Questions about Ultrasonic Welding

by Ken Holt, Hermann Ultrasonics, Inc.

What effects and influences will a glass reinforcement in a plastic have upon the ultrasonic welding process?
This is an excellent question as these types of additives are common in semi-crystalline parts. Unfortunately, the question has two considerations that need to be contrasted and compared. Both positive and negative influences can be seen.

Positive Influences. In ultrasonic welding, glass reinforcement stiffens the plastic part and thereby lessens the amount of amplitude or vibration lost within the part in contact with the horn/sonotrode. Why is this good? By getting as much vibration to the weld joint area as possible, the best weld is assured to occur between the parts to be welded. This effect is most notable in ‘high-loss’, softer semi-crystalline materials, particularly polypropylene (PP), polyethylene (PE), polyamides (Nylons or PA), and Acetal (POM). The welding of these types of lower modulus plastics benefits from the addition of glass reinforcement or even a talc or CaCO3 (calcium carbonate).

Percentages of 10 to 33 percent are most commonly seen and are proven ratios for Nylons and Acetal in ultrasonic welding, particularly in the automotive industry where good stiffness is added and tighter molding tolerances are allowed.

Negative Influences. Plastic ultrasonic welding cannot melt those materials commonly added to plastic materials, only the plastic matrix in which they occur. The reinforcements do detract from the amount of plastic available to weld at the weld joint area and steps must be taken in the molding process to either produce a ‘resin-rich’ surface (which also results in shinier parts) or a surface with no more than the base ratio of reinforcement, e.g., 33 percent. In other words, keep the glass levels at or below the specified percentage in the weld joint area. If temperature profiles in the molding machine are run at low levels, a resultant and undesired glass-rich surface will occur.

Another negative influence is the wear evoked on the face of the horn/sonotrode by these materials, so proper coatings or hardenings of the tooling should be undertaken.

There seem to be many process parameters involved in the welding process. How can the process be controlled most thoroughly and accurately?
Fortunately, process control techniques and mechanisms are numerous. These can be broken down into three main categories for clarification: electronic, mechanical, and management.

Electronic. A computer-controlled welding system has many options to allow process control by allowing multiple limits of outputs to be set. For example, if welding by energy or, in other words, energy mode, it can be specified that only parts within a certain overall height tolerance be accepted. Likewise, those with weld times outside a researched, good range can be rejected. Typically, modern units allow for the setting of limits in most all other weld parameters than those programmed as main weld parameters. Once again, a computer-controlled welder is required for these.

Mechanical. There are many mechanical aspects of the welding process, including tooling, mechanical settings of the press, and auxiliary equipment influences (part clamps, protective film usage, pick and place part movement, etc.). The most common problem I have experienced in this regard is the adjustment of accessible controls by well-intentioned individuals who do not have a full understanding of the cause and effect relationships of the changes they make. For example, to increase the throughput of a machine, someone may change the weld force to shorten the weld time needed for the weld, but by doing so, inadvertently produces an unwanted and adverse strain within the part. Settings that are easy to adjust should be documented and verified on a regular basis and kept at the welding station for easy reference and comparison.

Welding units that do not allow such easy-to-reach adjustments will be changed in software settings, and proper password protection in software is common. In this way, only those knowledgeable in the cause/effect relationships are allowed to make adjustments. Medical manufacturing, in particular, can benefit from such systems as changes can be tracked and assigned for FDA-compliance purposes.

All mechanical adjustments to tooling relationships, likewise, should be recorded and should be repeatable. Leveling height adjustments are a common example and heights of the corners of the fixture plates can be measured easily and recorded.

At the root of this question is not only what adjustments are to be made but also, and most important, why are these adjustments necessary? What input to the process has changed and how can that factor be minimized? Is a variation in part warpage causing problems? How can that be minimized? Is a rise in moisture content of the plastic part to blame? How about the material lots? Has the melt index of the resin been changed with an addition, regrind, or a reformulation of the base product? Is parting line flash now interfering with the fit of the part into the fixture? Take appropriate steps upstream to alleviate downstream adjustments.

Management. As with all quality assurance programs, managerial ‘buy in’ is tantamount to success. Proper training of operators on the equipment, including access and reading of manuals, and good communication paths between the secondary operations department and molding departments are both items that, in my experience, can help people de-bug processes.

I have heard the term ‘amplitude’ used frequently regarding ultrasonic welding but have never been given a good definition of it, or its influence on the process.
Amplitude is nothing more than the amount of vibration, or strain, seen at the end of the horn/sonotrode and is expressed in microns or 10-3 mm. All sonotrode faces move out and in from an ‘at rest’ position (reciprocal movement). Amplitude can be expressed as a peak-to-peak excursion measurement of the faces movement; or it can be expressed as simply the peak amount of movement forward from the ‘at rest’ position of the face. This is called peak amplitude. It is the amount of travel that the face of the horn/sonotrode experiences in a cycle of vibration. This vibration, whose frequency is dictated by the welder’s operating frequency, e.g., 20 kHz, results from the activation of the piezoelectric crystals within the converter or transducer. It is amplified typically by the component called the booster and also by the shape of the sonotrode. The mechanism of amplification is outside the scope of this writing but can be discerned easily by visits to manufacturers’ web sites.

I am a big fan of using analogies in trying to explain new terms to people in training. As basic as this may sound, the following is a true and good analogy: If you wish to melt butter in a pan faster, you increase the flame height or heat introduction rate. Likewise, if you wish to perform a weld faster, you analogously increase the amplitude. Amplitude is a huge factor in dictating how fast plastics melt or reach their glass transition temperature.

However, we cannot go ‘full throttle’ for a variety of reasons; each plastic resin has a unique value or range of amplitude in which it effectively melts yet is not degraded. Likewise, each plastic part design will have a subset range within the aforementioned, in which parts weld well. Similarly, values of amplitude ranges can be found in manufacturers’ literature or by contacting their application labs.

Tooling engineers at the manufacturers’ shops should provide tooling capable of running correct amplitudes given a certain booster ratio, as well as long-running, robust designs. Also worthy of consideration is that the horn/sonotrode will have a maximum amount of strain that can be expected repeatedly.

Amplitude is an excellent variable to use in the design of experiment study of a welding process. The effect of too low of an amplitude in a welding process is commonly seen as an incomplete weld (even with proper leveling of the fixture), or a ‘cold forming’ of the weld joint. This is where the joint design appears to have been physically deformed but not melted. (Note: silicone weld release present in the weld area will exhibit the same look.)

It is possible to have too much amplitude. This is seen in processes having very low weld times (for instance, less than 0.1 seconds for a small part) or a process that produces flash from the weld area even when it is not fully melted down (or even physical part damage such as cracking of the parts). Lower amplitudes within the previously discussed ranges will provide a more repeatable and robust process. Welding distance versus time graphing can provide important and useful information in the application of the correct amplitude.

Ken Holt has been in the plastics industry for over 20 years and has written for various publications, as well as conducted training and educational seminars. He is the applications lab manager at Herrmann Ultrasonics, Inc., in Bartlett, Ill., where he deals with all facets of the ultrasonic welding process from initial design reviews to final field testing of equipment. He can be reached by e-mail at