Steam Cleaner. Re-Birth of a Karcher MPDS

[awemawson]

The last few times I've used the Karcher MPDS I've had an annoying water leak from down low, apparently from the water inlet arrangement,  certainly not high pressure as in pumped water. So the unit has been in the workshop for a few weeks waiting for me to get round to finding the problem.

So first thing this morning I at last gave it a good going over, and what did I find - a pin prick hole in a 1/4" male / male elbow. A quick look through the bits box failed to find one so a repair was called for !

A quick clean up in the cabinet grit blaster of course made the hole bigger, and I decided that a nice blob of SIF Bronze brazing rod would cure the problem and if I ended up filling the elbow with bronze it would be easy to drill it out again.

The surface tension of the bronze was in fact sufficient for it not to wick inside, and it's all now back together with Loctite 542 sealant curing as I type - perhaps now it can be evicted from the workshop to gain a bit more space !

[Sea.dog]

I'm not sure I'd call that a pin-prick.

The elbow has definitely seen better days judging by the thread on one side 

[AdeV]

I'm not sure I'd call that a pin-prick.

Exactly what I was going to say! 

Makes you wonder - why do they make plumbing parts out of stuff that goes rusty in contact with water? 

[awemawson]

In fairness it was probably galvanised when installed - but that was quite possibly 40 years ago !

This unit is a survivor from the early Cold War when 'limited battlefield nuclear weapons' were in vogue, and was intended to decontaminate vehicles in the field.

So it was probably retired from active service around 1980 but made quite a while before then.

But you are right - bronze would have been a better choice.

[awemawson]

I'm not sure I'd call that a pin-prick.

The elbow has definitely seen better days judging by the thread on one side 

It got bigger as I poked it 

[aeromad11]

Thanks for the great pics and info on the Karcher MPDS! I recently picked up an LMDS which is very similar to the MPDS. I am having similar issues with the burner and ignitor.
 Due to congestion in the control panel, I am looking for a  user/parts/service manual to make troubleshooting easier. I did see the training pdf which was very helpful but doesn't quite answer all of my questions. One question I had was in reference to the knob on the flow /pressure safety block, I'm not quite sure what the function of the knob is. Possibly a lockout when pulled and a bypass when pushed? Do you by any chance have any manuals? If so, could you share or refer me to a source? Thanks in advance.

[awemawson]

All the documentation that I have is attached to the first post in this thread, and I don’t think mine has the knob that you are referring to.

Can you please post a few detailed photos of your unit?


Thanks

[aeromad11]

I will upload some pics later today. The safety block I was referring to is the same as the valve in your pic  "Water-Flow-Sensor-Housing.jpg". It has a pull-up position that latches, and a push-in position that is momentary.
   I spoke with Karcher and they are trying to get me technical 
manuals. Thanks

[awemawson]

I recently wanted to bring the Karcher out of storage to steam off the considerable hard grease build up on my newly acquired JCB 3CX. I knew it needed a new battery but also that there was a pressure feedback issue from the HP side back to the feed water so it would need a bit of fettling.

After jumping through a whole load of hoops to extract the battery and fit a replacement (covered on the JCB thread here, post #82 https://www.madmodder.net/index.php/topic,13429.75.html) and coming then to the back feeding issue I realised that the pump needed a full strip down and rebuild.

First issue was access - all the bits are at floor level - solved by putting it on the forks of the JCB and raising to a convenient height.

I could then start removing bits to examine the water valves. It's a twin cylinder pump, each having an inlet and exhaust valve but with some complicated plumbing between them to form other functions.

The valve are a plain flat disks pressed onto raised seats by a weak spring, the seats appear to be stainless steel inserts pressed into the brass casting.


So lets start pulling it apart . . . .

[awemawson]

Now looking at the valve seats it was obvious that they need some form of re-finishing. I went through several ideas - perhaps a custom HSS cutter hand driven - maybe a spigot of suitable size with silicon carbide paper glued on - maybe a standard 'spot face cutter' ?

All these ways of doing it would leave debris in the pump bore that would be pretty impossible to remove, so I discounted them all in favour of a total strip down so that proper cleaning could be applied as the final stage before re-assembly.

So today the strip down began. The construction really only allows the brass casting that comprises the bore and valves to be withdrawn leaving the pistons attached to the crank, crank case and electric clutch unit. Not sure how easy it's going to be to get the pistons back in those bores. No rings, just VERY precise fits, with packing glands to keep the water and grease inside.

[awemawson]

I then decided to give the casting a thorough de-grease, using Toluene (as I happen to have a few gallons of it)

It was at this stage I discovered two things:

A/ Toluene dissolves the paint - no big issue - happy to lose  the paint.

B/ More seriously - the bores of the cylinders are not machined brass as I had thought, but are brass at the outer end with a very thick sleeve of some composite material for the majority of their length.

Quickly stopping the use of the toluene in case it affects this composite material adversely, I had intended to blow off the casting with compressed air . . .

. . . oh no chum . . .your compressor has packed up 

It was at this point I decided to call it a day - I do have a reserve compressor that I can bring into service but really DIDN'T want the distraction !

[philf]

I had intended to blow off the casting with compressed air . . .

. . . oh no chum . . .your compressor has packed up 

It was at this point I decided to call it a day - I do have a reserve compressor that I can bring into service but really DIDN'T want the distraction !

"There's a hole in my bucket dear Liza"

[modeng200023]

It couldn't happen to a nicer person!! Perhaps you should use your 'spare time' to write a book about your experiences with your collection of equipment going wrong just when you need it.
John

[RussellT]

Now looking at the valve seats it was obvious that they need some form of re-finishing. I went through several ideas - perhaps a custom HSS cutter hand driven - maybe a spigot of suitable size with silicon carbide paper glued on - maybe a standard 'spot face cutter' ?

Tap reseating tool?

equipment going wrong just when you need it.

Yes, I find that stuff seldom goes wrong when it's sitting on the shelf.

Russell

[awemawson]

Your'e right there Russell!

So first thing I gave the Hydrovane another chance, but no it wasn't having it. Pumps up to about 50 PSI then struggles to get any further. So the reserve Hydrovane was brought out to play. First one is single phase, second one is 3 phase, but both have been connected previously where the compressors live (my Portakabin) so the wiring and plumbing was straightforward. Reserve plumbed in, plugged in, powered up, and brought my compressed air network rapidly to the 100 psi that it's regulated at.

OK time to go searching spot face cutters. (I knew that I had a box of them stored unused for years), and sure enough not only did I find them, but measurement show that they were tolerably flat across the cutting edges and  there was one of just about exactly the right 20 mm diameter. But there were no arbors. OK have to make an arbor, but it would be an oddity, as the drive is 8 mm square with rounded corners, but I'd need to make a guide bush to ride in the 14 mm through bore of the valve to keep concentricity.

I think I've come to the conclusion that I may as well turn up an arbor to take washer shaped discs of silicon carbide paper that has a spigot smaller than the 14 mm through bore but riding is a 14 mm bush in that bore. This way the cutting isn't taking place at the edge of the paper but in the middle. Also depending on how hard the valve seats are I can make as many silicon carbide washers as I like (I have a 'washer punch' that cuts i/d and o/d simultaneously dependant on which punches you assemble)

. . . enough excitement . . coffee time . . .

[awemawson]

This afternoon, having first performed a bubble / leak detector test on the mornings air plumbing (all OK  ) I started on the valve grinding / lapping / recutting gizmo.

First I tested that cutting silicon carbide paper with the washer maker was practical - it is ! My 20 mm washer cutter is producing 21 mm washers so I had to go down a size.

Then I turned a 20 mm bar end down to 19 mm to clear the threads on the valve block, with a 6 mm spigot to locate the washer of  wet and dry paper. I drilled and tapped the spigot M4, and made a 14 mm o/d 6mm i/d collar that slid onto the spigot to pinch the silicon carbide paper, and also form the guide bush in the through bore for the valves. This bush is retained by a cap socket screw and washer.

First tentative tests show that it seems to work quite well but I've run out of time and now need to do other things. The results in the picture are literally 5 turns of the tool lightly pressing it into place by hand.

[tom osselton]

Not bad!

[awemawson]

This morning I had a lapping session - firstly on the four valve seats that are fixed in the pump casting. They have come out OK'ish but there is still pitting apparent. I may have  a further go at them tomorrow if I wake up full of enthusiasm !

Then I moved on to the disk valve face seals that press onto the valve seats by way of a weak spring. The first water inlet valve did have a distinctive shaped bit worn out of it which took quite a bit of work to bring up to a fair surface, but the other three weren't too bad - minor pitting.

There is a ball valve built into the casting of the pump, that apparently returns the output of one cylinder to the input side to halve the pumped volume when producing dry steam - I assume that the boiler cannot cope vaporising the full 1450 litres per hour that the pump produces and this seemingly reduces the flow to 740 litres per hour. Burner output is 420,000 kJ so some mathematically gifted person can no doubt confirm that!

I could find absolutely no way of removing the ball element, which has one quadrant as a plain sphere, two quadrants with a through drilling at 90 degrees the 4th quadrant being apparently not involved. This of course could be another path by which high pressure water is playing havoc with the input, so I wanted to examine the valve seat - sadly not possible.

I've decided to remove and blank off the twiddly bits that allow sucking in of chemical agents, and dosing with softening chemicals for the time being, I can easily add them back if and when I find I've solved the problem that I'm chasing and I don't want the complications that they add to the party.

Going through all my documentation for the hundredth time I've come across a drawing that implies that the pump may actually be a 'displacement pump'. This means no super tight cylinder fits, just tight packing to keep the water in. This technique is sometimes used with single acting hydraulic cylinders where the rod acts as the piston and the oil just displaces it.

I am fairly sure that the hydraulic accumulator that cushions pump pulses needs re-charging but I have no idea how to test it. The system operates (in theory) at 60 bar - does this mean that the compressed nitrogen in the accumulator should be at 60 bar ? If anyone is skilled in this direction please shout out.

[awemawson]

I did in fact sneak out again last night and have another go at the valve seats - they are not perfect but will have to do - I'll see if they give issues when re-commissioning.

It occurred to me that it would be sensible to test the safety blow off valve before putting everything back together, so I rigged up my 'PRC1' test pump that goes up to 60 bar. Oops - the valve doesn't open at any pressure that I can give it - actually a bit more than 60 Bar according to the gauge 

So I dismantled it for examination. It's a fairly conventional spring loaded face seal valve that probably would work very nicely were it not totally encrusted in lime scale. Now the book says that the safety or Pressure Relief valve should blow a bit when the pump starts - obviously it hasn't been doing this - is THIS our smoking gun ?

So a good soaking in citric acid and a clean up later it was restored to better health. I've set it to a conservative 40 Bar as it can easily be adjusted when things are back together and running.

I will also test the pressure switch that disconnects the electric clutch driving the pump before powering up. As I mentioned in an earlier posting I've never been happy that it's trigger point is correctly set, and with the PRC1 (which I'd forgotten that I had !!!) I can test it, but a job for another day.

[hermetic]

after all this tension build up, only a video of you in a Santa outfit cleaning the JCB with the steam cleaner is going to assuage the yearning! I look forward to it!
Phil

[awemawson]

Don't hold to many hopes Phil !

This morning I set up to check the pressure switch that controls the stopping and starting of the pump via an electric clutch.

AS I think that I mentioned before this has given me trouble in the past. A vertical threaded rod is one end of a piston exposed to water pressure, and it carries a 'spool' that rises up and down with changing water pressure. Into the space in the centre of the spool is the operating leaf of a microswitch. But this isn't 'just' a microswitch in the ordinary sense, it has two stable states. Flick the leaf upwards and it toggles into the off state, and flick it downwards and it changes to the on state to start the pump.

When I got the machine the spool was broken, and actually consisted of two 'top hat nylon washers' of the type often used to insulate things passing through panels. This almost certainly wasn't original, and it's been guesswork getting the design of one right. The set point can easily be adjusted by nuts above and below the spool, bit it always seems that if the high pressure limit is set right it fails to turn back on when the pressure drops.

Today I've made yet another spool of slightly narrower centre section and although it is working it's upper pressure is somewhat lower than i'd intended -  but it is switching reasonably reliably so good during the pump recommissioning phase so I'll leave it as it is for now.

To connect the test pump I needed to adapt from 1/2" BSP to the karcher pipes of 18mm x 1.5mm pitch and the only thing that I had to hand was the pressure relief valve, so it got pressed into service !

Next step - reassemble the pump!

[awemawson]

So - how did it go?

Well, like the Curates Egg, it's 'good in parts'

Putting it back together I was concerned that lining up the two pistons and bores might have been an issue, so initially I put the packing glands on the pistons to leave less to align. This was a mistake! Once I put the pump casting back on it's studs it proved impossible to start the threads that pull the packing glands into the cylinders. So the pump casting had to be removed, glands loosely screwed in, then the casting put back and bolted to the crankcase. After that it was just a case of putting all the gubbins back on facing in the right directions!

Starting up, the engine leapt into life, and the pump started pumping - no high pressure back feed to the water inlet, so that at least is cured. However the pressure switch is again playing up. It started switching the clutch off at about 15 bar - I did tweak the spool nuts but I couldn't get it to be consistent in where it switched off the clutch.

Pump is producing pressure up to whatever I set the relief valve to (yes I tweaked that as well!) it's just that bally pressure switch 

The hot water side is working, producing boiling water with no problem - the flame failure circuit is a bit iffy - probably needs the window on its cadmium sulphide cell cleaning.

I am very tempted to design a replacement pressure switch arrangement if I can find anything suitable - stuff in this pressure range tends to get a bit pricey. I suspect that the current piston mechanism and it's seals are perhaps getting a bit tired and not responding as fast as they should. Remember, it's always been temperamental.
 

[awemawson]

I have tracked down a 60 bar adjustable pressure switch - actually 0 - 220 bar - (eBay £15) which has a 1/4 BSPT male thread so my intention today was to remove the original pressure switch from its mounting flange, and make up a matching thick flange and gasket to mount the new switch.

. . . best laid plans of men and mice . . . ! Having removed the original switch it became apparent that it's mounting flange would leave far too narrow a 'land' to satisfactorily seal against 900 psi. The original switch seals not by the flange but by a series of O rings in its bore. Rather than commit myself to endless blown gaskets another solution was needed - a bit of lateral thinking called for.

Well, it matters not a jot WHERE in the pressure system the switch lives - were I to re-install the original to 'seal the hole' and find a suitable pipe joint to insert an adaptor then all will be well with the world . . maybe !

It turns out that the flexible hydraulic hose feeding the original switch housing has enough room round it to be a suitable candidate, and this means that the wiring should even just about reach.

So the hunt is on for an M18 x 1.5 mm pitch Male / Female cone tee fitting with a 1/4"BSPT female branch. I very much doubt that they exist but I'm sure something can be made !

(amusingly I noticed that the cast body of the original pressure switch has it's alloy composition cast into it!)

[vintageandclassicrepairs]

Hi Andrew,
Time to make some more swarf 

John

[awemawson]

I’m waiting on delivery of the pressure switch which probably won’t be until after Christmas now John.