DIY Metcal 13.56 MHz RF Supply

[KeepItSimpleStupid]

Interesting questions:

There's a tool called a "transfer punch".  The idea is the tool fits in the hole to be transferred and has a point in the center.  That gets you the location of hole #1 (from the wrong side), of course.  To drill from the "bad side" get an "aircraft bit".  e.g. http://www.mscdirect.com/browse/tn/Holemaking/Drilling-Drill-Bits/Metalworking-Multipurpose-Drill-Bits/Aircraft-Extension-Drill-Bits?navid=12106188

The "transfer punch" is not meant to center punch.  Use a real center punch for that.  Automatic center punches are the way to go.  You just push rather than hot with a hammer.

Always when manually transferring something, transfer one hole and secure it.  Then transfer another, secure.  Transfer another and secure.

One of the best drills for sheet metal is a pilot point or brad point bit.   The hoes, you will get will be round and not these triangular holes.

You may be able to find a bezel that will work.  i.e. http://www.hobbytronics.co.uk/lcd-bezel-16x2 that will hide your cutting skills.

See https://www.lowes.com/projects/woodworking-and-crafts/hand-saw-buying-guide/article

You can do a lot with a hack saw and a coping saw.  Generally you drill the corners with a hole big enough to insert the coping saw blade or even the hack saw blade or even an electric jig saw.  The blade in the coping saw is removeable, so you insert it through the hole and then attach it back to the saw.

You can clamp some hard steel as a straight edge.  If your dealing with sheet metal, there is  a nibbler you can use. 

Clean the edges up with a file and paint.

The next step up is a moto-tool like Proxxon or Dremel.  The Dremel has way too much runout (side to side motion).  there are youtube videos comparing the two.

Large round holes are best done with a step drill bit.

[Monkeh]

Can anyone help me to find some good chinese manufacturer of ceramic pcb ?
I want to order some ceramic pcb for the rf stage .

Example :

Uh.. why do you want a ceramic board for this?

[Monkeh]

Can anyone help me to find some good chinese manufacturer of ceramic pcb ?
I want to order some ceramic pcb for the rf stage .

Example :

Uh.. why do you want a ceramic board for this?

Because i have already build my own unit ,but the problem in my case is the heat.
So i need an type of pcb which would be resistant to heat.

.. No, you need to get rid of the heat.

[az113]

Hello again to all.
In search of incredible, i return to identify a MX-RM3e or newer MX-H1-AV tip's plug.
May be anyone identify it?

I buy some SMB connector, and i was very close - but no, it is slightly different.
Outer diameter of SMB female socket is 3.7mm instead 4mm in Metcal's tips, and inner pin diameter is 0,5mm, instead 1mm in Metcal ;(

[David Hess]

Because i have already build my own unit ,but the problem in my case is the heat.

So i need an type of pcb which would be resistant to heat.

Like others said, this is the wrong way to go about it unless there are other considerations like physical size or operating environment.

What kind or type of RF transistor did you use?

High power transistors including RF ones have provisions to heat removal beyond only the printed circuit board.  It would be better to select a different part which is packaged in a way which makes this possible.

[macboy]

Hello again to all.
In search of incredible, i return to identify a MX-RM3e or newer MX-H1-AV tip's plug.
May be anyone identify it?

I buy some SMB connector, and i was very close - but no, it is slightly different.
Outer diameter of SMB female socket is 3.7mm instead 4mm in Metcal's tips, and inner pin diameter is 0,5mm, instead 1mm in Metcal ;(

It is an "F" connector, most commonly used for 75 ohm TV antenna connections in North America.

[Monkeh]

Hello again to all.
In search of incredible, i return to identify a MX-RM3e or newer MX-H1-AV tip's plug.
May be anyone identify it?

I buy some SMB connector, and i was very close - but no, it is slightly different.
Outer diameter of SMB female socket is 3.7mm instead 4mm in Metcal's tips, and inner pin diameter is 0,5mm, instead 1mm in Metcal ;(

It is an "F" connector, most commonly used for 75 ohm TV antenna connections in North America.

Not the handpiece connector, the tip connector. Fs are rather larger than 4mm.

[Nitrous]

Hi guys,

This is a great project. In more ways then the Metcal soldering station.
Has anyone "recently" picked up the challenge of putting a kit together?
Someone purchasing SMD parts in bulk with a simplified PCB bound be great!
Anyone else interested in someone offering a kit?

Nitrous

[nowlan]

Just wondering how this rates vs buying a hako or weller. Do they perform? What is total cost with handle?

[richard.cs]

The performance of mine is very good, and I like the smart sleep function which isn't supported on a standard one.

[NavyBOFH]

Great, we are back. 

What's new, My RF-Borad is finely working, the problem seem to be the XFMR, chris send me one of his and all problems went away. He only explanation that I can find is that RS send me the wrong ones. I will test them both next week with a network analyser to see what's the diference. I will also test a few Metcal tips.

Meanwhile I finished my controller board design placed an order at iTead, my board are already on there way to Portugal.

Paulo

Resurrecting the dead - but I like your layout for the control board! I was always wondering if you ever finished up your RF board design...

As for some of the more recent posts, I am thinking of grabbing the more recent fixes and stuffing them into the PCB somewhere and giving it a crack at making a handful to see how it works out. Mostly I am not a software person so I hope the firmware from 2013 is still the best for this project. I have a legit Metcal MX-500 at work with handle and about a half dozen tips to 1) assemble my board with and 2) compare the board with the "real deal" before ordering a Hakko handle and some tips for my own home use!

[fireworks]

  If you do that, please post your results here.
  I'm interested in this project and I think others are too. 

[NavyBOFH]

I'm getting ready to order a set of PCBs to give it a spin. I had to re-read the entire thread but Paulo seemed to design his own control board to run off Arduino and his own code, while Chris' board ran a PIC controller and its own code. I think though my scouring I have found the most settled on BOM, PCB designs, and firmware.

Now the voltage protection circuit mentioned as a fix was probably intended for Chris' control board but I don't know if it was meant for Paulo's board as well - I'd have to look at both sets of schematics and see what the case might be. I also need to double check that there isn't conflicting information between build sources.

In the end I'd like to have at least one working setup to use on my own - barring any issues coming from making PCBs or sourcing the right parts. The debate I had for the longest time was which board to use for control and I think I'm going to print both just to experiment. Sadly I'm not much in the programming arena so I am going to zip up everything I found and post it here so I can get suggestions and ideas before sending out for PCB orders and taking orders for them on here.


Sent from my iPhone using Tapatalk

[NavyBOFH]

This is what I have come up with so far: I have zipped each functional section into its own ZIP. One is the RF board, which seems to be the latest as of June 2013. The BOM is from zoltan and seems to be what Chris references as the "best list" available.

The control boards are the PIC board which Chris designed, along with his latest firmware, and the Arduino version from Paulo, with his firmware. Neither firmwares have been checked by myself so I don't know where we stand on those or if improvements can be made. There is also no BOM for either board, which might be helpful to some.

Let's see if we can complete these archives, get a good BOM for all of them, make some boards and test them out and refine the firmware as needed. If anyone want to try their hand at putting the over voltage protection circuit in the board that would be awesome since my free time is sparse in the next several weeks.

https://drive.google.com/open?id=0B9MnfIuHNf_lNFlZaEwzbXdwOGc

[NavyBOFH]

Here's my rough draft of including richard.cs' over voltage fix into the controller PCB. It's not the cleanest work I can do - nor am I sure I routed everything 100% (never used Eagle before so learning curve plus everything else)... but I hope someone will either approve it or tweak it so we can move to making some PCBs and getting this project started again for those that want it!

[NavyBOFH]

Here's some screenshots of the PCB and the addition to the board for those that are mobile or don't have Eagle. I amended the schematic file for all of it too, which helped slightly with the routing process.

Other issue is the 2N7000 isn't in an SMD package until you go to the 2N7002 which changes resistance values as well. I don't know if it will make a large difference for this use or not, but if not then I would also think that a SOT23-6 package for both transistors would save some wiring and space as well. The diodes seemed to be a 1-for-1 between a 12V, 10V, and a 5V6 model, so hopefully there won't be much more to change.

EDIT: I noticed I left out a trace from Q2 gate to Q3 drain. I corrected that issue in my files here and posting here for continuity. I will send the files to anyone interested just to make sure if someone is changing or fixing the files we can all be on the same page.

[richard.cs]

Your undervoltage cutout seems to match my schematic, and I can confirm that a 2N7002 would be fine (or pretty much any other low power SMT MOSFET, it's not critical.

I don't remember the failure mechanism, but I remember several people reported blowing up the RF FETs, and Chris said cleanly shutting down the supply would always prevent it.

[NavyBOFH]

Your undervoltage cutout seems to match my schematic, and I can confirm that a 2N7002 would be fine (or pretty much any other low power SMT MOSFET, it's not critical.

I don't remember the failure mechanism, but I remember several people reported blowing up the RF FETs, and Chris said cleanly shutting down the supply would always prevent it.

Thank you for getting back to me with that answer! I saw some people reporting it as well, and some of it had to do with the FET not being cleanly controlled, and others seem to have been from power supply issues.

I am glad that the 2N7002 will work, which makes life a little easier instead of making the board with through-hole components. I think next will still be to get both of the MOSFETs together onto an SOT23-6 package so it can limit some of the board space as well - if anyone can solder the PIC18 or ATMEGA168 to the board, then a SOT23-6 shouldn't be any worse.

My next move is probably going to be to see if I can design both the Arduino board with the same fix as the PIC board and give hobbyists an option between the two. To me, it seems like either processor works well and should have no bearing on the final product - but for experimentation sake and builders' preference I am hoping to update both.

I enjoy bringing this project back to life, considering Metcal and Hakko stations are STILL outside "hobbyist" price range... and all things willing if I can get enough bugs sorted out I would love to see this become a "finalized" hardware spec with the software being community-fixed via GitHub (which I will be setting up this week as well).

[NavyBOFH]

My next move is probably going to be to see if I can design both the Arduino board with the same fix as the PIC board and give hobbyists an option between the two. To me, it seems like either processor works well and should have no bearing on the final product - but for experimentation sake and builders' preference I am hoping to update both.

In my personal opinion this seems to be pretty useless ,because i think the rf-stage has more priority for now.
For example :
You can check what you can tweak or optimize on the rf-stage to make it less hot,you can optimize swr parameters etc, etc, etc ..... so basically leave for now the work on mcu's to later

Just my personal opinion .

I'd love any input on the RF stage. I plan on checking output power and SWR at some stage but it seems from previous posts that the system was decently SWR matched when it was designed. The IRF510 seems to run hot too but I can't find anything in that range that works any better without going up to another design.


Sent from my iPhone using Tapatalk

[NavyBOFH]

Further update: some messages and suggestions are that the IRF510 is not sufficient to drive the RF stage of the project and have been seeing either excessive heat and/or failures. I had to spend some time digging around to see that Metcal themselves (and not sure how Hakko or other "clones" engineered theirs) used an IRF640N for the driver and the IRF530 as the power amplifier.

The first suggestion I was given was to change the IRF510 to a IRF530 which can seemingly work and is rated for double the thermal dissipation the IRF510 offers. This sounds like the most interesting experiment to me since the design of the two is similar with slight changes in the characteristics of the MOSFET's input resistance and capacitance which would have to be double-checked through the actual MOSFET driver and circuit. Since this is where my experience is lacking, I would hope someone on here would know better how to go through this process versus a "trial and error" method I would end up going through. However, it is compelling to try since the IRF510's thermal dissipation is 43w while the 530's is 88w - that alone should lend to a cooler running and more efficient circuit - I would much rather run a MOSFET at 75% of its rated output than another at 100-105% output.

The other option was an IRF540 but that seems like too big of a step up and a complete re-design of the RF stage. It's thermal dissipation and power handling is up at 120w but has radically different characteristics to the FET itself.

I would love input on this... since my option at this point is to produce some boards with the over voltage fix and see what improvement can be made through trial and error which is hard to do on a hobbyist budget lol.

[richard.cs]

Datasheets for convinience:

IRF510: https://www.vishay.com/docs/91015/sihf510.pdf

IRF530: https://www.vishay.com/docs/91019/91019.pdf

I haven't looked properly but I notice that the IRF530 has 3x more gate charge and all the switching parameters are slower. This is expected given you get the improved thermal performance by having a bigger die. It may well be a suitable MOSFET but I wouldn't just throw it in without re-examining the gate drive and testing on a prototype. If you unintentionally slow the switching then you will increase dissipation and it may very well run hotter despite the factor of two in R_thJC. Essentially the real figure to consider for thermal performance is the change in sum R_thJC+R_thinterface+R_thheatsink which is around a change of 6.5 C/W to 5 C/W with the 100x100mm heatsink and a greased mica insulator. Don't pay too much attention to the headline power figures in the datasheet, they're not really representative of real-world use (as in you can't usually get anywhere close).

I suspect the MOSFET-killing on switch-off relates unintended linear operation dissipating hundreds of Watts for milliseconds and taking it outside it's safe operating area. That should be fixed by the undervoltage lockout circuit.

I personally wouldn't change it, mine seems pretty robust and thermal performance seems sufficient with the 100x100mm heatsink everyone else has used. During testing I had it running continuously into a 50 Ohm load (equivalent of a cold tip that never warms up) for about half an hour and the MOSFET was only warm - maybe 40-50C.

If you do want to change it the datasheets offer hints but as they're not really specified for use as an RF power amplifier you will probably have to do it by experiment. I would stay away from the IRF540, as a minimum you would need to seriously beef up the gate drive (9x more charge needed, so 9x the current to do it in the same time), but the rise and fall times might still catch you out. Paralleled smaller MOSFETs might be simpler.

[NavyBOFH]

Datasheets for convinience:

IRF510: https://www.vishay.com/docs/91015/sihf510.pdf

IRF530: https://www.vishay.com/docs/91019/91019.pdf

I haven't looked properly but I notice that the IRF530 has 3x more gate charge and all the switching parameters are slower. This is expected given you get the improved thermal performance by having a bigger die. It may well be a suitable MOSFET but I wouldn't just throw it in without re-examining the gate drive and testing on a prototype. If you unintentionally slow the switching then you will increase dissipation and it may very well run hotter despite the factor of two in R_thJC. Essentially the real figure to consider for thermal performance is the change in sum R_thJC+R_thinterface+R_thheatsink which is around a change of 6.5 C/W to 5 C/W with the 100x100mm heatsink and a greased mica insulator. Don't pay too much attention to the headline power figures in the datasheet, they're not really representative of real-world use (as in you can't usually get anywhere close).

I suspect the MOSFET-killing on switch-off relates unintended linear operation dissipating hundreds of Watts for milliseconds and taking it outside it's safe operating area. That should be fixed by the undervoltage lockout circuit.

I personally wouldn't change it, mine seems pretty robust and thermal performance seems sufficient with the 100x100mm heatsink everyone else has used. During testing I had it running continuously into a 50 Ohm load (equivalent of a cold tip that never warms up) for about half an hour and the MOSFET was only warm - maybe 40-50C.

If you do want to change it the datasheets offer hints but as they're not really specified for use as an RF power amplifier you will probably have to do it by experiment. I would stay away from the IRF540, as a minimum you would need to seriously beef up the gate drive (9x more charge needed, so 9x the current to do it in the same time), but the rise and fall times might still catch you out. Paralleled smaller MOSFETs might be simpler.

You essentially read my mind on the issue. I think your protection circuit would fix most of the issues people were complaining about. At that point it would also be down to whoever used the Arduino vs PIC boards since I have a feeling there might be differences in how the controller handled heating and ramping down. My thought was the same as yours that at that point if there's still issues with the MOSFETs that having an IRF510 and an IRF9510 in push-pull configuration would probably be the way to handle that issue. I'm more inclined to test boards with your circuit and see if/when a failure occurs. I am pretty sure with my line of work if I passed around the unit to several of my coworkers they'll kill it in short time if given the opportunity for a failure to occur.

As for the 100x100 heatsink - part of my thought is (once validating everything in the two-board configuration), was to see if I could get everything stuffed into a 100x150 board or even smaller, with the ability to then add your input protection circuit and feed the entire setup off an external DC brick instead of stuffing a transformer inside the unit - like you ended up doing.

Have you continued working with your unit? It would be nice to have a long-term review of the setup since it seems most other people that made one have disappeared!

[richard.cs]

All I can really say is that I use mine fairly regulary and I haven't blown it up yet. I haven't looked at the mosfet failure in detail and it's not happened to so I'm really just speculating about the cause. I have even misused it as a 13 MHz source for immunity testing with no harm done.

The only bug it has is sometimes the power metering seems off, reporting that it's finished heating before it really has, etc. It may be a software bug but probably I should find time to tweak the thresholds which Chris has made configurable from the UI before I complain about it.

What I haven't done is tried a wide selection of tips.

[NavyBOFH]

All I can really say is that I use mine fairly regulary and I haven't blown it up yet. I haven't looked at the mosfet failure in detail and it's not happened to so I'm really just speculating about the cause. I have even misused it as a 13 MHz source for immunity testing with no harm done.

The only bug it has is sometimes the power metering seems off, reporting that it's finished heating before it really has, etc. It may be a software bug but probably I should find time to tweak the thresholds which Chris has made configurable from the UI before I complain about it.

What I haven't done is tried a wide selection of tips.

Luckily my workplace has a decent assortment of tips and a spare handle that I can use - though I'm tempted to buy a Hakko handle and stand to try out, and play with an assortment of tips to get the "best value" for the controller.

I'm glad yours has been standing up to the use! That to me says your over voltage fix has been the key part that was missing and these boards I eventually get made should be the "version 1.0" boards finally!

I sent you a PM regarding some component values on the controller board, once I get all that filled in I am going to try a prettier version of the over voltage circuit on the board, post the files/gerbers/BOM, and see if we can get a working setup!

[NavyBOFH]

Boards have been ordered! I ordered 10 sets which I will build one, keep one as a spare, and might have a friend interested in a set - which leaves 7-8 sets up for grabs when they come in! Anyone that is interested speak up!

I am thinking of doing what others did here and order the enamel wire and toroids for the inductors from the "preferred" source and sell the boards as "kits" with those hard-to-find pieces together. I don't know how much the toroids will cost, but right now I am assuming the boards plus toroids won't cost more than $10-15 (my cost), plus whatever shipping costs to you, which I will try my hardest to make it an inexpensive option!

I am also ordering up all the bits and pieces for my board and updating the BOM as I go - while all the parts are still the same, the example part numbers are to parts that have been marked obsolete so I am trying to find better options for them all. Right now DigiKey has all but 1 part in stock which I am sourcing a replacement.