DIY Metcal 13.56 MHz RF Supply

[Precipice]

Everybody has received their package safe and sound?

I have, and am assembling the pile of parts... Thanks!

[AndersAnd]

Also Thermaltronics (in Australia and USA) make induction soldering stations, and quite well priced replacement cartridges and tips / handles .
http://www.thermaltronics.com/replacement_tips.php
They are color coded into 600F 750F 800F , blue yellow red. 
There is  a keyhole shaped slot in their soldering stand so you can change tips/ cartridges while operating.

Basically any induction soldering iron that has a coaxial connector in the lead should work with the 13.56 system, It's usually a "F" connector = same as used on US television sets for the antenna.

Thermaltronics have already been mentioned and tested to work with this DIY supply earlier in this topic. Thermaltronics was founded by former Metcal employees once the Metcal patents expired.
Thermaltronics are also sold rebadged as Easy Braid: http://www.easybraidco.com/hand-soldering
As far as I know EasyBraid branded solder stations are only sold by DigiKey and Amazon US.
Amazon only sell the top series EB9000 though.
EasyBraid at Digi-Key: www.digikey.com/Suppliers/us/Easy-Braid.page

So so far there's 4 compatible brands to choose from for 13.56 MHz solder pens and tips:

• Metcal
• Thermaltronics
• Easy Braid
• Hakko

[martin5256]

Hello could somebody make a partslist for mauser or another store it would be nice if you could order a kit just an idea

[zayance]

Hello could somebody make a partslist for mauser or another store it would be nice if you could order a kit just an idea

There is a BOM on Reply #330 that was made for the Boards i had ordered, so you have parts refs there,
but unfortunatly the Silkscreen is different from original File of this one, Silkscreen is different for the RF Generator.

Alos all parts for building this one are not available from Mouser, you need some parts that can be found @ Reichelt,
if you haven't read the latest posts i made.
+ Not many are interested on this one, so not sure about a KIT.

Best,


T.

[zapta]

For the people that build these power supply, please post pictures of your soldering stations when done.

[martin5256]

Hello could somebody make a partslist for mauser or another store it would be nice if you could order a kit just an idea

There is a BOM on Reply #330 that was made for the Boards i had ordered, so you have parts refs there,
but unfortunatly the Silkscreen is different from original File of this one, Silkscreen is different for the RF Generator.

Alos all parts for building this one are not available from Mouser, you need some parts that can be found @ Reichelt,
if you haven't read the latest posts i made.
+ Not many are interested on this one, so not sure about a KIT.

Best,


T.

You im back

so im only 15 but im on an electrotechnician school and so on i wouldn´t say people arent interested on a kit but they are to lazy to ask i was looking at parts on farnel and i wasn´t on the half and there were 200€ i mean if you could put the cost down get some boards a nice case and you could create a good brand the design is nice and mamalala had some work with it but deam if he could make a kit i think if he puted it on dangerous prototipes  and the pcb on seedstudios he could make some money


i am interested in metcal irons becouse my curent is a tesla one an it works nice but you know you have to move on

so if we could put all the components in one place to sell it would be more then good and i think people would buy it when it hac a cost like a used metcal

guys it is nice what you do and you are great at it so i wana see some progres


greatings form slovakia

[zayance]

Hello could somebody make a partslist for mauser or another store it would be nice if you could order a kit just an idea

There is a BOM on Reply #330 that was made for the Boards i had ordered, so you have parts refs there,
but unfortunatly the Silkscreen is different from original File of this one, Silkscreen is different for the RF Generator.

Alos all parts for building this one are not available from Mouser, you need some parts that can be found @ Reichelt,
if you haven't read the latest posts i made.
+ Not many are interested on this one, so not sure about a KIT.

Best,


T.

You im back

so im only 15 but im on an electrotechnician school and so on i wouldn´t say people arent interested on a kit but they are to lazy to ask i was looking at parts on farnel and i wasn´t on the half and there were 200€ i mean if you could put the cost down get some boards a nice case and you could create a good brand the design is nice and mamalala had some work with it but deam if he could make a kit i think if he puted it on dangerous prototipes  and the pcb on seedstudios he could make some money


i am interested in metcal irons becouse my curent is a tesla one an it works nice but you know you have to move on

so if we could put all the components in one place to sell it would be more then good and i think people would buy it when it hac a cost like a used metcal

guys it is nice what you do and you are great at it so i wana see some progres


greatings form slovakia

True i get your point, parts at Farnell are usually expensive compared to some other suppliers, some would say, "well if you want genuine and no problems of counterfeit etc... go for those Big supplier" but anyway the BOM gives hints on what to use, package etc... so you can bring parts cost down depending on where and what you choose, giving that you use correct package/model, anyway you got my point.
Of course a Kit would save cash, since you'll have exact count parts and no extras, plus the advantage of buying by multiple etc...
BUT who got the time for this haha , not sure if mamalala would have, and neither would i, plus the fact that it might bother some brand out there who knows???
I made a partial kit for helping my extra boards to get out of here, and that was some time put to it etc...
Interest on these weren't off the hook, and we can understand why, most won't even bother buidling a Solder Station   , since you already need one to do so haha  .

I still need to build mine YET! (soon tough)


Cheers,


T.

[richard.cs]

I've spent some time hunting back through the thread as I'm about to order components. Could someone please confirm if my understanding is correct here.

1) The BOM from post #330 matches the silkscreen boards that I got from zayance?
2) It's electrically the same circuit as posted way back in #103?
3) The component identifiers on the board/bom do not match the schematic?
4) No schematic with matching component identifiers has been posted?

If that's all correct then zayance, would you be able to post a matching schematic please? I assume it must exist since you had the CAD files to get the board made. If it doesn't it's not a big problem but it would save me quite a bit of time.

Many thanks
Richard

[zayance]

Wait didn't you get the email i've sended with the files for it?
Meaning updated Schematic etc...

Basically it's exactlly the same latest board as mamalala only diff here is that there was a prob in the silk on the Generator side,
since i made one PCB to have both designs, Eagle pushed the Silkscreen numbers beside of just copy/pasting (logic and bummer).
I also moved the silk for some components here and there to make the assembly and it's visualisation easier, that's all.
So i made a Corrected BOM AND a corrected Schematic + Build doc according to the right Tname and values etc....
Something that i've sended to people who bought the boards, including you if my memory is not failing already??
Please let me know

I repeat, All the rest is the same

Best,

T.

[richard.cs]

The quick-release style F-plug grips onto sockets with screw threads just fine. :-) For the record my supply will probably end up with a N connector on the output and a N-F adaptor living on it the whole time, it'll make life easier if I ever design my own handpiece.

First solder on my RF board today (did the LM22676 on the hotplate at work in my lunchbreak).

[richard.cs]

I'm currently building up the RF board. The first test with the buck converter worked perfectly putting 11.74 V into my dummy load.

I've hit a problem in the area around IC7 - In the schematic and BOM R40 is listed as 22k, in the build guide it is 2k2, which is correct?

Also having gone through all the documentation again there is green comment text next to R41 saying "200R" despite this being a 1k. Whilst examining this it also occurred to me that the choice of R39 (the pot) or R54 and R55 (fixed divider) will make the circuit perform differently. With the fixed resistors R41 is effectively 1050 Ohms to 2.5V, with the pot fitted and in the middle R41 is effectively 3500 2250 Ohms to 2.5V and, this being a non-inverting amplifier where the value of R41 matters, this has a direct effect on the gain. I have fitted R39.

I'm using zayance's board and the design pack he sent me.


Edited to correct a couple of mistakes

[mamalala]

I've hit a problem in the area around IC7 - In the schematic and BOM R40 is listed as 22k, in the build guide it is 2k2, which is correct?

Also having gone through all the documentation again there is green comment text next to R41 saying "200R" despite this being a 1k. Whilst examining this it also occurred to me that the choice of R39 (the pot) or R54 and R55 (fixed divider) will make the circuit perform differently. With the fixed resistors R41 is effectively 1050 Ohms to 2.5V, with the pot fitted and in the middle R41 is effectively 3500 Ohms to 2.5V and, this being an inverting amplifier where the value of R41 matters, this has a direct effect on the gain. I have fitted R39.

I'm using zayance's board and the design pack he sent me.

Yea, all that stuff was left in that state while i was still experimenting with the performance of the unit. In the actual unit i have here, as well as the ones i gave away, i have (using zayance's part numbers):

R40 = 22k
R41 = 200R
R54 = 100R
R55 = 100R
R39 is not installed.

Sorry for the confusion.

Greetings,

Chris

[mamalala]

BTW, that part of the circuit doesn't have that much influence anyways. A RF signal is picked up before the last L of the output filter and rectified, then fed through an adjustable pot. That is then fed into the OpAmp to generate a volatge that is injected at the center of the divider at the buck regulator. When the tip heats up it causes a mismatch, which makes that voltage rise. Since that is then injected into the feedback network of the regulator, the supply voltage to the RF stage will be lowered.

That particular section was me experimenting with different offsets to that voltage, in an attempt to get a much sharper regulation curve, but turned out not to be really needed that way (thus the footprint of the adjustable R still there, while not being used anymore and instead replaced by the 100R/100R divider).

It will basically work either way. The only important thing is that this stage has enough gain to make the signal large enough, so that the buck converter will regulate down sufficiently. If it would not then the tip will continue to absorb quite some energy, causing it to glow dark-red after a while right at the top of the  slug inside the cartridge (that is, right where the coil in the cartridge is).

Feel free to experiment a bit with that section. It's not perfect, and i guess still some room for improvement there. But then, i'm not that much of an RF guy, so i left it at "good enough to be usable".

Greetings,

Chris

[richard.cs]

Thanks for the reply. I think given that I've fitted all the other resistors, and also just in case I feel like playing with the offset I will leave R39 installed and R41 at 1k as it is now and fit 200k to R40. This would give me a gain of 90 instead of 89, so no significant change. It would rise a bit if the pot is pulled away from centre but I'm probably unlikely to do that and it doesn't really sound like it matters anyway.

Unless anyone can see a good reason not to do that?

[richard.cs]

Hi All,

I will soon be placing another Farnell order to get some of the parts I missed last time. I've decided to build up the controller board too, originally I was just going to add a cheap ebay dc-dc module for the 12V rail and call it done but the auto sleep functionality is just too nice. :-)

I've just re-read the whole thread and been left with a few questions:

1) In zayance's schematic there is, from transformer to output, 47nF x6 series, 22pF x2 shunt, 1uH series, 56pF x5 shunt, 1uH series, 56pF x3 shunt, 400nH series, 56pF x4 shunt. I've read various posts about missing out or fitting extra 56pF capacitors, is the above correct or should I add/remove some capacitors?
2) Is the firmware attached to post #204 the most recent available?
3) Have any binaries been posted or could someone send me one please? I will be using the PIC18F2620.
4) What's the optimum/min/max supply voltage? I've seen 24 mentioned as usable but low, and 26, 28 and 30V mentioned various times. I will probably build with a 50 Hz transformer but I'd like to pick the best secondary voltage.

[mamalala]

1) In zayance's schematic there is, from transformer to output, 47nF x6 series, 22pF x2 shunt, 1uH series, 56pF x5 shunt, 1uH series, 56pF x3 shunt, 400nH series, 56pF x4 shunt. I've read various posts about missing out or fitting extra 56pF capacitors, is the above correct or should I add/remove some capacitors?
2) Is the firmware attached to post #204 the most recent available?
3) Have any binaries been posted or could someone send me one please? I will be using the PIC18F2620.
4) What's the optimum/min/max supply voltage? I've seen 24 mentioned as usable but low, and 26, 28 and 30V mentioned various times. I will probably build with a 50 Hz transformer but I'd like to pick the best secondary voltage.

Attached are three images from my rather dusty development unit that show how many caps i have installed. I did the same cap placement on subsequent units. Again, feel free to fiddle with that a bit, i went for "as much like a sine at the output as possible" plus "large Vpp output into a 50 ohms RF dummy load". Also again, it probably isn't perfect, and some room for improvement left in the filter section. Did you note the update to the coil winding scheme that happened later? Should be documented, if you can't find it i will make some pics of the current ones as well.

Also attached are two ZIP files, one containing only the compiled .cof and .hex files, the other contains the sources (the whole MPLab-X tree, in fact). The project compiles with Microchips C18 compiler.

About the voltage, 24 volts is enough to get it working well, i use an old 24V transformer from an old soldering station. Just don't go too high, the buck regulator can't take too much.

Greetings,

Chris

[mamalala]

Oh, and as i had mentioned back then, be careful with the LCD connection. I did the original design using a display that had VCC and GND swapped, compared to the "standard". So make extra sure that yours does fit the pinout on the board.

Greetings,

Chris

[richard.cs]

It looks like you have, from transformer to output, 47nF x6 series, 22pF x2 22pF x1 shunt, 1uH series, 56pF x5 shunt, 1uH series, 56pF x3 56pF x4 shunt, 400nH series, 56pF x4 56pF x3 shunt. That's essentially what I've got in the schematics except for a missing 22pF shunt (I assume, it looks identical to the 56pF) and one of the final capacitors moved to the stage before. This is consistent with your post 82, zayance's board has an extra pair of pads at final output for another shunt capacitor which it looks like I don't need and I'll still need to stack an extra capacitor on the group before. I appreciate some tweaking might be needed here.

The most recent update I saw on the coils was taking a few turns off when you went from 11 MHz to 13.56 MHz. The instructions I have say 11 turns L3, 14 turns L5, 10 turns L6 but the photos show 13, 15 and 11. The text for the transformer says 13 turns but the photo shows 14. I guess this means the text was updated and the photos were not, do these numbers look right to you?

I imagine a 24 V rms transformer will give about 32 V dc after the rectifier so I'll go with that. LCD pinout noted, that'll get swapped in the ribbon cable if necessary.

Many thanks for all your help.

Richard


EDIT: After comparing the schematics, my PCB and the photos I was wrong about capacitors. Zayance's extra capacitor is in the right place to be the same as in Mamalala's photos
The black soldermask looks very good but it does make it hard to see the tracks, I was looking at the final inductor backwards.

[mamalala]

Here are images of what i used for the RF transformer and the coils, numbered from first to last.

Greetings,

Chris

[richard.cs]

I make that 14 turns bifilar, 12 turns, 15 turns, 9 turns. That's different again 

I'll use your photos as a starting point and go from there. When I'm done I plan to post a log of tweaks and changes I made to get it going.

[mamalala]

I'll use your photos as a starting point and go from there. When I'm done I plan to post a log of tweaks and changes I made to get it going.

Sounds great!

Greetings,

Chris

ETA: And if you are going to experiment there, make sure to get a bunch of spare IRF's, they are easy to fry

[richard.cs]

I still intend to roll my own design at some point, I'm still toying with the idea of a mains-derived DC bus (probably using either phase-angle control or a PFC chip to get the variable voltage), generating the RF at high voltages and getting the isolation from an RF transformer. When first conceived that combination of voltage and frequency put it in valve territory with all the associated problems but someone at work recently drew my attention to these: http://uk.farnell.com/cree/c2m1000170d/mosfet-n-ch-1700v-4-9a-sic-to247/dp/2361497 - A 1700V, 4.9A silicon carbide MOSFET with relatively low gate capacitance for only £3.07. Now that does look interesting...

I'm very tempted to put a few in my next order and see what they're capable of. It might even drop acceptably into your design as a near-indestructible output stage, it would be a nice quick test of if they work at all at 13 MHz.

[extide]

So, has anyone put together an updated guide on the part values to use? I have read through the whole thing. I am planning on building 1 or 2 of these, using the original gerbers uploaded towards the beginning of the thread. If not I might go ahead and put together an updated set of instructions.

Thanks,

This is AWESOME by the way!


EDIT: I am aware of the BOM on post 33 -- however it uses zayance's part numbers. Is there a table to convert them back to the original numbers?

[richard.cs]

I'm building one up at the moment and was planning to put some details together when finished. For now I would suggest you look at posts #378, #385 and #387. An equivilence table between the two sets of component IDs would be nice but I don't have one. There are various conflicts I have found (in zayance's build pack anyway) between the BOM and the build instructions but nothing serious.

Whilst you could use the origonal gerbers I would suggest a few of changes:
1) Add in pads for the extra 56 pF capacitor.
2) Put some thermal relief in. I've found a lot of areas where something that carries no real current (d.c. or RF) like a logic pulldown is unsolderable with a small iron because one end is hard-down onto a large area of ground plane.
3) Consider swapping the LCD pinout to the standard one.
4) A little bit of BOM rationalisation would be nice. For example the Aux supply capacitor for the switcher (can't remember the number now) is currently the only 1206 10nF in the design and could have been the same as the 10nF 0805's scattered everywhere.
5) Depending on your skill with surface mount soldering the pads might benefit from tweaking slightly - they're perfect for reflow but a little tight for hand work.

These aren't criticisms of the design or the work that's been done so far, just things that would be nice-to-have if you're making more boards anyway.

[richard.cs]

Hello All,

I've now got mine to the point where it can be tested as soon as an output FET and heatsink are fitted. Rather disappointingly I've just realised my only M3 tap is broken so I think I'll stop here for tonight and hopefully get the heatsink finished tomorrow.

One comment on that - the mounting holes on the board are 3.2 mm, this is a perfectly normal size for clearance M3 but I suspect that for many people with an electronics-only background getting four tapped holes in position to within perhaps 0.2 mm will be difficult, and being tapped the heatsink holes can't simply be opened out a little bigger. Might I suggest that loose-fitting 3.5 mm or even 4 mm holes might be preferable on a future board revision.