Coffee Forums banner
24K views 190 replies 20 participants last post by  FairRecycler 
#1 · (Edited by Moderator)
Hi Everyone,

I thought I start this topic, to gather as much information as we can, in the hope for an ultimate recipe, to eliminate the vibration of the needle.

I'm far from expert in this topic, so every input would be greatly appreciated.

I'm aiming to start the experiment next week, where I'm willing to test the the following variables on a freshly rebuild machine with a stainless steel 41.3mm (1 5/16") glycerine filled gauge:

- Water filled / air filled capillary tube

- Length of capillary tube (short straight / short single coil / longer 5-10 coils)
I would appreciate some input on theoretically ideal length, if such exists.

- Height of gauge in reference of sampling point.

- Sampling point: boiler in-feed / outfeed
Sorry I'm not planning to test the pump outfeed (T piece on PTFE tube), as I can't see it any chance for it being any better by going closer to the vibratory pump.

- Routing
I would also appreciate some input on this.

I'm open to test other reasonable variables too, feel free to give suggestions

For your reference, I attach photos of my actual kit.

Also link a couple of videos I made last year

Bad one:

https://drive.google.com/file/d/1L7AMSugEXKaGLcr-v-F42RROKnk1gIBC/view?usp=drivesdk

Good one (I suppose):

https://drive.google.com/file/d/1XolCMfJ8JVIQKaKB8Ju-BlzN5GO1QZ0J/view?usp=drivesdk

Thank you in advance for all your support and expertise.

Peter

Product Automotive lighting Cable Auto part Electrical wiring


Circuit component Motor vehicle Electrical wiring Computer hardware Electronic engineering


Motor vehicle Automotive design Automotive lighting Bicycle part Rim


Automotive lighting Motor vehicle Automotive design Rim Automotive exterior
 

Attachments

See less See more
4
#2 ·
I like your approach for making an adapter for OPV. Unfortunately for other points I cant say much, maybe only I would use quality gauges like Orman or similar. Do you sell them?
 
#21 ·
Good question, I think it worth to try. I'll include this one too.

Why don't you just use the spare connection on the steam valve found on the Gaggias without the OPV? It wouldn't be too difficult to remove the ball and spring and find a suitable adapter for the capilliary tube. It would then be a straight swap.

View attachment 42483
It is probably the best tapping point.

I only have one concern from a manufacturing perspective, it needs - in most cases - a new steam valve, and that's well more expensive than such a kit should be :(

I was thinking of a piggyback for the steam valve, but then the cutout for the knob won't line up :(
 
#6 ·
Good idea for a thread.

As i don't like the idea of meddling with the boiler safety valves in these espresso machines, and, still can't get my head around using the PRVs to control pressure in the puck, i am looking myself at putting a pressure gauge on my Classic.

I've seen a few attempts and mostly, they all work on the principle of taking a tap from the pump discharge and mounting the capillary tubed gauge low down in the machine.

I have a few issues with this.

A) Taking the pressure tapping from the pump side of the boiler is not, i think, best practice. It exposes the gauge to the vibrations and shock from the pump and is as far away from the group head (where you want to be) as possible.

B) Capillary tubing is so small that normal flow into/out of it isn't going to happen. However, i don't think zero flow happens either. Its a dead space/void in the flow path and that's always a bad idea where hygiene or hold up is an issue. I think you will have some movements of fluid but it will be by convection, vibration etc and will be slow. Its a bug pot imho.

C) Putting the Gauge low is again a problem for the issues noted in (B) I see it is common to tee off in a manner which is not free draining, usually with the tee pointing down from the pump tap. Another sump created with a dead head.

I have looked to see how other manufacturers have added pressure gauges to their machines and how they resolve the issues noted above.

Sage, in their Dual boilers take the pressure tap on the boiler side of the system from the multiport block valve (3 way solenoid).

Here is a good video showing the guts of the DB:



I like the gauge fitting etc, but it fails on all of the points i noted above and is not a good idea i think. It suffers from hold up and is on the wrong side of the boiler.

I think the best approach is the Sage DB, but without remanufacturing the valve block footprint, its not possible.

So i would take the tapping of the steam valve or any other boiler port (make one yourself!) and run a normal tube without kinks, up to the gauge which is mounted at a high point above the boiler tapping. I've been looking for slim, vertical pressure indicators, but they are not common.

I shall watch this thread with interest!
 
#9 ·
Good idea, but for those without brazing equipment there shouldn't be a need to braze anything. There is an internal thread in the valve (1/8 BSP?), it should be possible to find a standard adapter to fit this at one end with an external thread at the other end (1/4 BSP?) to fit the capilliary tube fitting.
 
#26 ·
Good idea, but for those without brazing equipment there shouldn't be a need to braze anything. There is an internal thread in the valve (1/8 BSP?), it should be possible to find a standard adapter to fit this at one end with an external thread at the other end (1/4 BSP?) to fit the capilliary tube fitting.
Gas Electrical wiring Circuit component Machine Household appliance accessory


I took a steam valve from a Gaggia Evolution but I'm not sure what connections might be used. The previous nipple has 7mm threads while the male threads that insert into the capillary tube nut I believe is a BSP 1/8 male.
 

Attachments

#11 · (Edited by Moderator)
It's good that we can take some of our conversations onto an official thread @FairRecycler.

Just to be clear, the first video should worry anybody who's doing an install -- a glycerine filled gauge that is vibrating pretty badly. Do you have a guess at the reason for that?

Routing 1: I believe that @MartinB has suggested that the best route is one that goes to the right of the boiler. However, because of the distances involved, this basically means you have to bend the capillary tube close to 90 degrees if you're going to mount it to the left. Cue @ratty in here who just answered my question this morning about the 90 degree bend issue.

https://www.coffeeforums.co.uk/topic/52816-gaggia-classic-2013-refurb/page/3/?tab=comments&do=embed&comment=755752&embedComment=755752&embedDo=findComment#comment-755752

I've stolen his pictures below so we have an immediate reference image. However, I also think that his gauge is different---on the glycerine ones you use, I think the distance to the boiler is much more limiting. Also, if I had to guess, his gauge is more centred.

Motor vehicle Rim Gas Engineering Automotive wheel system


Routing 2: One idea that I have not seen discussed, for those that use a T-junction, is to simply move the T-junction further downstream from the pump. Most installs I assume people are cutting their pump-to-OPV pipe and just choosing a location. Some people cut it closer to the pump, some further away. Has anybody tried to simply extend the pump-to-OPV piping and attach the T-junction further away?

It could be that this has little effect because the vibrational issue seems to be a function of the flow itself and not the actual vibration of the pump in the Gaggia case.

Length of capillary tube: Unfortunately I bought a gauge from shock_waves_shop recently had entered into a debate with him about the shortened capillary tubes. It is unclear to me the magnitude of the effect of a longer coil, but there is plenty of anecdotal evidence to suggest that a longer coil leads to reduced flutter. Here are a number of people who note that replacing their capillary tubes with a longer version reduces vibrations.

If you are curious the snubber that was linked a bit lower is here -- I believe a 1/4" male to 1/4" female porous snubber; basically a valve with a porous medium in it that should reduce the vibrations. It would probably cost about £10-15 more to install once you deal with all the pipe changes.

The post by HB here explains how flutter is a function of the vibration pump and notes the difference in flutter observed over different machines.

Wet/dry capillary tube: In your various install guidance @FairRecycler you note the importance of filling the capillary tube with fluid in order to eliminate air. This is slightly opposed to the links above that have noted that people have taken efforts to dry their capillary tube using a heating gun or butane torch. However, it is not clear to me the effect of either suggestion. See below on the fridge discussion, but it seems you either want a completely liquid filled or completely air filled tube and that's essentially it. If this is the case, then your suggestion is not contradictory to the advice of trying to remove all liquid.

Block on the radiator: As most people have noted, you have a unique design of a block on the radiator rather than a T-junction from pump-to-OPV. Again I don't have a huge insight into this choice; as noted, it seems more stable because it is further away from the pump, so at worst, it doesn't seem to hurt. At best, it helps a lot. But the video you linked of the needle flutter (your "bad" video) is interesting because it shows that this isn't a simple fix.

If the effect of the snubber is to essentially remove non-uniformities in the flow, then placing your block after the OPV might have a positive effect. I'm not sure.

Tapping into the steam valve: @hugoread has done it and might comment: https://coffeeforums.co.uk/topic/48961-gaggia-classic-pressure-gauge-install-video/

Also see this reference by the Polish blogger and includes clear images of how the tapping can be done: https://gaggiaclassicmods.blogspot.com/2015/12/manometr-raz-jeszcze-podaczanie-do.html

Crucially, note that the blogger stated that he did the steam valve mod because his previous T-junction pump-to-OPV gauge indicated -2 bars less than expected. This was later discovered to be an issue of a malfunctioning gauge, not because of the position.

Scientific literature: I tried to turn to the scientific literature for guidance. The main application area for capillary tubing and vibration that is discussed and that might be applicable for us is for minimising noise from refrigerators. For example [1], [2], [3]. However, I'm not sure the applications/results are exactly the same. In the fridge, the capillary tubing is used to pass through the liquid so that it is compressed and pressurised, then moved into an expansion chamber where it turns into a gas. Basically, if I understand the research, then the main theory is that noise is primarily created due to bubbles created at the capillary tubing inlet, which are then 'released' at the outlet and so this produces the noise.

The work, for example, by Daniel Hartmann [2] above and also these people seek to change the inlet to reduce this issue -- i.e. either reduce the diameter of the capillary inlet or lengthen it.

I do not know how this 'jives' with the above questions, but there is some reason to believe that the main vibrational issue is can be dealt with by 'pre-treating' the flow prior to entering the capillary tube, and that this can have a larger effect than what you do with the capillary tubing itself.
 

Attachments

#14 · (Edited by Moderator)
Is there room to fit a 0-20 bar Pressure Gauge Snubber to the back of the gauge before connecting the capillary.
Moving on this idea, does anyone know what type of snubber might work with the least number of extra attachments/tube conversions from the normal T-junction route? If I understand correctly, it's silicone tubing with 6mm internal diameter and 9mm external diameter.
 
#13 · (Edited by Moderator)
Capillary tubes are primarily for Gauge positioning. The best place for the gauge is on the vessel itself. (mostly).

The longer the tube, the greater the lag between action and response. This may be explaining why some notice a difference in needle twitch.

Coiling the capillary tube can be a way of providing heat loss (or gain) which will effect the readings accordingly.

I suspect the main difference people are seeing in needle response may be down to some localised flooding in the capillary tubes, with those flooded systems behaving worse then the non flooded ones ( you can't compress fluid)

If you are trying to come from the pump side, perhaps an expansion chamber is needed ( which is the boiler in effect) to dampen out the hydraulics coming off the pump.

In the OP's instance, the brass block is a great idea but on the wrong port. If it was on the solenoid port, you would have the boiler as a buffer.
 
#15 · (Edited by Moderator)
I tested (non-scientifically) adding additional tubing from pump to T fitter to opv and this seemed to have extremely minimal effect.

https://streamable.com/wg4ld3

The video is misleading because of the camera frame rate. The needle is bouncing very rapidly between the bounces and you're not seeing it. The flutter is about 0.5bars in extent.

Previously when operating this gauge I had about the same (smaller) fluttering as was observed in @ratty recent thread. Then something happened (I've no idea what) that caused a change to worsen the flutter. Air/water mix getting in? Who knows. It's tricky.

I wish I had the money/time/patience/desire to conduct a proper test.
 
#17 · (Edited by Moderator)
Well...haven't we established that? See the various references in the thread that have discussed the effects of the vibration pump.

Multiple people have tapped into the pump and observed different levels of vibration. So the question is whether that can be minimised and to what extent and via what factors...

I apologise as I guess I have not been writing very clearly.
 
#19 · (Edited by Moderator)
This may help..

Here is a picture of my OPV setting gauge

Automotive exhaust Gas Nickel Auto part Gauge


There are two ball valves, one controlling flow out of the outlet and one controlling flow to the gauge. To use it I start with both valves open and start the pump. Once water flows out of the outlet (purging the air) i shut the valve. Pressure builds up and registers on the gauge. At this point the needle vibrates wildly, I slowly close the ball valve to the gauge and the oscillations die down and stop completely when the valve is shut. I mention this because there is a sweet spot just before the valve is closed when there is very little oscillation (and it's not a glycerine filled gauge). That may be a possible answer to the problem, find a way to restrict the size of the connection to the gauge.
 

Attachments

#20 ·
This may help..

Here is a picture of my OPV setting gauge

There are two ball valves, one controlling flow out of the outlet and one controlling flow to the gauge. To use it I start with both valves open and start the pump. Once water flows out of the outlet (purging the air) i shut the valve. Pressure builds up and registers on the gauge. At this point the needle vibrates wildly, I slowly close the ball valve to the gauge and the oscillations die down and stop completely when the valve is shut. I mention this because there is a sweet spot just before the valve is closed when there is very little oscillation (and it's not a glycerine filled gauge). That may be a possible answer to the problem, find a way to restrict the size of the connection to the pump.
Norvin, this is excellent.

This post is maddening for the simple reason that I saw a setup where someone installed a flow limiter in a non-Gaggia machine in order to limit needle flutter and I never clocked its possible importance. In your ball valves, how small is the sweet spot? Do you envision this might be a permanent solution for built-in gauges?

By the way, regarding the theory of using the steam valve in order to reduce vibrations, this is exactly what this person does. The resultant flutter is about the same as I've seen for other pump-OPV T-fitter installs. See the video here. Based on this single video, it's not obvious that the steam gauge location fixes things.
 
#24 ·
14 hours ago, Blue_Cafe said:

Good idea for a thread.

As i don't like the idea of meddling with the boiler safety valves in these espresso machines, and, still can't get my head around using the PRVs to control pressure in the puck, i am looking myself at putting a pressure gauge on my Classic.

I've seen a few attempts and mostly, they all work on the principle of taking a tap from the pump discharge and mounting the capillary tubed gauge low down in the machine.

I have a few issues with this.

A) Taking the pressure tapping from the pump side of the boiler is not, i think, best practice. It exposes the gauge to the vibrations and shock from the pump and is as far away from the group head (where you want to be) as possible.

B) Capillary tubing is so small that normal flow into/out of it isn't going to happen. However, i don't think zero flow happens either. Its a dead space/void in the flow path and that's always a bad idea where hygiene or hold up is an issue. I think you will have some movements of fluid but it will be by convection, vibration etc and will be slow. Its a bug pot imho.

C) Putting the Gauge low is again a problem for the issues noted in (B) I see it is common to tee off in a manner which is not free draining, usually with the tee pointing down from the pump tap. Another sump created with a dead head.

I have looked to see how other manufacturers have added pressure gauges to their machines and how they resolve the issues noted above.

Sage, in their Dual boilers take the pressure tap on the boiler side of the system from the multiport block valve (3 way solenoid).

Here is a good video showing the guts of the DB:

Thank you for the lot of details

A/ totally agree, this made me make the piggyback to the opv.

B+C/ Thank you for the explanation, I got it now.

Sage/ Yes the front layout allows you to do so. Lucky, unlike the Classic :(

Gaggia TS/ I wouldn't mix in HX machines as those boilers half empty, or half full depending on your point of view :) So tapping to the top of the boiler eliminates the water issue in the tube.

"This guy's" Classic on YouTube/

Unfortunately not many Classic in use has this old type of steam valve. :(

just below that, my version:

As in that time I was stick to the water filled tube, I didn't wanted to let the hot water to the gauge as most of them not rated for such a temperature. I never thought of the cooling effect of the coils, my bad. I'll definitely try one with a sufficient coil size, tapped to a piggyback at the boiler outfeed. This would - unlike the steam valve tapping - allow a self draining system too.

Thanks again
 
#28 ·
11 hours ago, phario said:

And since we might be in the mood for contradictory videos:

Haha.

Everything is wrong with this setup, but it works like a treat.

There is also a 90° elbow at the gauge end...

The key is probably the gauge, I can't remember seeing this kind of gauge before.

In his blog he usually details every components - including the polish supplier - used in his mods.
 
#29 ·
Haha.

Everything is wrong with this setup, but it works like a treat.

There is also a 90° elbow at the gauge end...

The key is probably the gauge, I can't remember seeing this kind of gauge before.

In his blog he usually details every components - including the polish supplier - used in his mods.
One variable no-one mentioned before is the amount of air in the boiler, that highly affects the dampening of the whole system.
 
#40 ·
It might be silly, but I'm trying to think out of the box

Would this trap all the water when placed near to the tapping?

Mini 1/4" AF2000-02 Air Compressor Line Water Separator Trap Water Trap Filter

https://www.ebay.co.uk/itm/383595649914
Were you looking to ensure water doesn't get into the capillary?

My understanding is that these are for separating moisture from air, rather than impeding liquid systems.

By the way (not to ignore your question) but one theory I have seen proposed is that the vibrations arise due to water-air mixing in the cap tube. People have advised to use a blow torch to dry any liquid in the cap tube. I have wondered what effect might be from using the boiler to keep heat on the tube so keep it dry
 
#45 ·
I realise that I may be excessively posting, but it is better to collect thoughts here rather than lose them, I suppose.

So the question here is if we can simply isolate the main needle flutter as a function of the vibration pump. Specifically, is it the case that the needle flutter is a consequence of pressure pulsations (and not, say, other mechanisms like air-liquid flow going unstable). Some additional reading from Chap. 5 of the book by Nakamura et al.

Pulsations in piping can be reduced by various methods: installation of an orifice that adds fluid damping, adjusting piping length to change the resonance frequency, installation of a pulsation suppression device (snubber) that reduces the pulsation itself, and installation of a 1/4 wavelength branch pipe that aims to reduce a particular frequency component
See the image below of their Table 5.7 that indicates various ways of reducing pressure-induced vibrations.

Font Material property Screenshot Document Parallel
Font Screenshot Rectangle Number Parallel


Some of the suggestions are quite interesting, and they largely relate to pre-treatment of the flow before it hits the gauge or the modification of the pump-OPV distance. Some of these recommendations seem unrealistic for the community (like those suggestions that relate to knowing exact properties), but suggestions of inserting an orifice, snubber, or side branch might be applicable.
 

Attachments

#50 · (Edited by Moderator)
I've been following this thread with interest.👍

Some of the suggestions [to damp vibrations] are quite interesting, and they largely relate to pre-treatment of the flow before it hits the gauge or the modification of the pump-OPV distance. Some of these recommendations seem unrealistic for the community (like those suggestions that relate to knowing exact properties), but suggestions of inserting an orifice, snubber, or side branch might be applicable.
This reminded me of pre-heater mods where a coil of tubing was wrapped around the boiler or a little pre-heater sump installed. You could easily modify the diameter, length or composition of the pump-OPV and see what effect on damping out the vibrations. 👍

edit: if the pressure is fluctuating - isn't that something you would want to see on the gauge?
 
#49 · (Edited by Moderator)
In a PM I mentioned to @FairRecycler that we should take care not to assume that the liquid filled gauges are free from defects and to ascribe vibrations to the system rather than the gauge (this has largely been what we have done---trying to diagnose the system rather than consider if the gauge is, itself, the culprit).

To that end, here is a cheap dry gauge, mounted direct to the pump-to-OPV line with almost no vibration perceptible to the eye.

https://streamable.com/0ixshv

This is the same gauge I mentioned in my "contradictory" post:

https://www.coffeeforums.co.uk/topic/53171-pressure-gauge-discussion-experiment-improvement/?do=embed&comment=757635&embedComment=757635&embedDo=findComment

Additional details for my reference:


  • The liquid-filled gauges linked in the original post has hole diameter 41mm but requires about 37mm of space behind the gauge when you install it into the Gaggia case. This means you have to be careful with rear clearance to the boiler.

  • The gauge above is linked here but I have not been able to find a Chinese supply source (as you would expect there to be). The necessary case hold diameter is marginally smaller, at 40.6mm, but it has the advantage of much less needed rear clearance, and you only need about 30mm of space behind.
 
#57 ·
trying to diagnose the system rather than consider if the gauge is, itself, the culprit)
And it is confirmed there is an orifice of 1.0mm on my gauges, which is miles away the desired diameter.

If anyone knows an affordable machinist who turns and mills as well please PM me.

I know you prefer non-inline examples, but what are examples of inline flow limiters
As seen on this or on @rattys topic inline taps/valves been used to reduce flow.

However, when I did my (poor!) test, I had a suspicion that gauge vibration was not because of object vibration, i.e. it doesn't much matter how you hold the pump or hold the gauge. It is coming from within the flow itself.
I can't see it's caused by object vibration as my piggyback version should not suffer of that, being bolted to the brew group (well it's only an assumption , but I can't see much chance for that)

Following on from this idea. I've just come across this: https://ashcroft.com/productinformation_pdf/upload/pip-ash-pi-7b-gauge-throttle-plug-selection.pdf

You can buy throttle plugs or screws to vary the size of the orifice, it seems some already have throttle plugs built in.
Wow that's a great one. I thought the 0.5mm orifice will be sufficient, but according to this I should go down to 0.15mm with an air "filled" capillary tube.

Good job I've just finished the 0.5mm piggyback :)

@phario I have a few good reason to stick with the gauges I'm using now, like build quality (full stainless steel body), aesthetics (streamlined professional look)

I admit the gauge you linked could work 10/10 times out of the box, but that chrome effect plastic screams at it's designer, no mention of the stickout at the front. I would rather do the extra mile to sort the 41.3mm one out. Obviously it's my personal preference.
 
#62 · (Edited by Moderator)
The boiler infeed and boiler outfeed piggybacks ready to test.

This morning I'll get a machine done and start this test.

It's not a fair comparison as in the outfeed version there is a 0.5mm orifice as well to reduce the brew flow (I wanted to try it for so long), I'm hoping to get good results with that setup. I left the length of the capillary tube similar for now.

I'm struggling to source push fittings for the PTFE tube version (which I believe to be a great solution) as not many fittings standing both 10-15bars and 150°C.

I found these, but they're sold in a box of 10 or 100. Iv tried several resellers and FESTO UK too with no joy :(

QSF-1/8-6-B-100

NPFC-R-G18-M5-MF

NPQH-D-M5-Q6-P10

NPQH-D-M5-S6-P10

NPFC-R-G18-M5-FM

MA-40-16-1/8-EN-DPA

If anyone has a clue on getting 1 of each, that would be great.

Gesture Electricity Electronic device Electrical supply Elbow


Wood Rectangle Beige Audio equipment Tints and shades
 

Attachments

#64 ·
The boiler infeed and boiler outfeed piggybacks ready to test.

This morning I'll get a machine done and start this test.

It's not a fair comparison as in the outfeed version there is a 0.5mm orifice as well to reduce the brew flow (I wanted to try it for so long), I'm hoping to get good results with that setup. I left the length of the capillary tube similar for now.

I'm struggling to source push fittings for the PTFE tube version (which I believe to be a great solution) as not many fittings standing both 10-15bars and 150°C.

I found these, but they're sold in a box of 10 or 100. Iv tried several resellers and FESTO UK too with no joy :(

QSF-1/8-6-B-100

NPFC-R-G18-M5-MF

NPQH-D-M5-Q6-P10

NPQH-D-M5-S6-P10

NPFC-R-G18-M5-FM

MA-40-16-1/8-EN-DPA

If anyone has a clue on getting 1 of each, that would be great.

View attachment 42956

View attachment 42957
Fantastic work
 
#63 · (Edited by Moderator)
The other question is that I've wondered exactly how hot the pump to OPV water gets. A simple experiment would be to just measure the OPV ejection. It can't be any hotter than that. The point I'll make is that the pump is not rated for hot water. I doubt it gets higher than 100C but it would be good to know so we can choose the tubing accordingly.

I'm unreasonably excited to see your test. I hope it sheds light.
 
#65 ·
The other question is that I've wondered exactly how hot the pump to OPV water gets. A simple experiment would be to just measure the OPV ejection. It can't be any hotter than that. The point I'll make is that the pump is not rated for hot water. I doubt it gets higher than 100C but it would be good to know so we can choose the tubing accordingly.

I'm unreasonably excited to see your test. I hope it sheds light.
By experience it's around room temperature.

Why is it unreasonable??? 😅😅

At this point I'm not too keen on shedding lights light, but promising results. The amount of time in this project makes me feel, I'm aiming for something impossible... I want that baby with no more labor. 😂😂

Yes I do know it doesn't work like that, but the amount of research on applicable push fittings with no joy (knowing I found them, but those not for sale individually), did put me off a bit.
 
This is an older thread, you may not receive a response, and could be reviving an old thread. Please consider creating a new thread.
Top