0.04mm.

I would like to hear your advice on:

1. whether Increasing the journal bearing clearance more 0.03-0.04mm is effect much to gearbox operation or not?

2. The guy from OEM said that the contact of gear is not correct could make temp increasing, what do you think about this point? 

3. The temperature is fluctuated unstable from 83 to 102oC. Sometimes, it suddenly decrease from 100 to 89oC without any changing load. What do you think about this situation and do you know why (I have not found the convinced cause for this).

Thanks John.

Those bearings do not have a high sliding velocity.  The pinion is about 46 m/sec and the gear about 48 m/sec.  Not knowing the bearing details, by my rather rough calculation, I would want around 0.24mm (pinion) and 0.25 mm (gear) clearance in the bearings.  But having that doesn't address the erratic nature of your temperature and I think you'll have to look further to find the cause.  As you likely are aware, that can be due to several reasons.  Look carefully at your data.  This is a double helical gearbox so both bearings on any given rotor should have similar temperature behavior.  On any given rotor, is the erratic nature to the data apparent on both bearings or only on one?  Do you see the erratic temperature behavior on all bearings, both shafts, perhaps indicating a problem with the supply temperature?

Thanks John and William,

Regarding to the oil, I think it is not a cause. The reason is this oil system used for three compressor trains which 2 the rest compressor do not showing the similar phenomenon (just for information that at the beginning start up the plant, all compressor train have bearing temp increase around 8 to 10oC compare to the previous time. one of the compressors was trip by high bearing temp fault, we have to increase the bearing clearance out spec and it is solved with temp below 80oC - at the time, the bearing temp of gearbox is not high due to load not increased yet). at the moment the problem is just serious to gearbox (HP and LP compressor, turbine bearing temp also increase 7-10oC comparison to the value before revamp but it stable with temp below 91oC).

John, could you advise me how you can calculate from speed of journal to bearing clearance, is it formula for this (sorry, because I do not know).

 

Ngo Dinh posted:

 

John, could you advise me how you can calculate from speed of journal to bearing clearance, is it formula for this (sorry, because I do not know).

 

Sorry, I used proprietary methods from a former employer that I still have access to.  However, there are plenty of references on the web that can give some guidelines.  https://www.kingsbury.com/pdf/catalog-FPJ.pdf, page 4 Figure 1 gives some guidelines from a bearing manufacturer.

If you seek out more info, just keep in mind that gearbox bearings have an L/D of approximately 1.

 

Looking at the fluctuations in the bearing temp, is there a fault with the instrument you are using to measure it? I have seen this on similar Parallel Shaft G/Boxes which was due to damaged wires on the RTD. What is being used to measure bearing temp, and where on the bearing is it located? Is there a 'spare' instrument next to it to check that the temperatures are real?

If you are measuring a temperature of 111 deg C, this would indicate that the white metal of the bearing has reached it's plastic limit. This also questions how far away from the white metal the temperature is being recorded.  When inspecting the bearing, has  the white metal been 're-shaped' within the bearing?

Looking at the contact markings, it is difficult to say with out knowing the OEM's specification for this and how they expect the markings to increase with load and torque, (I'm assuming about 95% contact at 100% load) but they do look to have a high proportion of the tooth in contact with no load. Where these checks done in the gear case, and with the contract bearings?

A final thought is that when the decision was made to increase the load to 110%, did the OEM re-design the bearings to take this increased load?

Hi John,

We open gearbox and do two things: 

1. Increase bearing clearance to 0.20 at HSS and 0.23 at LSS.

2. Adjust gear contact as OEM

the temp now is low with max at HSS 75oC and LSS 82oC.

do you think that the gear contact not good will make the high bearing temp problem? Have you meet a similar case before?

Ngo Dinh posted:

Hi John,

We open gearbox and do two things: 

1. Increase bearing clearance to 0.20 at HSS and 0.23 at LSS.

2. Adjust gear contact as OEM

the temp now is low with max at HSS 75oC and LSS 82oC.

do you think that the gear contact not good will make the high bearing temp problem? Have you meet a similar case before?

I have not seen a case that can be directly attributed to tooth contact.  But as you likely know, sometimes on a disassembly and reassembly things happen that can also provide explanation.  I have seen for instance a case where RTV compound was used on the split line and it was theorized that where the two halves of the casing were bolted up, the RTV caused an increase bearing clearance.  That resulted in increased bearing end leakage and higher than normal temperatures.  Because rotors were not now running with parallel centerlines, the contact suffered.

The scenario is a "stretch" but in theory it could happen.  But after a rebuild when you are scratching your head sometimes all you have is potential theories that are difficult to prove.

But you should not in general use RTV by the way for any components that get bolted together and control a clearance.

Thanks John,

actually, in my experience,  i have not seen a case that high temp due to bad contact. And i think that it seem to effect more to vibration but a little to temp and if any it could be a minor effect that contribute to make temp fluctuate but not a root cause of high temp.

the other thing, it also make me supprise that it is just in last turnaround 3 months ago. The contact seem to be good, of course, at the time  the field maintenance take at wrong rotation ( it shoud be active side). But my question that if the inactive side is contact good, is it active side good also? For double helical, is the contact more area of contact more better? Or...?

A double helical gearset can have poor contact if there is an axial spacing error in the total machine.   As you likely are aware the gear has a thrust bearing and the pinion is almost always free to float.  But any thrust applied to the pinion will cause the desirable 50%/50% split of power between the two helices to be upset.  You might for instance have a 70%/30% split.  You can even have a sufficient thrust to make one helix carry all the load with backside contact on what you call inactive.

Go to https://pdfs.semanticscholar.o...b3d287cea0eb630c.pdf for a good paper on the subject.

 

Thanks John,

I have read this magazine already and I am sure that the axial position of both gear is correct. About inactive side I mean above, it is just when do blue contact check. I mean that the field operator applied the blue contact in back side or inactive side with reverse rotation to get gear contact pattern. Is it ok to take the contact? and if the pattern is ok, is it mean that active side ok?

The other things I think it could effect to axial position of pinion is wrong pre-stretch setting up. in our case, I also find the wrong prestretch.

 

 

Ngo Dinh posted:

But my question that if the inactive side is contact good, is it active side good also? For double helical, is the contact more area of contact more better? Or...?

In high speed double helical drives you can get differing tooth contact patterns on the active vs. inactive side.  The active side is expected to see the transmitted torque and bending moment and thus there may be a tooth modification.  That same tooth modification is not needed on the inactive side because torque and bending are minimal.

It is also important that you be aware of any tooth modifications that may be present as they may affect the static and dynamic tooth patterns differently.  The OEM is aware of any tooth modifications since they have access to the manufacturing drawings.  End user maintenance departments usually don't have the drawings and all to often apply "handbook" criteria that is intended for slow speed machinery.   

 

Normally, I usually apply that the gear contact is good when the contact area pattern is over than 80%; however, for double helical, sometimes it is not the same. Some manufacturers stipulate the contact pattern. I think it could be for ensuring share the load at both helical when gear on load and unload.

 

Attachments

Photos (1)

Add Reply

×
×
×
×