Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]

Groups > comp.lang.forth > #33413 > unrolled thread

Question to Albert

Back to article view | Back to comp.lang.forth

Contents

  Question to Albert rickman <gnuarm@gmail.com> - 2014-09-17 20:27 -0400
    Re: Question to Albert Paul E Bennett <Paul_E.Bennett@topmail.co.uk> - 2014-09-18 08:08 +0100
      Re: Question to Albert rickman <gnuarm@gmail.com> - 2014-09-18 03:55 -0400
      Re: Question to Albert albert@spenarnc.xs4all.nl (Albert van der Horst) - 2014-09-18 10:37 +0000
        Re: Question to Albert rickman <gnuarm@gmail.com> - 2014-09-18 16:30 -0400
        Re: Question to Albert Paul E Bennett <Paul_E.Bennett@topmail.co.uk> - 2014-09-19 01:11 +0100
      Re: Question to Albert Brad Eckert <hwfwguy@gmail.com> - 2014-09-27 17:10 -0700

#33413 — Question to Albert

I reading a bit of the blog by Yossi Kreinin and found the following 
comment by Albert van der Horst...

http://yosefk.com/blog/my-history-with-forth-stack-machines.html#comment-2536

"And the pinout! Only hobbyist would try those chips out. There are no 
pins,, just solderareas, with a stride of .4 mm (not mils, 400 
micrometer). Hobbyist trying those out would be probably my age (over 
sixty)."

The context is the GA144.  I have to agree with most of what he said, 
but I don't understand the portion quoted above.  I assume he is 
complaining about the package rather than the pin out.  In reality I 
think the package was one of the best choices they could have made.  In 
the real world size is a *huge* factor for ICs these days and this was a 
great tradeoff between size and usability.

While many devices are shipped in BGA sorts of packages (FPGAs and many 
CPUS) this device is in a no lead quad flat pack.  No, this is not so 
great for prototyping by hand, but prototyping by hand is virtually gone 
in the professional world.  Nearly every project uses a PCB on the first 
revision of the board and plan for two or even three spins of the board 
total.

I like the package because the dimensions are compatible with relaxed 
6/6 mil trace and space design rules.  This allows a lower board cost. 
Heck, I can get protos from oshpark.com for $10 a square inch with 5/5 
design rules.  (a great deal btw, 3 boards, 2 weeks @ $10/sq inch)

I hope I understand what Albert meant and am not just ranting about 
nothing.

-- 

Rick

[toc] | [next] | [standalone]

#33415

rickman wrote:

> I reading a bit of the blog by Yossi Kreinin and found the following
> comment by Albert van der Horst...
> 
> http://yosefk.com/blog/my-history-with-forth-stack-machines.html#comment-2536
> 
> "And the pinout! Only hobbyist would try those chips out. There are no
> pins,, just solderareas, with a stride of .4 mm (not mils, 400
> micrometer). Hobbyist trying those out would be probably my age (over
> sixty)."
> 
> The context is the GA144.  I have to agree with most of what he said,
> but I don't understand the portion quoted above.  I assume he is
> complaining about the package rather than the pin out.  In reality I
> think the package was one of the best choices they could have made.  In
> the real world size is a *huge* factor for ICs these days and this was a
> great tradeoff between size and usability.
> 
> While many devices are shipped in BGA sorts of packages (FPGAs and many
> CPUS) this device is in a no lead quad flat pack.  No, this is not so
> great for prototyping by hand, but prototyping by hand is virtually gone
> in the professional world.  Nearly every project uses a PCB on the first
> revision of the board and plan for two or even three spins of the board
> total.
> 
> I like the package because the dimensions are compatible with relaxed
> 6/6 mil trace and space design rules.  This allows a lower board cost.
> Heck, I can get protos from oshpark.com for $10 a square inch with 5/5
> design rules.  (a great deal btw, 3 boards, 2 weeks @ $10/sq inch)
> 
> I hope I understand what Albert meant and am not just ranting about
> nothing.
 
Like Albert, I am also over 60. However, I do not see what the problem is 
with the GA144 even for hand soldered prototype. It probably takes a 
properly laid out PCB as your basis but the soldering in task is quite easy 
to accomplish. My most recent experience with this was on a tiny IC package 
(20 pins) with similarly close pin-spacing.  The technique is to flood 
solder across all pins then use solder-wick to pull the excess away. I find 
a decent angle-poise magnifier is handy when doing this sort of thing and 
access to a microscope for final inspections.


-- 
********************************************************************
Paul E. Bennett IEng MIET.....<email://Paul_E.Bennett@topmail.co.uk>
Forth based HIDECS Consultancy.............<http://www.hidecs.co.uk>
Mob: +44 (0)7811-639972
Tel: +44 (0)1235-510979
Going Forth Safely ..... EBA. www.electric-boat-association.org.uk..
********************************************************************

[toc] | [prev] | [next] | [standalone]

#33416

On 9/18/2014 3:08 AM, Paul E Bennett wrote:
> rickman wrote:
>
>> I reading a bit of the blog by Yossi Kreinin and found the following
>> comment by Albert van der Horst...
>>
>> http://yosefk.com/blog/my-history-with-forth-stack-machines.html#comment-2536
>>
>> "And the pinout! Only hobbyist would try those chips out. There are no
>> pins,, just solderareas, with a stride of .4 mm (not mils, 400
>> micrometer). Hobbyist trying those out would be probably my age (over
>> sixty)."
>>
>> The context is the GA144.  I have to agree with most of what he said,
>> but I don't understand the portion quoted above.  I assume he is
>> complaining about the package rather than the pin out.  In reality I
>> think the package was one of the best choices they could have made.  In
>> the real world size is a *huge* factor for ICs these days and this was a
>> great tradeoff between size and usability.
>>
>> While many devices are shipped in BGA sorts of packages (FPGAs and many
>> CPUS) this device is in a no lead quad flat pack.  No, this is not so
>> great for prototyping by hand, but prototyping by hand is virtually gone
>> in the professional world.  Nearly every project uses a PCB on the first
>> revision of the board and plan for two or even three spins of the board
>> total.
>>
>> I like the package because the dimensions are compatible with relaxed
>> 6/6 mil trace and space design rules.  This allows a lower board cost.
>> Heck, I can get protos from oshpark.com for $10 a square inch with 5/5
>> design rules.  (a great deal btw, 3 boards, 2 weeks @ $10/sq inch)
>>
>> I hope I understand what Albert meant and am not just ranting about
>> nothing.
>
> Like Albert, I am also over 60. However, I do not see what the problem is
> with the GA144 even for hand soldered prototype. It probably takes a
> properly laid out PCB as your basis but the soldering in task is quite easy
> to accomplish. My most recent experience with this was on a tiny IC package
> (20 pins) with similarly close pin-spacing.  The technique is to flood
> solder across all pins then use solder-wick to pull the excess away. I find
> a decent angle-poise magnifier is handy when doing this sort of thing and
> access to a microscope for final inspections.

Yes, it shouldn't be that bad.  I've never used this technique as my 
time for soldering has passed.  My hand is *very* unsteady these days. 
I bought three of the GA144 from Schmartboards with the, not so well 
thought out, adapter boards but had them solder at least one of them for 
me, maybe it was two... lol  They only charge a couple of bucks extra.

But then their adapters are special.  They have channels in the top of 
the PCB with the traces in the groove and solder already pooled in them. 
  I think it is intended for leaded parts, but they seem to think it 
works well for QFN parts.  I think the idea is the raised PCB material 
acts as a better separator than solder mask and the bridges don't form 
so easily.

-- 

Rick

[toc] | [prev] | [next] | [standalone]

#33418

In article <c7venqFigupU1@mid.individual.net>,
Paul E Bennett  <Paul_E.Bennett@topmail.co.uk> wrote:
>rickman wrote:
>
>> I reading a bit of the blog by Yossi Kreinin and found the following
>> comment by Albert van der Horst...
>>
>> http://yosefk.com/blog/my-history-with-forth-stack-machines.html#comment-2536
>>
>> "And the pinout! Only hobbyist would try those chips out. There are no
>> pins,, just solderareas, with a stride of .4 mm (not mils, 400
>> micrometer). Hobbyist trying those out would be probably my age (over
>> sixty)."
>>
>> The context is the GA144.  I have to agree with most of what he said,
>> but I don't understand the portion quoted above.  I assume he is
>> complaining about the package rather than the pin out.  In reality I
>> think the package was one of the best choices they could have made.  In
>> the real world size is a *huge* factor for ICs these days and this was a
>> great tradeoff between size and usability.
>>
>> While many devices are shipped in BGA sorts of packages (FPGAs and many
>> CPUS) this device is in a no lead quad flat pack.  No, this is not so
>> great for prototyping by hand, but prototyping by hand is virtually gone
>> in the professional world.  Nearly every project uses a PCB on the first
>> revision of the board and plan for two or even three spins of the board
>> total.
>>
>> I like the package because the dimensions are compatible with relaxed
>> 6/6 mil trace and space design rules.  This allows a lower board cost.
>> Heck, I can get protos from oshpark.com for $10 a square inch with 5/5
>> design rules.  (a great deal btw, 3 boards, 2 weeks @ $10/sq inch)
>>
>> I hope I understand what Albert meant and am not just ranting about
>> nothing.
>
>Like Albert, I am also over 60. However, I do not see what the problem is
>with the GA144 even for hand soldered prototype. It probably takes a
>properly laid out PCB as your basis but the soldering in task is quite easy
>to accomplish. My most recent experience with this was on a tiny IC package
>(20 pins) with similarly close pin-spacing.  The technique is to flood
>solder across all pins then use solder-wick to pull the excess away. I find
>a decent angle-poise magnifier is handy when doing this sort of thing and
>access to a microscope for final inspections.

Good grieve. You're one of the most accomplished experts in the field
of embedded programming, and you do it for a living. You can't expect
that what you find easy, is easy for others. I've the solder-it-yourself-
prototype board for GA144, and I dare not touch it, because I made
such a mess of the example boards of the same type with twice the
pin distance.

By the way, I don't own a microscope.

I can't agree with rickman. There is so much experimenting needed to
get anything going with those boards that designing a circuit board
around each experiment is out of the question. Moreover I for me would
not manage to design such a board.

"Prototyping by hand is virtually gone in the professional world"
My argument was that only Forth enthusiast, mostly hobbyists, try
those chips out.

>
>
>--
>Paul E. Bennett IEng MIET.....<email://Paul_E.Bennett@topmail.co.uk>


Groetjes Albert
-- 
Albert van der Horst, UTRECHT,THE NETHERLANDS
Economic growth -- being exponential -- ultimately falters.
albert@spe&ar&c.xs4all.nl &=n http://home.hccnet.nl/a.w.m.van.der.horst

[toc] | [prev] | [next] | [standalone]

#33423

On 9/18/2014 6:37 AM, Albert van der Horst wrote:
> In article <c7venqFigupU1@mid.individual.net>,
> Paul E Bennett  <Paul_E.Bennett@topmail.co.uk> wrote:
>>
>> Like Albert, I am also over 60. However, I do not see what the problem is
>> with the GA144 even for hand soldered prototype. It probably takes a
>> properly laid out PCB as your basis but the soldering in task is quite easy
>> to accomplish. My most recent experience with this was on a tiny IC package
>> (20 pins) with similarly close pin-spacing.  The technique is to flood
>> solder across all pins then use solder-wick to pull the excess away. I find
>> a decent angle-poise magnifier is handy when doing this sort of thing and
>> access to a microscope for final inspections.
>
> Good grieve. You're one of the most accomplished experts in the field
> of embedded programming, and you do it for a living. You can't expect
> that what you find easy, is easy for others. I've the solder-it-yourself-
> prototype board for GA144, and I dare not touch it, because I made
> such a mess of the example boards of the same type with twice the
> pin distance.
>
> By the way, I don't own a microscope.
>
> I can't agree with rickman. There is so much experimenting needed to
> get anything going with those boards that designing a circuit board
> around each experiment is out of the question. Moreover I for me would
> not manage to design such a board.
>
> "Prototyping by hand is virtually gone in the professional world"
> My argument was that only Forth enthusiast, mostly hobbyists, try
> those chips out.

Ah, that was what I misunderstood.  I thought you were implying that the 
package meant only the hobbyists would use the chip which of course 
makes no sense.  I understand now, but I don't think that is the point.

The chip was not designed with hobbyist in mind.  I think the push into 
the hobbyist market via Schmart Boards is an afterthought to make it a 
bit easier for the curious to try out the GA144.  The part was designed 
for commercial apps even though you would not think that by looking at 
the features on the die itself.  However *for commercial* uses, the 
package is a pretty good compromise (nearly everything in engineering is 
compromise) between size, cost and ease of use.

-- 

Rick

[toc] | [prev] | [next] | [standalone]

#33427

Albert van der Horst wrote:

[%X]

> Good grieve. You're one of the most accomplished experts in the field
> of embedded programming, and you do it for a living. You can't expect
> that what you find easy, is easy for others. I've the solder-it-yourself-
> prototype board for GA144, and I dare not touch it, because I made
> such a mess of the example boards of the same type with twice the
> pin distance.

Thanks for that accolade Albert. Even though you think I am an accomplished 
professional, and I certainly do this stuff for a living and have been in 
practice for over 40 years, I do not get as much time at the solder bench 
these days as I used to. There was a time when the soldering iron used to be 
hot all day. I am just in the habit of doing the first few prototypes just 
to check things out then letting the companies that have the robotic 
assembly line handle the final quantity at high quality.
 
> By the way, I don't own a microscope.

The Angle-poise I have is an LED lit one which can be found for as little as 
£25.00 and the microscope is a Maplin job (400x magnification) for about 
£39.00 IIRC.
 
> I can't agree with rickman. There is so much experimenting needed to
> get anything going with those boards that designing a circuit board
> around each experiment is out of the question. Moreover I for me would
> not manage to design such a board.
> 
> "Prototyping by hand is virtually gone in the professional world"
> My argument was that only Forth enthusiast, mostly hobbyists, try
> those chips out.

Depends which part of the professional world, but generally that is the case 
these days. It is often a value judgement about whether you need to use high 
level engineers as prototype board assemblers or let them carry on with the 
higher level engineering tasks instead.

-- 
********************************************************************
Paul E. Bennett IEng MIET.....<email://Paul_E.Bennett@topmail.co.uk>
Forth based HIDECS Consultancy.............<http://www.hidecs.co.uk>
Mob: +44 (0)7811-639972
Tel: +44 (0)1235-510979
Going Forth Safely ..... EBA. www.electric-boat-association.org.uk..
********************************************************************

[toc] | [prev] | [next] | [standalone]

#33572

On Thursday, September 18, 2014 3:08:45 AM UTC-4, Paul E Bennett wrote:
> (20 pins) with similarly close pin-spacing.  The technique is to flood 
> solder across all pins then use solder-wick to pull the excess away. I find  
> a decent angle-poise magnifier is handy when doing this sort of thing and 
> access to a microscope for final inspections.
> 
Nowadays you can get steel stencils for not too much, and mylar or polyimide stencils for less. Check out videos on sparkfun to see how to apply solder paste. Reflow can be done on a hotplate or in a toaster oven. No controller needed if you can watch the solder reflow and stop the process at that point.

[toc] | [prev] | [standalone]

Back to top | Article view | comp.lang.forth

csiph-web