Vibrations on BSP Parallel G Thread plumbing fittings 1

[Jack Benson]

In our system we have a small diaphragm pump that vibrates when in use.

We use swivel fittings in our system, and the manufacturing process i have taken over involves adding a drop of glue to every join as the previous system designer thought it would stop vibrations loosening the joins

This video is from the production line.


I want to remove the glue as it makes servicing the device much harder as disassembly is much harder as the glue is strong.

Also, if there are mistakes in the production process and joins are not connected properly cannot be easily fixed with a simple tightening.


How can i find best practice for systems that have to maintain water tightness where there are vibrations?

thank-you

 

[TugboatEng]

Pipe needs brackets and hangers to keep it from loosening. Thread lockers in this case provide no benefit. This is especially true with swivel joints because the adhesive doesn't prevent the swivel joint from moving.

 

[Jack Benson]

everything is bolted down and supported.

there are short flexible hoses up to G1/2" 50 cm long that connect various parts of the system. They are help in place with cable ties

are you saying the glue is pointless?

We are using a very small amount of LOCTITE® 577 General Purpose Thread Sealant.

I am happy to stop adding it.

 

[btrueblood]

Have you done some long duration testing with the pump running and seen if leaks are a problem?

Frankly, for a production device I would be looking for ways to minimize/eliminate threaded joints, and especially gasketed swivel joints like that - are there ways to use manifolds, brazed or welded joints, etc. If you can't eliminate the swivels, then you need to look at SAE Oring boss fittings and/or Oring face seal fittings. Both have the advantage of being metal-to-metal contact so vibration loosening is less of an issue (no soft gasket interface, and you can put full preload on the threaded connection without fear of damage the seal element), and you can use regular loctite (like the removable blue stuff) to prevent threads backing off.

 

[LittleInch]

Why don't you just use snap fit hose connectors like this

https://uk.rs-online.com/web/p/hose...8DvtHTz-SIoeFNoZepBoCJ-oQAvD_BwE&gclsrc=aw.ds

Sure they cost a bit more but think of the time saved from screwing all those connector together and leak free vibration resistant fittings?

Hydraulic hoses and gas fittings are often these sorts of connectors.

Remember - More details = better answers
Also: If you get a response it's polite to respond to it.

 

[Prometheus21]

For simple water tightness I would follow Littleinch's recommendation - snap fit connectors are by far the easiest solution.

We have similar problems in the industrial gas industry. If the product require frequent maintenance then we usually use Loctite 243 " aka the blue stuff" as a thread lock, for "permanent" locking we use Loctite 270.

Now this is only if the product allows the use of Loctite, and it is always used sparingly; but it does the trick and allows for frequent maintenance (every 2 weeks in our case).

Now as btrueblood states, I would look at alternate sealing methods. An O-ring seal would allow for a quick retightening by hand when needed, and is heavily employed in the industrial gas sector. If the assembly/disassembly of the product is infrequent then the use of a metal to metal seal is recommended (bull nose and cone seal - described in EN560 for BSPP threads) - For brass I would recommend 15-30 Nm of torque for a seal to be made - now this is only a recommendation and its effectiveness is obviously affected by the amount of vibration you experience. The industrial gas sector uses this kind of metal to metal seal almost everywhere, and can be found in almost any European gas installation.

Alternatively go for the cheapest solution: use a gasket and 60 Nm of torque. For most brass alloys this is acceptable for threads with TPI of 14 (often used for cylinder connections at 230 bar using G 1/2 up to G7/8 threads).

 

[zeusfaber]

We are using a very small amount of LOCTITE® 577 General Purpose Thread Sealant.

I'm always a bit suspicious when I see sealant being used on threads that don't form part of the pressure boundary. Would one of the thread-lockers mentioned by other posters be easier to work with?

You seem to have the same fitting on the outlet end of the elbow. Is the whole system assembled with swivel fittings and fibre gaskets?

A.

 

[Jack Benson]

Thread lockers would be much easier to work with. I didn’t know they existed. I have started researching them this afternoon.

The whole system is assembled with swivel fittings and fibre gaskets.

We do not have enough data to know if the risk is real of the join opening due to vibrations. At the moment we are erring on the side of caution.

I don’t know if there is a scale to measure vibrations to help assess the risk.

This is the pump we are using SeaFlo 42 series 3GPM. This is the only source of vibrations.

No ridgid pipe work connects to the pump

 

[Jack Benson]

I forgot to reply that we do not need to disassemble the product.

I will look into other sealing options.

In 6/12 months we hope to have an injection moulded version of the system to remove all the pipework, but until then we need to be able to produce efficiently and be able to iterate the design as needed

 

[hydtools]

Use the fittings used in recreational vehicles. Straight female swivels with internal rubber cone washers. They are used in your attached pump video.

Ted

 

[TugboatEng]

That is what they are using.

 

[electricpete]

> I forgot to reply that we do not need to disassemble the product.

that seems in conflict with original problem statement

> I want to remove the glue as it makes servicing the device much harder as disassembly is much harder as the glue is strong.

But fwiw it makes sense to me the threads are used to compress the washer, so you don't need thread sealant, but thread locker may help to prevent vibration loosening

 

[Jack Benson]

We only need to adjust a joint if it was not sufficiently tightened in the production process.

The glue we are using at the moment it hold the joints in place is very strong and makes corrections very difficult.

I have ordered the Loctite Blue & Red thread locker to test. If they provide a balance between preventing vibrations opening the swivel joints, but also not requiring too much force to open, you have solved my problem

 

[TugboatEng]

Do you even have a problem? Everything you are using is commonly used without thread in many applications and doesn't have problems with vibrating loose.

IF your fittings appear to be loosening it is likely due to the gasket taking compression set. That is not going to be solved with a thread locker.

What is the gasket material? What is your operating temperature? What is your process fluid?

If you use thread locker while your problem is caused by compression set you're setting yourself up doubly for failure.

 

[3DDave]

Tugboat - Look at the other threads.

thread404-512745
thread404-512396
thread404-510845

 

[TugboatEng]

Ahh, I didn't realize I had responded to OP before. I guess I'll reiterate, the water is too dang'd hot for buna/nitrile rubbers and that is causing the "loosening" without the nut rotating.

 

[Jack Benson]

TugboatEng

- the max water temperature is 45C.

- I do not know if there is a real issue with vibrations or not. We have always used glue on the joints so we do not have systems in the wild that have not been glued. I inherited the process when i took over my current role and it causes unnecessary problems.

- We either is a fibre gasket, unless we are using a flexible EPDM hose that are supplied with rubber gaskets



Prometheus21

- We are using G1/2" female swivel connectors (14 TPI). I have checked with the production manager and the torque that is being used is 15 Nm which is a lot lower that your post suggests we should be using. We are using a Facom R.306-25D torque wrench. Before we used the torque wrench, we had problems with under tightening and over tightening. Over tightening damaged the washer. I have not been able to find a torque charge online for G fittings. Would you know where i could find one?

thank-you

 

[TugboatEng]

The fiber gasket has a rubber binder. This rubber binder needs to be compatible with your fluid. Generally, SBR would be used for hot water applications.

 

[Prometheus21]

I don't believe an official chart is available, but I might be wrong. The torque I recommended is purely based on information gathered by myself and other engineers in the company I work for; as well as information gathered by senior engineers I have had the pleasure of working with in the industrial gas industry. Note that we deal in shock loads, rarely in continuous vibration configurations; (Although for cylinder filling with flexible hoses you have vibration propagating throughout the entire system - this has never been an issue however).

That being said:

1. You say you don't know if vibration is an issue. I can't imagine testing it and finding out would require much effort and resources. There is no need to fix an issue that might not even exist.

2. The gasket will get damaged long before the threading system is overstressed. I have no experience using fibre gaskets, as they are not permitted in the industrial gas industry. For repeated connect/disconnect applications we use gaskets made of PA6 or polyurethane. For "permanent" connections we use copper or aluminium gaskets. If you ever need to disconnect this kind of system you would also need to replace the gasket (not a huge cost all things considered).

The image below shows a typical connection on a cylinder valve outlet; e.g. a pressure regulator connected to the valve of a gas cylinder.

For a similar configuration with G 1/2 threads and gasket as sealing method we use 60 Nm of torque to make a secure seal at 230 bar cylinder pressure. Note the nut is made of stress relived, surface treated CW614N/CZ121 (CuZn39Pb3) and the gasket material is PA 6. The pressure regulator experiences quite the shock load when the cylinder valve is first opened. For a repeated test of 1000 pressure shocks no leakage was detected and no retightening was required. Note that everything stated above also holds true if the seal is an O-ring or bullnose and cone seal (metal to metal). (For in-house applications we have abused our equipment with similar configurations for years - I estimate around 30000 cycles in the product lifetime - no leakage at the gasket as of yet)

Considering your systems low working pressure I would recommend trying 25 Nm and call it a day if the gasket shows no sign of damage (test it - repeatedly). If needed you can use Loctite 243 or another thread locker.

Then again don't trust me, I'm just a stranger on the internet. Testing it out is half the job and all of the fun

 

[saplanti]

Diaphragm pumps will always produce pulses and vibration. Therefore you need to use pulsation absorber/accumulator to eliminate the pulsation effect. Perhaps the pump supplier can recommend/provide an accumulator.

Pulsation flow may/will cause problem in the flow path in case you do not eliminate/reduce pulsation.

Additionally the pump mechanical frequency is another that the piping system should be avoiding. If your system is within the 25% of pump frequency and its multiples you may expect mechanical resonance in the system. That is why the vibration causes some fitting get loosen.

So the system is small or big the pulsation systems gets this kind of problems unfortunately. You may need to do some engineering or get consultation from the pump supplier.