Homebrew PCBs

I have only done three or four toner-transfer boards, with acceptable results. But recently I acquired a laminator, and I picked up an inexpensive hotplate, so I did some experimenting to test my techniques. Here are some results; if you read all the way to the end, I have a couple of questions. :)

Test results showing patterns for 10/10, 8/8, 6/6, and 5/5 (left to right):

http://home.earthlink.net/~a_wake/tt_test_closeup.jpg

The patterns have two inner blocks and traces of the designated size and spacing spiraling outward to two outer blocks. If the inner block and the matching outer block show continuity, then the trace is intact; if there is continuity between two inner or two outer blocks, then there is a short. All of these test patterns worked except the 8/8 -- there is a break in one of the traces, but not the other. No shorts anywhere. Surprisingly (to me, anyway!) the 5/5 pattern worked perfectly. I'm not sure I'd be ready to trust it for a real project, though! :)

Here is what I used to produce the patterns.

HP LaserJet P2015dn printer with HP toner

HP Presentation Paper, Laser Glossy, 34lb. I can't find the part number anywhere ...

GBC 2130 laminator. This has a built-in temperature control that will go up to 320 degrees -- which is where I set it to do this test. I ran it through on speed 2 (next to slowest) three times. Here is a picture of the laminator:

http://home.earthlink.net/~a_wake/laminator.jpg

*** very important: I peeled the paper off the board immediately after removing from the laminator! In the past, I have found that it is nearly impossible to get this paper off if you let it cool and soak it, but it seems to come right off, leaving the toner behind, if you peel it off immediately.

I etched in 2 parts drug-store variety hydrogen peroxide plus 1 part muriatic acid. Usually this is super fast, but today it took more than 10 minutes ... maybe because the chemicals were really cold, and I etched it outside in the cold.

After cleaning the board with acetone to remove the remaining toner, I put it on a small hotplate -- a 110v single burner with an 8" x 8" piece of 1/4" thick aluminum on top. I fluxed the board and then heated it up until I could tin it with some solder. I wiped the excess solder off while still hot.

Now for some questions: 1) The PC board turned a chocolate brown color after heating on the hot plate. If you look at the second picture above, you'll see the now-brown board at the bottom of the picture next to a light-green board that came from the same stock. IOW, the light green is the color it was before heating on the hot plate. Is this color change normal? Ominous?

2) Does anyone else use the glossy laser paper as I've described above, peeling it off immediately after laminating?

3) Any suggestions about any of the technique described above?

Many thanks to anyone who has actually read this lengthy post all the way to the end!

Andy

On 05/02/2010 22:25, awakephd wrote:
> I have only done three or four toner-transfer boards, with acceptable results. But recently I acquired a laminator, and I picked up an inexpensive hotplate, so I did some experimenting to test my techniques. Here are some results; if you read all the way to the end, I have a couple of questions. :)
>
> Test results showing patterns for 10/10, 8/8, 6/6, and 5/5 (left to right):
>
> http://home.earthlink.net/~a_wake/tt_test_closeup.jpg
>
> The patterns have two inner blocks and traces of the designated size and spacing spiraling outward to two outer blocks. If the inner block and the matching outer block show continuity, then the trace is intact; if there is continuity between two inner or two outer blocks, then there is a short. All of these test patterns worked except the 8/8 -- there is a break in one of the traces, but not the other. No shorts anywhere. Surprisingly (to me, anyway!) the 5/5 pattern worked perfectly. I'm not sure I'd be ready to trust it for a real project, though! :)
>
> Here is what I used to produce the patterns.
>
> HP LaserJet P2015dn printer with HP toner
>
> HP Presentation Paper, Laser Glossy, 34lb. I can't find the part number anywhere ...
>
> GBC 2130 laminator. This has a built-in temperature control that will go up to 320 degrees -- which is where I set it to do this test. I ran it through on speed 2 (next to slowest) three times. Here is a picture of the laminator:
>
> http://home.earthlink.net/~a_wake/laminator.jpg
>
> *** very important: I peeled the paper off the board immediately after removing from the laminator! In the past, I have found that it is nearly impossible to get this paper off if you let it cool and soak it, but it seems to come right off, leaving the toner behind, if you peel it off immediately.
>
> I etched in 2 parts drug-store variety hydrogen peroxide plus 1 part muriatic acid. Usually this is super fast, but today it took more than 10 minutes ... maybe because the chemicals were really cold, and I etched it outside in the cold.
>
> After cleaning the board with acetone to remove the remaining toner, I put it on a small hotplate -- a 110v single burner with an 8" x 8" piece of 1/4" thick aluminum on top. I fluxed the board and then heated it up until I could tin it with some solder. I wiped the excess solder off while still hot.
>
> Now for some questions: 1) The PC board turned a chocolate brown color after heating on the hot plate. If you look at the second picture above, you'll see the now-brown board at the bottom of the picture next to a light-green board that came from the same stock. IOW, the light green is the color it was before heating on the hot plate. Is this color change normal? Ominous?
>
> 2) Does anyone else use the glossy laser paper as I've described above, peeling it off immediately after laminating?
>
> 3) Any suggestions about any of the technique described above?
>
> Many thanks to anyone who has actually read this lengthy post all the way to the end!

You must have overheated the board for it to turn brown.

Leon

"awakephd" <a_wake@...> writes:

> The patterns have two inner blocks and traces of the designated size
> and spacing spiraling outward to two outer blocks.

I have bigger patterns like this here: http://www.delorie.com/pcb/spirals/

> Now for some questions: 1) The PC board turned a chocolate brown
> color after heating on the hot plate.

Hot plates keep getting hotter and hotter.  I found that I *can't* use
the hotplate for a second board unless I let it cool completely first,
else the board burns (charred epoxy!).

"awakephd" <a_wake@...> writes:
> http://home.earthlink.net/~a_wake/tt_test_closeup.jpg

A hint for photographing very close up: Use a sheet of plain white
paper to diffuse the flash.  Hold it above the object such that one
edge of the paper is against the camera, between the flash and the
lens.  That way, the flash illuminates the paper, which illuminates
the object from above.

--- In Homebrew_PCBs@yahoogroups.com, DJ Delorie <dj@...> wrote:
>
> 
> "awakephd" <a_wake@...> writes:
> > http://home.earthlink.net/~a_wake/tt_test_closeup.jpg
> 
> A hint for photographing very close up: Use a sheet of plain white
> paper to diffuse the flash.  Hold it above the object such that one
> edge of the paper is against the camera, between the flash and the
> lens.  That way, the flash illuminates the paper, which illuminates
> the object from above.
>

Excellent hint, thanks!

> 
> Hot plates keep getting hotter and hotter.  I found that I *can't* use
> the hotplate for a second board unless I let it cool completely first,
> else the board burns (charred epoxy!).
>


You could attach a thermistor to the hot plate and build a simple temperature control circuit to keep the temperature FAR more stable than the controls, if any, that are built in. 

I've never used a hot plate anyway though, I just transfer, soak, clean, etch, optionally tin with the liquid stuff, then solder the parts.

"James" <jamesrsweet@...> writes:
> You could attach a thermistor to the hot plate and build a simple
> temperature control circuit to keep the temperature FAR more stable
> than the controls, if any, that are built in.

I've thought of that, but mine is cast iron - it has way too much
thermal mass to be accurately controlled *at all*.  It takes about
three minutes to get up to temperature, and the ramp-up is pretty good
for reflow soldering, so as long as I don't mess with it it works for
one board at a time, or multiple boards simultaneously, just not for
multiple boards consecutively.

If I'm going to start hacking up an mcu-controlled reflow device, I
won't start with a cast iron hotplate.

--- In Homebrew_PCBs@yahoogroups.com, DJ Delorie <dj@...> wrote:
>
> 
> "James" <jamesrsweet@...> writes:
> > You could attach a thermistor to the hot plate and build a simple
> > temperature control circuit to keep the temperature FAR more stable
> > than the controls, if any, that are built in.
> 
> I've thought of that, but mine is cast iron - it has way too much
> thermal mass to be accurately controlled *at all*.  It takes about
> three minutes to get up to temperature, and the ramp-up is pretty good
> for reflow soldering, so as long as I don't mess with it it works for
> one board at a time, or multiple boards simultaneously, just not for
> multiple boards consecutively.
> 
> If I'm going to start hacking up an mcu-controlled reflow device, I
> won't start with a cast iron hotplate.
>

In my case, it is *literally* a hot plate -- a piece of aluminum sitting on a portable one-eye electric burner. I am wanting to move towards surface mount components and reflow, so that is why I thought I'd do some experimenting to see what I could do with something along these lines. I was looking for an electric skillet at all the thrift stores near here ... apparently they sell quickly when they get one. Finally one store had this electric burner, so I decided to give it a try. Clearly, I need to do some more experimenting with the temperature settings ...

Meanwhile, I don't have any of the liquid tin stuff, nor am I in a position to get any anytime soon, so I thought I'd see if I could manually tin a board quickly this way. I fluxed the board, heated it, dabbed on some solder, and brushed it out. Worked great ... except for the small problem of over cooking the board!

On 06/02/2010, at 3:17 PM, awakephd wrote:
> In my case, it is *literally* a hot plate -- a piece of aluminum sitting on a portable one-eye electric burner. I am wanting to move towards surface mount components and reflow, so that is why I thought I'd do some experimenting to see what I could do with something along these lines. I was looking for an electric skillet at all the thrift stores near here ... apparently they sell quickly when they get one. Finally one store had this electric burner, so I decided to give it a try. Clearly, I need to do some more experimenting with the temperature settings ...
> 
> Meanwhile, I don't have any of the liquid tin stuff, nor am I in a position to get any anytime soon, so I thought I'd see if I could manually tin a board quickly this way. I fluxed the board, heated it, dabbed on some solder, and brushed it out. Worked great ... except for the small problem of over cooking the board!

I haven't tried the tinning part, but I've soldered SMD parts using an alu frypan over a gas flame - no board discoloration, take it off when all the joints look like they've gone shiny.

PG

So what? Just because it has mass it can't be controlled?

It'll work just fine even with a simple on/off controller. Put an
aluminium plate on top of it to spread the heat, the cast iron plate
is unevenly heated by the coil. Drill a hole inside the aluminium for
the thermocouple.

Of course the temperature will overshoot that way with a simple on/off
controller, but after the initial overshoot it holds pretty steady.
The mass is high enough that adding a board won't cool it a bit, so no
quick control is necessary.


If the boards get brown you overheat them, I agree. However If you
order some boards without soldermask and leadfree immersion tin from
PCB Pool you will also find them slightly browned towards the edges,
but much less than this one. Dipping them in tin just does that and
leadfree has made it worse. I guess it should be OK, PCB Pool aren't
amateurs and the boards have worked fine so far. With soldermasked
boards you just don't see the color any more.

ST

On Sat, Feb 6, 2010 at 4:41 AM, DJ Delorie <dj@...> wrote:
>
> "James" <jamesrsweet@...> writes:
>> You could attach a thermistor to the hot plate and build a simple
>> temperature control circuit to keep the temperature FAR more stable
>> than the controls, if any, that are built in.
>
> I've thought of that, but mine is cast iron - it has way too much
> thermal mass to be accurately controlled *at all*.  It takes about
> three minutes to get up to temperature, and the ramp-up is pretty good
> for reflow soldering, so as long as I don't mess with it it works for
> one board at a time, or multiple boards simultaneously, just not for
> multiple boards consecutively.
>
> If I'm going to start hacking up an mcu-controlled reflow device, I
> won't start with a cast iron hotplate.
>
>

Is this the HP paper you are using?

HP color Laser Presentation Paper, Glossy  34#

HP Color Laser Presentation Paper
<http://www.officedepot.com/a/products/136435/HP-Color-Laser-Presentatio\
n-Paper-Glossy/>

Craig







[Non-text portions of this message have been removed]

Stefan Trethan <stefan_trethan@...> writes:
> So what? Just because it has mass it can't be controlled?

The loop is too big to be useful that way, esp when the initial
ramp-up works just fine for reflow anyway.  It's like a three MINUTE
response curve.  I'd rather invest my energy into something more
responsive and leave this one alone.

That's the one I'm using. When I first tried it, I thought I had made a big mistake in buying it -- when I let it cool and then tried to soak it off. It never soaked off!!

But now that it seems to be working pretty well just peeling it off immediately after laminating ... wow, that saves a lot of time!

I'd be very interested if this works for anyone else!

--- In Homebrew_PCBs@yahoogroups.com, "designer_craig" <cs6061@...> wrote:

>
> 
> 
> Is this the HP paper you are using?
> 
> HP color Laser Presentation Paper, Glossy  34#
> 
> HP Color Laser Presentation Paper
> <http://www.officedepot.com/a/products/136435/HP-Color-Laser-Presentatio\
> n-Paper-Glossy/>
> 
> Craig
> 
> 
> 
> 
> 
> 
> 
> [Non-text portions of this message have been removed]
>

Some followup --

I made a board last night, and etched, drilled, and tinned it this morning. The process was the same as described below, except I managed to regulate the temperature on the hot plate much better. Here is the new board:

http://home.earthlink.net/~a_wake/tt_test_board2.jpg

and here it is in comparison to the over-cooked board from yesterday:

http://home.earthlink.net/~a_wake/tt_test_comparison.jpg

There is still a little browning of the new board, but nothing like what happened to the one from yesterday. I took an old board and put it on the hot plate, and kept turning the heat up a little at a time until I saw the solder melt. When I put the new board on, I wound up still needing to turn the heat up a bit more -- basically, it was melting the solder only on the heating cycle. I wonder if it browned a bit in part because it took me so long to get it all tinned, because it wasn't quite hot enough.

Now that I know where the setting is, I may try the next time with a 3/4" thick plate of aluminum, with the idea that it should keep a pretty stable temperature across the heating cycle.

In any case, I am very pleased with the latest results. BTW, the fine traces on the board are 10 mil, and the lettering is using .004" thick strokes.

Thanks for all the input, and for all that you all have taught me to get me to this point!

Andy

--- In Homebrew_PCBs@yahoogroups.com, "awakephd" <a_wake@...> wrote:

>
> I have only done three or four toner-transfer boards, with acceptable results. But recently I acquired a laminator, and I picked up an inexpensive hotplate, so I did some experimenting to test my techniques. Here are some results; if you read all the way to the end, I have a couple of questions. :)
> 
> Test results showing patterns for 10/10, 8/8, 6/6, and 5/5 (left to right):
> 
> http://home.earthlink.net/~a_wake/tt_test_closeup.jpg
> 
> The patterns have two inner blocks and traces of the designated size and spacing spiraling outward to two outer blocks. If the inner block and the matching outer block show continuity, then the trace is intact; if there is continuity between two inner or two outer blocks, then there is a short. All of these test patterns worked except the 8/8 -- there is a break in one of the traces, but not the other. No shorts anywhere. Surprisingly (to me, anyway!) the 5/5 pattern worked perfectly. I'm not sure I'd be ready to trust it for a real project, though! :)
> 
> Here is what I used to produce the patterns.
> 
> HP LaserJet P2015dn printer with HP toner
> 
> HP Presentation Paper, Laser Glossy, 34lb. I can't find the part number anywhere ...
> 
> GBC 2130 laminator. This has a built-in temperature control that will go up to 320 degrees -- which is where I set it to do this test. I ran it through on speed 2 (next to slowest) three times. Here is a picture of the laminator:
> 
> http://home.earthlink.net/~a_wake/laminator.jpg
> 
> *** very important: I peeled the paper off the board immediately after removing from the laminator! In the past, I have found that it is nearly impossible to get this paper off if you let it cool and soak it, but it seems to come right off, leaving the toner behind, if you peel it off immediately.
> 
> I etched in 2 parts drug-store variety hydrogen peroxide plus 1 part muriatic acid. Usually this is super fast, but today it took more than 10 minutes ... maybe because the chemicals were really cold, and I etched it outside in the cold.
> 
> After cleaning the board with acetone to remove the remaining toner, I put it on a small hotplate -- a 110v single burner with an 8" x 8" piece of 1/4" thick aluminum on top. I fluxed the board and then heated it up until I could tin it with some solder. I wiped the excess solder off while still hot.
> 
> Now for some questions: 1) The PC board turned a chocolate brown color after heating on the hot plate. If you look at the second picture above, you'll see the now-brown board at the bottom of the picture next to a light-green board that came from the same stock. IOW, the light green is the color it was before heating on the hot plate. Is this color change normal? Ominous?
> 
> 2) Does anyone else use the glossy laser paper as I've described above, peeling it off immediately after laminating?
> 
> 3) Any suggestions about any of the technique described above?
> 
> Many thanks to anyone who has actually read this lengthy post all the way to the end!
> 
> Andy
>

What sort of solder are we talking? Paste, "normal" resin-core wire, or somethig else?

What are you brushing it (excess solder) off with? I have never satisfactorily tinned a whole board yet.

PG

On 07/02/2010, at 6:33 AM, awakephd wrote:

> Some followup --
> 
> I made a board last night, and etched, drilled, and tinned it this morning. The process was the same as described below, except I managed to regulate the temperature on the hot plate much better. Here is the new board:
> 
> http://home.earthlink.net/~a_wake/tt_test_board2.jpg
> 
> and here it is in comparison to the over-cooked board from yesterday:

Yep, that is the HP paper I'm using also.  The cover looks a bit different, but the Q2546A part number is the same.

I'll have to try peeling it off while hot, I've been soaking it off.  But of a hassle to soak off, but does eventually come off, just not anywhere near as easy as the Pulsar paper - that stuff just "floats" off in a few seconds.

Ken H>

--- In Homebrew_PCBs@yahoogroups.com, "awakephd" <a_wake@...> wrote:

>
> That's the one I'm using. When I first tried it, I thought I had made a big mistake in buying it -- when I let it cool and then tried to soak it off. It never soaked off!!
> 
> But now that it seems to be working pretty well just peeling it off immediately after laminating ... wow, that saves a lot of time!
> 
> I'd be very interested if this works for anyone else!
> 
> --- In Homebrew_PCBs@yahoogroups.com, "designer_craig" <cs6061@> wrote:
> >
> > 
> > 
> > Is this the HP paper you are using?
> > 
> > HP color Laser Presentation Paper, Glossy  34#
> > 
> > HP Color Laser Presentation Paper
> > <http://www.officedepot.com/a/products/136435/HP-Color-Laser-Presentatio\
> > n-Paper-Glossy/>
> > 
> > Craig
> > 
> > 
> > 
> > 
> > 
> > 
> > 
> > [Non-text portions of this message have been removed]
> >
>

Piers, I am using plain old rosin-core solder from Radio Shack -- nothing fancy. I'm using a small "flux brush" to brush on the flux and to brush the solder around -- I picked up a package of 20 or so on sale cheap at Harbor Freight. At the end I am also wiping it with some paper towel.

My technique is still developing, but on this board, I put the bare board on the hot plate; I brushed flux on the whole board (I have a pint container full of rosin flux that I picked up from a yard sale); I cut some snips of solder and scattered them on the board. When they start melting, I brush the molten solder over every bit of copper.

Undoubtedly the liquid tin stuff would be a whole lot faster and easier, but this does seem to work for me for the moment. Hopefully you will improve on my technique and let me know how! :)

Andy

--- In Homebrew_PCBs@yahoogroups.com, Piers Goodhew <piers@...> wrote:

>
> What sort of solder are we talking? Paste, "normal" resin-core wire, or somethig else?
> 
> What are you brushing it (excess solder) off with? I have never satisfactorily tinned a whole board yet.
> 
> PG
> 
> On 07/02/2010, at 6:33 AM, awakephd wrote:
> 
> > Some followup --
> > 
> > I made a board last night, and etched, drilled, and tinned it this morning. The process was the same as described below, except I managed to regulate the temperature on the hot plate much better. Here is the new board:
> > 
> > http://home.earthlink.net/~a_wake/tt_test_board2.jpg
> > 
> > and here it is in comparison to the over-cooked board from yesterday:
>

My aluminum top-plate is only 1/4" thick, works great.

Hi Guys
My technique on small one off boards is to use a used piece desoldering wick and push it over the tracks with a soldering iron

Craig

  ----- Original Message ----- 
  From: awakephd 
  To: Homebrew_PCBs@yahoogroups.com 
  Sent: Sunday, February 07, 2010 1:17 PM
  Subject: [Homebrew_PCBs] Re: TT results - HP 2015, HP Laser Glossy paper, GBC 2130, HCl + H2O2, hotplate


    
  Piers, I am using plain old rosin-core solder from Radio Shack -- nothing fancy. I'm using a small "flux brush" to brush on the flux and to brush the solder around -- I picked up a package of 20 or so on sale cheap at Harbor Freight. At the end I am also wiping it with some paper towel.

  My technique is still developing, but on this board, I put the bare board on the hot plate; I brushed flux on the whole board (I have a pint container full of rosin flux that I picked up from a yard sale); I cut some snips of solder and scattered them on the board. When they start melting, I brush the molten solder over every bit of copper.

  Undoubtedly the liquid tin stuff would be a whole lot faster and easier, but this does seem to work for me for the moment. Hopefully you will improve on my technique and let me know how! :)

  Andy

  --- In Homebrew_PCBs@yahoogroups.com, Piers Goodhew <piers@...> wrote:
  >
  > What sort of solder are we talking? Paste, "normal" resin-core wire, or somethig else?
  > 
  > What are you brushing it (excess solder) off with? I have never satisfactorily tinned a whole board yet.
  > 
  > PG
  > 
  > On 07/02/2010, at 6:33 AM, awakephd wrote:
  > 
  > > Some followup --
  > > 
  > > I made a board last night, and etched, drilled, and tinned it this morning. The process was the same as described below, except I managed to regulate the temperature on the hot plate much better. Here is the new board:
  > > 
  > > http://home.earthlink.net/~a_wake/tt_test_board2.jpg
  > > 
  > > and here it is in comparison to the over-cooked board from yesterday:
  >



  

  __________ Information from ESET NOD32 Antivirus, version of virus signature database 4842 (20100206) __________

  The message was checked by ESET NOD32 Antivirus.

  http://www.eset.com


[Non-text portions of this message have been removed]

Another option is to paint with solder paste (cheap plumbing variety
is sufficient) and reflow, then wash away the rest.

Or a fine brass brush might be used to thinly spread the solder (same
effect as wick).

The professional process is to dip it in molten solder quickly, and
then blow the excess off with hot air (hot air leveling). This doesn't
seem practical for homebrew use. Roller tinning is another option,
where the board is fed through two rollers one of which is copper and
rotating through a solder bath on it's lower portion. Could be built
but only makes sense for higher volume.

ST

On Sun, Feb 7, 2010 at 10:21 AM, Craig VK2ZCM <vk2zcm@...> wrote:
> Hi Guys
> My technique on small one off boards is to use a used piece desoldering wick and push it over the tracks with a soldering iron
>
> Craig
>

> 
> I've thought of that, but mine is cast iron - it has way too much
> thermal mass to be accurately controlled *at all*.  It takes about
> three minutes to get up to temperature, and the ramp-up is pretty good
> for reflow soldering, so as long as I don't mess with it it works for
> one board at a time, or multiple boards simultaneously, just not for
> multiple boards consecutively.
> 
> If I'm going to start hacking up an mcu-controlled reflow device, I
> won't start with a cast iron hotplate.
>


Thermal mass is no problem, it actually makes it easier, you just have to tune your temperature control to what you're controlling. With electronic control you aren't limited to long on/off cycles with loads of hysteresis, you can cycle the heater rapidly or even use phase angle control (think light dimmer) to vary the power to the heater. It can be done with a few op-amps and a handful of passives, or you can use the microcontroller of your choice.

My 400 gallon spa has a lot more thermal mass than your hot plate, and it has no trouble maintaining the temperature to within a fraction of a degree.

--- In Homebrew_PCBs@yahoogroups.com, DJ Delorie <dj@...> wrote:
>
> 
> Stefan Trethan <stefan_trethan@...> writes:
> > So what? Just because it has mass it can't be controlled?
> 
> The loop is too big to be useful that way, esp when the initial
> ramp-up works just fine for reflow anyway.  It's like a three MINUTE
> response curve.  I'd rather invest my energy into something more
> responsive and leave this one alone.
>


If you'd rather save the effort for something more to your liking that's perfectly ok, however there's really no particular hurdle to overcome with what you have, it doesn't matter how much thermal mass it has or how long it takes to heat and cool, the temperature can be very accurately controlled. Something more responsive will be hot and ready to go much quicker, but it's also harder to regulate the temperature. Less thermal mass means more prone to oscillation above and below the desired temperature, and it will be more greatly affected by external influence. It can still be done, but it's harder, you have to substitute the mechanical thermal mass with electrical damping.

To illustrate the point, think of why your car has shock absorbers. They are mechanical damping to slow down the response of the suspension, without them you would be bouncing all over the place on the springs. You might notice that when you have the added mass of several passengers and cargo, the car rides smoother as well, the response to variations in the road surface is slowed.

"James" <jamesrsweet@...> writes:

> it doesn't matter how much thermal mass it has or how long it takes
> to heat and cool,

It certainly does when you have to control the temperature *change
rate* as well as the hold temps.  Proper reflow wants a specific temp
vs time curve, not just a pre-heated surface.  The iron plate can just
barely heat up fast enough for reflow, and mcu can only slow it down.

If I start with a cold plate, put my board on it, and turn it on, I
get a pretty good temp/time curve - takes about 3 minutes for small
boards, or 6 minutes if I add an aluminum plate for large boards (heat
spreader).  Six minutes is a little long.  I'd rather have an aluminum
hotplate I can mcu-control that can do a proper temp/time curve.

If you put a board on a pre-heated plate, the flux boils and spatters
solder (and parts) all over the place.  You really need to heat the
flux up slowly so it activates and cleans the metal without disturbing
anything, then dries up before the solder flows.

I also don't want to shock the board by hitting it with that much heat
all at once.

Actually, it takes a little time for the heat to propagate through the
PCB. With a preheated plate I get very good results, no splattering or
such. You could always put a second plate right next to it for preheat
at a lower temperature if you really need to.

But If you want to drive a profile, of course I agree it's no use
having so much thermal mass.

ST

On Tue, Feb 9, 2010 at 5:40 AM, DJ Delorie <dj@...> wrote:

> It certainly does when you have to control the temperature *change
> rate* as well as the hold temps.  Proper reflow wants a specific temp
> vs time curve, not just a pre-heated surface.  The iron plate can just
> barely heat up fast enough for reflow, and mcu can only slow it down.
>
> If I start with a cold plate, put my board on it, and turn it on, I
> get a pretty good temp/time curve - takes about 3 minutes for small
> boards, or 6 minutes if I add an aluminum plate for large boards (heat
> spreader).  Six minutes is a little long.  I'd rather have an aluminum
> hotplate I can mcu-control that can do a proper temp/time curve.
>
> If you put a board on a pre-heated plate, the flux boils and spatters
> solder (and parts) all over the place.  You really need to heat the
> flux up slowly so it activates and cleans the metal without disturbing
> anything, then dries up before the solder flows.
>
> I also don't want to shock the board by hitting it with that much heat
> all at once.
>
>