Metcal MX500

[hpmaxim]

I recently purchased a Metcal MX500 and its misbehaving, and I was hoping someone might have some schematics, service manual, tips, etc.

I have plugged in what I believe is a working soldering wand (with tip) and desoldering wand.

When I turn it on, switched to soldering wand, one of the following happens: solid amber light, amber light comes on for a fraction of a second and goes out (and sometimes comes on again for a short time a couple seconds later), or nothing.  I may have seen it flash green once also.

When I turn it on with it switched to the desoldering wand, any of the above happens, but I've also had the green light come on for a fraction of a second and go out, or, on multiple occasions, I've gotten a solid green, and the desoldering tip gets hot (i.e. its working normally).

I've tried the soldering wand in both output ports (but not the desoldering wand), and I've tried multiple power receptacles (but I haven't tried multiple power cords).  I know the amber light indicates a ground fault/open circuit, which usually indicates a bad wand or tip.  I have used the wand/tip as recently as a couple days ago with an STSS power supply and they seemed to be working okay.

[nanofrog]

Amber LED = no ground, so I'd track backwards from there, and see. Could be as simple as a bad joint.

Supposed to be a schematic for the MX500 (here). Manual (.pdf download).

Also, here's a schematic for the SP200 (470kHz model, but should still be of some help if the schematic link doesn't pan out).

[hpmaxim]

Wow, that's great info.

The seemingly random behavior has me a bit stumped.  I certainly suppose it could be a cold solder joint in the ground path, and that might explain why sometimes it comes up working and sometimes it doesn't.  But that doesn't explain why it only works with the desoldering wand, and it doesn't explain whats with the light coming on and then going out (what causes this?)

[nanofrog]

Wow, that's great info.

The seemingly random behavior has me a bit stumped.  I certainly suppose it could be a cold solder joint in the ground path, and that might explain why sometimes it comes up working and sometimes it doesn't.  But that doesn't explain why it only works with the desoldering wand, and it doesn't explain whats with the light coming on and then going out (what causes this?)

You may have multiple issues, some playing into others. Such as thermal cycling on a weak solder joint causing an intermittent ground connection.

Your soldering handpiece may also be bad, so the desoldering tool would likely be your best bet to test the power unit post any fixes. Then see if the normal soldering iron works or not, and go from there.

[BurningTantalum]

Hi,  I would suspect the co-ax feeder to the wand. They are notorious for failing due to flexing, usually within a few cm of the handle. I have repaired a fair few and all have been at the same point.
 I assume that you haven't got an RM2 wand (marked on the F-connector), as this will not work with the MX500 as it doesn't have a DC path through the tip coil as the RM3 has (3R resistor?...), so it shows as a faulty wand.
Photos show a typical failure - pitch of fractures is about 1.25mm and all that I have repaired have been the same for some reason.
Regards,  BT

[BurningTantalum]

Of course, have a thorough and methodical pike around the solder joints on the main board with a good magnifying lens to start with.
The MX500s seem to be quite well assembled generally, apart from the unfriendly design of the hidden switch for the sleep function.

[hpmaxim]

I don't have an appropriate screwdriver to open the unit at the moment, so it might be a couple days before I get it open, but I'd like to point out something with regards to the accessories.

The MX500 I bought had no accessories with it.  The accessories (wand, cable, tip, etc) that I'm using are from my old STSS power supply, which I'm pretty confident takes the same accessories.  As far as I know, these accessories work fine with the STSS.  I will double check and test some stuff more extensively.  According to the documentation that nanofrog provided that the amber light should latch on when it comes on, but I've seen it come on and then go out, and I've seen the light simply not come on.  I really doubt that an accessory would cause that.  I'd think a bad accessory would cause the amber light to come on (and stay on) not come on and then go out or not come on.  As nanofrog said, maybe my soldering wand (or cable) is bad, but I suspect the other intermittent behavior is an indicator of a power supply fault.

[hpmaxim]

Interesting, I just looked at the F-connectors.  The soldering wand says: "STSS RM2 121790" and the desoldering wand has: "MX RM8E  032700?"  The "?" was a character I couldn't make out (it was partially rubbed off), and the "8" might not have been an "8", but I think it was.

I tried measuring DC resistance.  The desoldering wand (including "tip") was roughly 15 ohms (note, I said roughly... its an old Harbor Freight redometer, which  means 15 ohms could be .1 ohms or 30 ohms).  The soldering wand showed open, however the tip itself was also roughly 15 ohms.  So it would appear there is no DC path through the soldering wand/cable.  If you need a DC path to ground in order to avoid amber, that would explain that problem.  What a bummer, I thought the STSS and MX500 systems were interchangeable.  Is there a way to make this work on both?

The other question would be why does it sometimes seem to operate for a fraction of a second and then stop operating, and other times it operates normally? 

I've played with it a bit more since and found the following:  The behavior seems the same between the two output ports.  When the desoldering wand is attached (and selected) I either get solid green (correct behavior), nothing, or a green flash (or a few quick ones) followed by nothing.  When the desoldering wand is either not attached or not selected, I typically either get solid amber (correct behavior), nothing, or an amber light (typically followed by the green light coming on for < 1/10th of a second) and both lights going out and staying out.

The problem may be heat related. The more often its switched on and off the more likely it is to come up with the lights off.  If you let it just sit for a while it goes back to the random start up behavior.  Also, it appears more likely to function when the wand is not selected (in otherwords, repeatedly switching it on and off with no wand seems to work better than with the wand attached -- albeit, not much better).

[BurningTantalum]

This still points to an intermittent fault on the desoldering tool, and the RM2 wand not being compatible with the MX500.

The RM2 will definitely NOT work with the MX500 as it does not have a DC path through handle to the tip so an orange LED is displayed. I think that the RM2 is intended for the earlier RFG 30 power unit (not sure...)
Link out the circuit in the MX500 that is checking the continuity (defeat it) and the RM2 should work the same as an RM3.
OR if the RM2 does turn out to be O/C then you could add a resistor while you have it in pieces, making it into an RM3.

I don't have a desoldering wand (assume a Talon?) but it must be similar to the RM3 in principle.

I just remembered - I did have an RM3 that had an O/C screen braid just by the brass tube in the handle. Forgot that one !

Soldering irons of all types have a tough life in production environments, where most of these will have come from unless known otherwise. I have seen them swung like bull-roarers by assembly workers.

They are a pleasant little exercise to repair - pics available if needed.
Pic shows resistor across the cap in the brass tube of RM3E
BT

[hpmaxim]

This still points to an intermittent fault on the desoldering tool, and the RM2 wand not being compatible with the MX500.

My understanding is that if you have no wand plugged in and turn on an MX500, the amber light should come on and stay on.  It does this occasionally, but not often.  Just as when the desoldering tool is connected and selected, the green light comes on and stays on (and the tip heats up) sometimes, but not often.

This to me suggests there's an issue with the power supply, not the wand.

The RM2 will definitely NOT work with the MX500 as it does not have a DC path through handle to the tip so an orange LED is displayed. I think that the RM2 is intended for the earlier RFG 30 power unit (not sure...)
Link out the circuit in the MX500 that is checking the continuity (defeat it) and the RM2 should work the same as an RM3.
OR if the RM2 does turn out to be O/C then you could add a resistor while you have it in pieces, making it into an RM3.

Yep, at this point, I completely believe you that my soldering wand is incompatible with MX500 (still surprised by this, but I believe it).  Again, I was using the RM2 wand with an STSS-01 supply.

I'm a little confused by this whole resistance thing though.  I would assume that the lack of DC resistance is caused by some sort of series capacitor.  Not sure why they'd do this, but that's the obvious cause.  The tip itself clearly has a very low resistance and I had assumed the wand was essentially nothing more than a coax with an F-connector on one end and a plastic "handle" with a connector for the tip.  Is the purpose of the resistor simply to let the supply know a tip is attached?  It looks like the resistor is in series with the tip, and in parallel to something else (which I'd presume to be the aforementioned series capacitor, but looks more like a glass diode).  So it shows up as a low DC resistance, but the resistor is entirely shorted at at 13MHz? 

The desoldering tool is a "MX-DS1", you can google it for a picture.  It's basically like a little pistol shaped device, but has plug-in tips the same way the soldering handpieces work.  The trigger opens a valve which allows compressed air to flow into a venturi-based vacuum generator which sucks air through the tip (which has a hollow center) into a little paper trap.  It requires a connection to an external air compressor.

But anyway, at this point I think there are two separate issues:1) My soldering wand doesn't work with the unit because of the lack of a DC low-resistance path. 2) Something is funky with the power supply because it seems to work intermittently even when presented with a clearly open circuit output.

[BurningTantalum]

Sorry ! I must remember to re-read before I post.
 And I do know that a Talon doesn't desolder...(long hot drive today...)


The little glass item is a cap, with a resistor in parallel.
And yes, the RM2 definitely doesn't have the resistor. I have an RM2 that I cannot use but repaired some years ago, and took pics.

It seems as if the intermittent MX fault is in the DC sensing area and not the RF section so try removing the transistor that pulls down the error line Q7.
BT

[helius]

glass capacitors are little-known, but have been available for a long time as very stable and low tempco, low capacitance devices. I first saw them in a Digital Equipment LK201 keyboard used as decoupling. They were evidently widely used in spacecraft.
http://www.avx.com/docs/Catalogs/gl-cyintro.pdf has some data. The same part numbers (according to Octopart) were branded by Corning Glass Works, Cornell Dubilier, and AVX.

They aren't available from Mouser, Digikey, or Newark, I suppose due to low demand.

[hpmaxim]

Helius, funny thing is I've been lambasted as to how awful AVX caps are from two different companies I've worked for.  Both preferred Kemet for both performance and reliability.

Tantalum, looking at the schematic I'm a bit confused by how the circuit operates.  From what I can tell U2 latches in an open circuit fault and turns on Q7 (which along with Q9, Q10, or Q11) will switch off Q12, which in turn shuts down the switching regulator IC which powers the 18V linear regulator.  When the 18V regulator shuts down I'd think you'd lose the high-side of both amber and green LEDs shutting them off.  While this is what's happening (sometimes), its not the intended behavior.  My understanding is once the amber light goes on, its supposed to stay on until the unit is shut off (which happens sometimes, but not often).  What am I missing?

[BurningTantalum]

Hi,

Assuming that you have the same circuit drawing as me - there seems to be only one available - it isn't drawn very clearly. I just had a quick look and it seems that the LEDs and U2 are supplied from the monolithic reg U8 7818 18V. This is ON all the time that S1 is closed. It is supplied from a rather odd centre-tapped full-wave rectifier with the diodes in the low side. I repaired a US made car battery charger with the same quirk recently.

I will try disabling Q7 and plugging in my RM2 later today

I only have one unit here at present, I have others that are 10,000 miles away.
In this one I had a failed U4 and dead Q4, now the output of U4 needs to be very much higher than in the circuit description to get a decent heat in the tip.
When you have fixed yours I would appreciate some idea of U4 output V when cold and when idling - in case the description is wrong.

BT

[hpmaxim]

I think I have at least two issues, one is the lack of DC resistance of my RM2 wand.  I'm not that concerned about that at the moment because its an understood issue.

The second issue I *KNOW* I have is that when I turn on the unit with nothing attached, the amber light comes on (correct behavior) but often shuts off shortly after (incorrect behavior).  With my MXDS1 plugged in, it usually goes green (correct behavior) and often shuts off shortly after (incorrect behavior).  Since we don't know for sure the MXDS1 is working properly, I'm inclined to focus on the amber light and hope the green light issue is related.

I'm still trying to understand the schematic.  I can't figure out what U4 is actually doing.  It's a switching buck regulator but it doesn't seem to be feeding anything that requires any significant current.  The current path for the power amp seems to be in parallel to the center tap and through Q1 (which I think is some sort of over-current shutoff?)  The center tap seems to feed U8 (18V linear regulator) which is the DC supply for all the logic and linear ICs primarily for feedback control (open-circuit detect, sleep, etc).  What is bothering me is that the amber light is fed on the high side by U8, and on the low side by U2 acting as a latch.  Once U2 goes low, it should stay low and the amber light should stay on.  If its not staying on, it seems possible U2 failed, but its not likely.  If U2 failed causing the amber light to go out, then it would probably allow the green light to come on, but it doesn't.  The other alternative is that the 18V from U8 goes away.

However, its also odd that it typically fails within a second of turn on or not at all.  It's not like the thing operates for 20 seconds and then shuts off.  It behaves like its being latched into failure (and much like a latch, comes up in a random state).

[BurningTantalum]

Hi,
U4 supplies Q4 via a DC path through L3, L11 and T2 ie the heater current to the tip, chopped by Q4.

The rectifier is a bit odd -
The Bridge Rectifier (4 diodes) supplies the 56 Volts to U4.
If you then ignore that and look at the centre tap of the transformer in conjunction with the first 2 diodes of the bridge, it is a Full Wave Rectifier.
The usual layout of a Full Wave Rectifier would be to connect the centre tap to ground/0V, and each end of the winding to a diode anode- the cathodes being connected togather as the +ve rail.
This circuit is 'the wrong way up' to the normal configuration so that this supply, and the Full Wave Bridge supply, can share a common ground (I assume...)

This 30 odd volts supplies the 18V reg etc. An odd design, but allows one single-winding transformer and one bridge rectifier. The 7818 will probably not accept 56V on its input.

Re: the problems with the Green/Orange circuit - You could do with an El Cheapo scope or a decent test meter. I would leave the PCB in the case half for safety's sake - the mains inputs from the filter are not very friendly -  and solder small wire 'tags' to pads/tracks of interest to make connections easily, particularly Gnd connections for meter/scope. I use tiny test clips and wires to take the connection off a PCB under test to a 'tag board' or 'breadboard' to eliminate the chance of a stray prod doing some mischief on the PCB.  I have a pair of 1.0 mm meter prods that poke into the breadboard quite happily. Sorry if this is eggs and grandmothers to you !
BT

[hpmaxim]

Okay, feel free to call me an idiot, but I finally got a screwdriver and am now trying to open the case.  I took off all four screws T15 screws in the back, plus one in the center (which was probably a bad idea).  I can pull one end apart by maybe 2-3 mm, but not both sides simultaneously, and not more than a couple mm.  I tried prying on it with screwdrivers, and I tried asking a friend for help.  They just seem very stuck together.  Any advice?

[zapta]

Okay, feel free to call me an idiot, but I finally got a screwdriver and am now trying to open the case.  I took off all four screws T15 screws in the back, plus one in the center (which was probably a bad idea).  I can pull one end apart by maybe 2-3 mm, but not both sides simultaneously, and not more than a couple mm.  I tried prying on it with screwdrivers, and I tried asking a friend for help.  They just seem very stuck together.  Any advice?

IIRC, you also need to remove the nuts from the two front connectors.

As for the center screw in the back, you just need to release it a little bit but you should able to put it back.

[hpmaxim]

The F-connectors in front are in a recessed cavity and I do not see any nuts, I presume there are nuts but they are on the inside -- but without getting it open, I don't have acces to them.

[zapta]

The F-connectors in front are in a recessed cavity and I do not see any nuts, I presume there are nuts but they are on the inside -- but without getting it open, I don't have acces to them.

did you remove the little plastic cover in the front? Try to pull it out carefully.

[hpmaxim]

Wow, I didn't even realize it was a plastic cap, I thought it was part of the metal case.  Got everything apart.  Nothing appeared damaged, there were a few odd things I noticed:

There was a MAD1108P (8 parallel diodes), that seemed to be missing pins 1 and 16.  The PCB didn't have holes for it, so I guess it just wasn't used and Metcal for some reason decided to cut off or damage the legs.

There was a three terminal device that looked like it had a little plastic ring around it that had a crack running down it.  This might have been intentional.  It almost looked like a heatsink, but it seemed like it was plastic.

A number of the traces looked like the color had changed from green to green-brown.  Possibly from excess heat?  Nothing looked actually damaged though.

No signs of bad electrolytics or cold solder joints or burn marks, or anything else obviously damaged.

[zapta]

You do have that document from the internet with schematic and description, right?

Just making sure.

[hpmaxim]

Yeah, just looked through it again.  Everything that I thought looked weird appears to be normal.

After playing with it some, I think its likely the problem is either the power switch or a loose connection to the power switch.  I couldn't get it to work and then I moved the power switch a bit and it worked perfectly, when I tapped on the power switch, it stopped working, etc, etc.  Its made worse by the fact that I'm not real confident of my receptacle and/or power cord (if I jiggled the power cord in the receptacle it could stop working too).

Thanks for the help everyone, sorry it was such a boring (hopefully) conclusion.

[BurningTantalum]

Hi again HPMaxim,

Just a thought - it is easy (voice of experience...), on refitting the PCB into the case, to fail to engage the 1/4" tags on the end of the mains filter can with the receptacles on the PCB.

Regards,  BT

[hpmaxim]

Sorry, when I said "receptacle" I meant the receptacle in the wall, not on the Metcal.  I played with the switch some more, and noted that when I put ANY shear stress on the center pin the switch opened.  I got a replacement for the switch and used a different receptacle and the problem seems to be fixed.  Now, the only problem is I'm still getting the amber light.  I guess this leads to the next question:

Is it possible (and how) to open the wand in a non-destructive manner so I can perform surgery (add the resistor in parallel to the coupling caps) -- which in and of itself presents a challenge since the soldering wand that I'm modifying is my favorite soldering iron.