Printer Friendly

The struggle to make Q-1: a custom molder's story.

The Struggle to Make Q-1: A Custom Molder's Story

Delivering quality industrial blow molded parts isn't hard - all you have to do is throw the bad parts away. That's how new owners initially pulled tiny Fremont Plastic Products Inc. in Fremont, Ohio, up from a company no longer allowed to bid on jobs for Ford Motor Co. in late 1985, to Ford's coveted Q-1 rating, which Fremont was awarded this May.

"In a sense we got the Q-1 by throwing parts away," says Fremont president Paul Rothschild, who with associate Donald Smith and Rona Rothschild (married to Paul) bought Fremont four and a half years ago as a profitable but down-at-the-heel maker of toys, and auto and drainage parts. "We didn't have the quality processes and procedures then, so we threw parts away. It was expensive, but it was a long-term commitment." Now Fremont has the quality systems in place and is poised to tackle a far harder task.


"To be truly `world class' we have to get to where we don't make a bad part," Rothschild says. A lot is at stake for a small supplier like Fremont. The 1990s are going to be intensely competitive, he says, noting 35% excess capacity in the automobile business. Processing machines will have to run faster, more consistently and with less scrap in order to drive piece prices down.

In the initial quest for Q-1 status, Fremont's new owners redesigned the factory, upgraded existing equipment, bought five new blow molders (at $500,000 each) and expanded to 90,000 sq ft of space with 140 employees, up from 19,000 sq ft and 35 employees.

The original company had had quality problems with Ford, and just before it was sold had lost the right to bid on new Ford jobs, a fact its new owners weren't informed of. Equipment was poorly maintained. "If a machine leaked oil, they'd put a bucket under it and then put the oil back into the machine," recalls vice president and co-owner Smith. There were no computers, SPC systems, process improvements, or material-handling systems. Workers mixed colors by hand out of gaylords.

Fremont's new owners changed all that. They also redesigned the machinery layout, warehouse, office and maintenance areas. Where parts used to be finished anywhere there was space on the shop floor, the new owners built a well-lit, temperature-controlled room for secondary operations. One maintenance person became head of a new maintenance department, and a muddy field became a paved parking lot.

Now Fremont is an immaculate, modern plant four times its original size. Sales rose 350% to $11 million last year, an average of 30%-40% growth per year. Still, in a sense, the big, visible changes were "the easy hits," Rothschild says. Growth and quality are only half of Fremont's story. The other half, which they talk far less about, is streamlining processes for speed.


Because Fremont's new owners and several managers came from Owens-Illinois Inc. and Container Corp. of America, their experience was entirely in high-speed container blow molding lines. So they reshaped Fremont in that image, adapting systems and procedures from bottle and drum technology, which were unusual in industrial blow molding - certainly in a company Fremont's size. "There was an incredible quality improvement from adding smart controls," says engineering manager Steven Wagener. He installed a combination of linear potentiometers, LVDT's (linear voltage differential transformers) and Temposonic ultrasonic position sensors - the latter to control slow-close on the presses. "It was all commercially available and common procedure on bottle lines, but unusual in custom industrial blow molding," he says.

The new controls replaced manually set limit switches and allowed presses to run at or above nominal closing-speed capacity. Fremont thus runs presses rated at 1200 in./min, at 1200-1300 in./min. "A lot of people don't run them that fast, but a half second here, a half second there - it adds up," Wagener says. "In molding bottles, a number of 1/10-sec adjustments would add up to 4-5% volume gain." Fremont now gets 140 parts/hr on four cavities for one job that the first year ran at 51 parts/hr on two cavities; another job runs 120 parts/hr on two cavities that ran 75 parts/hr before, Smith says.

Fremont also built two well-equipped in-house labs, one for parts development and engineering, and one for quality control, with eight people total - five in the q-c lab, two product engineers, and a plant engineer. The q-c lab in turn works with eight utility people, two per shift, who monitor the parts and processes. "Most automotive parts suppliers would have at least as many or probably more people, but the structure of the way we operate lets us do more with less," says quality manager Jack Thomas.


Lab efficiencies are beginning to pay off, as Fremont installs an inexpensive robotic router of its own invention (see PT, Aug. '90, p. 63). The question was whether to get an expensive robot for $50,000 to $150,000, or an expensive, articulated router bit, and spend less money on the robotic manipulator, explains Wagener. Opting for the expensive bit, he devised a router, mounted on an overhead gantry crane, for under $40,000. The unit has four axes of simultaneous movement and can also interpolate two axes for contouring. The whole routing rig is computer controlled with very flexible software, Wagener says. It can, for example, plunge three holes for a hinge, come over and plunge two holes in front at a 70[degrees] angle, then cut and bevel the inside opening for a lid, he says, adding that because blow molded parts are never exactly the same, the router will actually make parts more uniform. Job setups will be done by calling up a menu and selecting the correct job, making setup quick and flexible.

Through in-house engineers working with an outside design consultant, Fremont also recently won its first major automotive contract for a part largely of its own design - a new light truck console made of HMW-HDPE. Fremont made consoles before the takeover, but from customer's designs. For the 1992 model year, Fremont won the bid with its own "more organic-looking" concept that has a lightly textured surface, says product manager Don Staczek.

To achieve this harder-to-mold surface, Fremont is running a series of "designed experiments" to develop methods for improving surface quality. "Rather than waste machine time, we run experiments during a color change," says Wagener. Test parts are marked, measured, analyzed, then reground and reused.

Fremont is very big on use of "design-of-experiments" (DOE) techniques, which alter several variables at once, thereby getting more results out of one experiment, an obvious efficiency for a small company that doesn't have machinery dedicated to R&D. Half a dozen employees have been sent for training in DOE multi-variant techniques at the American Supplier Institute in Dearborn, Mich., which is associated with quality "guru" Dr. Genichi Taguchi, a major proponent of DOE.

DOE research for the new console includes studying new cooling methods (variations on "huff-and-puff" on-off blow techniques) during color changes on the current console. Engineers have hooked up compressors to conduct cooling and blow pressure experiments. Surface quality is affected by more blowing air, not by cooler air, says Joe DeBortoli, machine processing supervisor. During the experiment, they also substituted HDPE's from other suppliers, not Fremont's usual suppliers, to see if the change affected the surface. "We're closing in on a better part," DeBortoli adds.

Previous designed experiments for appearance and cycle-time improvements gave conflicting results, DeBortoli says. Higher blow pressure improved surface appearance but appeared to slow cooling. Fremont's plant is now fed by a 100-psi air system, but if the DOE shows that 130 psi makes a significant difference, Fremont will add a compressor to raise psi to that level.


Once a part is in production, it's overseen by shift q-c people, but the whole shift is actively involved. Thomas heads a "Cpk improvement committee," which includes machine operators and utility people and meets every other week. The group discusses SPC printouts from each shift, comments from the shift supervisors and audit comments from the quality lab, Thomas says.

Cpk, or process capability index, is a mathematical expression of the process capability to produce a part that consistently falls near the center of the specification range (see PT, April '90, p. 116 for an explanation). Cpk also correlates with the statistical probability of producing a bad part. A Cpk of 1.33 (which correlates with a 0.0064% reject probability if process variation follows a "normal" distribution) is minimally acceptable by today's standards, Thomas explains, and the goal is to raise Cpk as high as possible.

For example, Fremont people concentrated on correcting deflection in a latch area of a console, which they monitored intensively. "We drove the Cpk on that characteristic up to 5, meaning variation was almost nonexistent. Now we only have to check every few hours, which drives the cost down," Thomas says. All employees undergo 6 hr of SPC training, while monitoring employees go through more in-house Cpk training. Large boards hang overhead in the plant as constant reminders to watch statistical parameters.

Getting small-company workers to accept such wrenching changes in quality and process systems hasn't been easy. Co-owner Rona Rothschild had a big impact on team building through a period of rapid growth and change. She recruited new employees from drug and alcohol rehabilitation programs and from Ohio state programs designed for workers laid off from other industrial jobs and families on public assistance.

"We don't do anything unique, and we haven't invented anything. It's just fundamentals and known technology," Paul Rothschild insists. In fact, Fremont has invented a patentable router and developed a new production auto part, and that's only the beginning. Fremont's most outstanding innovation so far is its adaptation of mega-company systems and container-making speed to a small custom industrial blow molding operation. "And it's never ending," Paul Rothschild adds. "You can always do something better, no matter how good you are."

PHOTO : `Smart' controls, more common to bottlemaking, doubled output for Fremont's industrial blow molding operation.

PHOTO : Before the takeover, machines were poorly maintained, leading to quality problems and downtime, which lost Fremont the right to bid on Ford jobs.
COPYRIGHT 1990 Gardner Publications, Inc.
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 1990, Gale Group. All rights reserved. Gale Group is a Thomson Corporation Company.

Article Details
Printer friendly Cite/link Email Feedback
Title Annotation:Ford Motor Co.'s quality rating; Fremont Plastic Products Inc.
Author:Schut, Jan. H.
Publication:Plastics Technology
Date:Sep 1, 1990
Previous Article:Materials and machinery advances will shine at polyurethane conference.
Next Article:CIM made simpler.

Related Articles
Blow molding: big changes to come.
World-class injection molding: how do you compare?
Injection molds: product lines reviewed.
CIM brings it to you 'live.' (computer-integrated manufacturing; blow molders monitoring)(includes related article)
Quality teams boost profits of small custom molder.
Hourly rates rose in 1st quarter.
Readers respond to blow molder's wall-thinning problem.
Why molders love all-electric machines.
Injection molding: the next 10 years.
Mixed Signals for Molders in 2nd Quarter.

Terms of use | Copyright © 2017 Farlex, Inc. | Feedback | For webmasters