Hi Julien,
We also use an Optimizer (like the one Alan Krepwick describes in his
response). We monitor dwell time and parallelism using sample average and
range charts for a sampling of 3. We also have a window on our carrier board
that we use to conduct SPC on the amount of flux deposited (by mass). All
this data is collected once per day (only one WS shift per day here). 

The levels of variation that we have seen very much resemble those Alan
mentions. The standard deviation for parallelism is just below 0.1s, much
like Alan's results, except that in an sampled average chart that results in
tighter bounds (+/- 0.2s instead of +/- 0.3s). Our st.dev for dwell time is
on a similar scale but slightly higher (0.125s). It is interesting that our
variations are so similar despite the fact that we are running the wave at
rather different heights. As Alan mentions, the Optimizer measures immersion
depth in increments of 12 mil. The maximum reading is 96 mil and we have our
process set up such that we view any non-maximum reading as a potential
issue. This is quite deep and is a result of our product mix. However, this
method does not allow me to share any accurate data on the variation we see
in immersion depth other than to say that we rarely see points below 96 mil.

Have you collected much data yet ? How can your process for parallelism have
a St.Dev of 0.2s but stay within the -0.2 to 0.4s range ? Assuming you are
not overestimating your actual standard deviation, it looks like you have
roughly twice as much variation (in st.dev terms) as Alan and I do. 

A suggestion on this last point. We do not use a marked section of the
fingers to run the optimizer. Although we note which fingers were used, we
do not control this. Hence the variation from the state of the fingers is
included in our numbers. A few years ago, concerned that there was too much
such variation, I switched the fingers to ones twice as thick as the
original ones (1/16). This, combined with a greater emphasis on finger
inspection and maintenance, resulted in the st.dev in our parallelism data
dropping over 30%. Since it appears as though your variation is higher than
Alan's or ours, perhaps this suggestion could help.

Hope this is of some use,
yves

Yves Dupuis
Process Analyst
Leitch Technology
Phone: 416-445-9929 X3389
Fax: 416-445-7927
25 Dyas Road,
North York, Ontario
M3B 1V7

> -----Original Message-----
> From: Julien Bouchard [SMTP:[log in to unmask]]
> Sent: Wednesday, July 11, 2001 11:20 AM
> To:   [log in to unmask]
> Subject:      [TN] WaveRider results on solder wave
> 
> Hi everybody,
>         We start using a waverider for monitoring two wave soldering
> machine. We
> have made some pretest run during one month. Our main concern is about the
> parallelism and the immersion depth of the solder. The parallelism is
> between -0.2 sec et 0.4 sec ( main wave ) and the immersion depth is
> between
> 1.47 mm and 1.93 mm. Our main question is about the variation we see not
> the
> absolute value. We have a standards deviation of 0.2 sec for the
> parallelism
> and 0.12 mm for the immersion depth. Since we have no real point of
> comparaison, we are unable to say if this it's ok or too much.  We want to
> know if other person use the waverider and which variation do you read
> from
> the immersion depth and the parallism of the wave solder. Thank you.
> 
> 
> Julien Bouchard ing. stag.
> Matrox - Spécialiste de procédés
> 
> --------------------------------------------------------------------------
> -------
> Technet Mail List provided as a free service by IPC using LISTSERV 1.8d
> To unsubscribe, send a message to [log in to unmask] with following text in
> the BODY (NOT the subject field): SIGNOFF Technet
> To temporarily halt delivery of Technet send the following message: SET
> Technet NOMAIL
> Search previous postings at: www.ipc.org > On-Line Resources & Databases >
> E-mail Archives
> Please visit IPC web site (http://www.ipc.org/html/forum.htm) for
> additional
> information, or contact Keach Sasamori at [log in to unmask] or 847-509-9700
> ext.5315
> --------------------------------------------------------------------------
> -------

---------------------------------------------------------------------------------
Technet Mail List provided as a free service by IPC using LISTSERV 1.8d
To unsubscribe, send a message to [log in to unmask] with following text in
the BODY (NOT the subject field): SIGNOFF Technet
To temporarily halt delivery of Technet send the following message: SET Technet NOMAIL
Search previous postings at: www.ipc.org > On-Line Resources & Databases > E-mail Archives
Please visit IPC web site (http://www.ipc.org/html/forum.htm) for additional
information, or contact Keach Sasamori at [log in to unmask] or 847-509-9700 ext.5315
---------------------------------------------------------------------------------