Continue to Site

Eng-Tips is the largest engineering community on the Internet

Intelligent Work Forums for Engineering Professionals

  • Congratulations waross on being selected by the Eng-Tips community for having the most helpful posts in the forums last week. Way to Go!

Ramp and Soaked

Status
Not open for further replies.

itsmoked

Electrical
Feb 18, 2005
19,114
So as you may have read elsewhere my reflow oven controller spazzed-out and fried some fully loaded circuit boards I was processing. I decided to get an OTShelf Ramp and Soak controller to replace it. I found a nice inexpensive one. I fabricated a whole enclosure around it and switched to SSR control and cool down assist. I hooked it up via 485 to my office comp to graph each cycle and to do the programming. I got the controller running and messed with it a bunch dialing in the autotune function. Once done I sat down a drew up the actual reflow profile I need in the ramp and soak to match the, pretty classic, 4300 profile a lot of reflowers use.

Once I had it drawn I turned to my new Solo Controller to program the Ramp and Soak only to see to my HORROR the damn controller only accepts Hours:Minutes! What a complete fail!
URL]
Reflow is based around seconds..

Looking around I see about 1/4 of the RandS controllers out there allow seconds. Most thru telling the timebase Minutes:Seconds verses Hours:Minutes. None appear very super fast available nor can I even find prices, meaning large circular time wasting ahead, asking price - "no, I'm not paying $600 for your controller you don't stock" type discussions.

Has anyone got any suggestions? Know of any PC SW that can use a regular PID controller and send time based setpoint values out to it via Modbus? Then my profile could just be sent to any'ol PID contoller moment by moment including the otherwise installed one I have now.

Keith Cress
kcress -
 
Replies continue below

Recommended for you

All of the R/S controllers I have used are expensive industrial units. These are programmable for hr/min or min/sec by flip of a switch (or digit). Are you sure yours is not selectable? Does it really mater? Must you have a dwell or ramp that is a fraction of a minute? 1 degree/sec is 60 degrees/min.

Also, make sure you have reliable over-temperature protection now. A PC is not reliable for that function.
 
Hi Comp;

The OTemp would be controller managed as a not to exceed temp.

When reflowing you generally ramp to a temp and hold it there for maybe 2 minutes.
Then you ramp to another temp and hold for maybe a 1 minute or so.
Then you dart up 25C in 30s then immediately retreat those same 25C in another 30 seconds
Then bail completely back to room temp very quickly but not overly quickly - often cracking-a-door-open speed.

It's that 30s that is very important - hence having a 1 minute controller is a problem.

They often even call reflow control 'Ramp and Spike' due to that profile.

I was trying to figure out if I could use an alarm. Then set the alarm to be tripped at the appropriate tip of the spike causing all heat to stop. The alarm relay could then turn on the cooling fan for the cool-down.

Keith Cress
kcress -
 
Outside the box, if your heating load is fairly stable from batch to batch, can you set your heat input (current to the heater?) at a rate that will take the required ramp time to reach the new temperature?
Maybe a little cut and try, or try and cut.
If you need more linear, divide your 30 seconds into 10 intervals and have the PC bump the setpoint of a pid controller every 3 seconds to get the linear ramp that you need. Note; The ramping setpoints will be higher than the soak setpoint. At the end of the ramp, return the setpoint to the desired soak temp.

Bill
--------------------
"Why not the best?"
Jimmy Carter
 
Compositepro; I failed to mention I did re-check: No minutes:seconds is not an option.

Bill; I'd love to have my PC bump the setting every three seconds. Problem is you need to talk Modbus and have a function timer to pace it all. I spent 4 hours looking for some SW that offers that. :/

Keith Cress
kcress -
 
Hi Keith. Is the 30 seconds critical? Does the ramp have to be linear? Have you measured the temperature time constant of your process?
If the heating time constant is several times more than 30 seconds, you may be in a usably linear portion of the heating curve.


Bill
--------------------
"Why not the best?"
Jimmy Carter
 
Hello again.
Will this product from Omega help?

COMMUNICATIONS
(USB Standard, Optional Serial and Ethernet)
Connection:
USB:

Ethernet:
Standard RJ45
Serial:
Screw terminals
USB:
USB 2.0 host or device
Ethernet Standards Compliance:
IEEE 802.3 10/100 Base-T auto-

Ranges and Accuracies
for Supported Inputs
Ramp and Soak Segment Times:
00.00 to 99.59 (HH:mm) or 00.00 to
99.59 (mm:SS)
See url:
You may find the software that you need here:
ftp://ftp.omega.com/public/TEMPERATUREGroup/products/CnPt/


Bill
--------------------
"Why not the best?"
Jimmy Carter
 
If it weren't so simple, I would build something around a quad opamp and some resistors/capacitors. I know that Jim Williams and Bob Pease do that in their heavenly labs.

Gunnar Englund
--------------------------------------
Half full - Half empty? I don't mind. It's what in it that counts.
 
Personally, I would just order another DIY kit from the web for <$100 and be done with it. Those controllers designed for hobby reflow ovens already take into account the need for profiles and sub-minute timing.

Take the ControLeo2, for example... $69 shipped, and even controls a stepper motor to open the door partially for those cool-down stages:
For $190 you can get it and a kit to help completely rebuild an oven.

The Zallus had a Kickstarter and costs $95, including a nice little color screen:

A boatload of other DIY controllers out there, but so many of them are darn near plug-and-play (and customizable) it's not worth the time trying to put together something on your own.

Dan - Owner
URL]
 
Status
Not open for further replies.

Part and Inventory Search

Sponsor