Most plasma equipment quotations are compared on two numbers: chamber size and rated power. Both are easy to put in a table, and neither predicts whether the machine will still be producing good parts two years after installation.
What actually decides the outcome is a combination of plasma uniformity, process control, vacuum performance, gas delivery, thermal behaviour and validation discipline — the parts of a specification that are rarely quoted and almost never compared side by side.
This guide is written from the buyer's side of the table. It lists the criteria worth specifying, the questions worth asking, and the failure modes that show up six months after acceptance when a decision was made on the wrong two numbers.
The one-line version
- Buy a validated process window on your own parts, not a machine with a bigger chamber.
- Acceptance should be defined in measurable numbers before you order, not negotiated after installation.
- Compare total cost of ownership, not purchase price — pumps, gas, fixtures and downtime usually outweigh the capital gap.
1. What You Are Actually Buying
A vacuum plasma cleaner ionises a process gas inside a low-pressure chamber using an RF or microwave source. The reactive species then act on the workpiece in four ways: chemical oxidation and decomposition of organics, ion bombardment and mild physical sputtering, removal of weak boundary layers, and surface activation that raises surface energy.
The important consequence for procurement: the machine does not deliver a result, it delivers a process window. Two systems with identical chamber dimensions and identical RF power can produce completely different uniformity, cycle time and material damage — because the electrode design, gas distribution, pumping arrangement and fixture strategy differ.
If you are still deciding whether plasma is the right process at all, start with our primer on plasma surface cleaning, which covers what the treatment does to a surface and how vacuum and atmospheric compare. This article assumes you have already decided to buy a vacuum system, and focuses on how to choose one.
2. The Six Specifications That Decide the Outcome
| Specification | What to ask for | What goes wrong if you skip it |
|---|---|---|
| 1. Chamber & loading | Effective loading area (not just volume), fixture and tray geometry, part spacing, uniformity map across the chamber | A large chamber that treats the middle well and the corners poorly; throughput that collapses once real fixtures are loaded |
| 2. Power & process control | Source type and frequency, usable power range, ignition stability, matching network, multi-step recipes, recipe locking and audit trail | Operators adjusting recipes by feel; discolouration, etching or embrittlement of sensitive polymers from over-treatment |
| 3. Vacuum system | Ultimate vacuum, process pressure range, pump-down time under real load, leak rate, pump type, maintenance interval | Cycle times that never reach the quoted takt; oil backstreaming contaminating the chamber; frequent unplanned maintenance |
| 4. Gas delivery | Number of channels, MFC accuracy, required gas purity, consumption per cycle, interlocks, recipe storage per gas | Poor run-to-run repeatability; gas cost that quietly exceeds the electricity bill; safety exposure with hydrogen or oxygen-rich mixtures |
| 5. Temperature & material compatibility | Measured workpiece temperature during treatment, thermal build-up in continuous operation, effect on your exact polymer/film | Warping, colour shift, dimensional change, or adhesion that degrades in ageing tests rather than at the moment of treatment |
| 6. Automation & traceability | MES / barcode / robot interfaces, data logging of every process parameter, user permission levels | No way to prove which recipe a suspect batch ran; manual loading becoming the bottleneck as volume grows |
3. The Three Specifications Buyers Skip
3.1 The vacuum system is the cycle time
Pump-down and venting often consume more of the cycle than the plasma itself. Ask for pump-down time measured with a realistic load — trays, fixtures, parts — not an empty chamber. Confirm the ultimate vacuum, the working pressure range, and the leak rate that is guaranteed at acceptance.
Pump type matters beyond speed. Oil-sealed pumps can backstream vapour into the chamber, which is exactly the contamination you are trying to remove; for cleanliness-critical work a dry pump and a clean chamber design are usually worth the higher price. Also ask for the maintenance interval and the cost of the consumables — this reappears in the TCO discussion below.
3.2 More power is not more cleaning
Rated power tells you the ceiling, not the useful range. What you need is the minimum effective power density for your material, and evidence that the system holds a stable plasma at the low end of its range. Over-treatment causes surface roughening, discolouration, dimensional change, embrittlement and damage to thin films — failures that appear weeks later in reliability testing, not on the day of acceptance.
A supplier who can show a stable process window at low power, with repeatable ignition and a documented matching network, is offering something more valuable than a bigger generator.
3.3 Gas delivery decides repeatability
Two machines with the same gas bottle will behave differently if one has a properly calibrated mass flow controller and the other has a needle valve. Ask how many independent channels the system supports, the flow accuracy, the purity required, and the consumption per cycle — gas is a recurring cost that is easy to under-estimate on a multi-shift operation.
For oxygen-rich or hydrogen-containing chemistries, insist on documented safety interlocks, leak detection and a gas cabinet design that a competent EHS reviewer would accept.
4. Validation: A Visible Discharge Is Not an Acceptance Test
The most expensive mistake in plasma procurement is accepting a system because the plasma lights up and the parts "look clean". Define the numbers before you sign, and test against an untreated control from the same lot:
- Water contact angle before and after treatment — fast, cheap, and a good proxy for wettability.
- Surface energy / dyne where a goniometer is impractical on the floor.
- Bond strength — peel, lap-shear, pull — on the actual interface you care about.
- Coating or printing adhesion, including cross-hatch where applicable.
- Uniformity across the chamber, sampled at corners and centre, not just where the parts usually sit.
- Batch-to-batch repeatability over at least several consecutive cycles.
- Ageing and reliability — thermal cycling, humidity, or whatever your customer's specification requires.
One more thing to validate explicitly: hydrophobic recovery. Many activated surfaces gradually lose part of the effect after treatment. Contact angle measured one minute after unloading can look excellent while the same surface measured after two days in a workshop is barely better than untreated. Fix the maximum waiting time between plasma and the next process step as part of the acceptance criteria, and write it into the work instruction.
5. What to Demand in the RFQ
A professional enquiry gets a professional answer. Group your requests so suppliers can respond point by point:
| Group | Request |
|---|---|
| Documentation | Full datasheet; chamber dimensions and effective loading area; utility requirements and installation drawings; materials of chamber, electrodes, seals and fixtures; pump and RF source brands with expected service life |
| Process | Pressure, power, gas flow and treatment time ranges; sample test report on your own material; standard and custom fixture options; recipe capacity, permissions and data traceability; uniformity data across the chamber |
| Commercial & service | FAT and SAT procedures; spare parts list and preventive maintenance schedule; warranty coverage and service response time; MES / barcode / robot interfaces; training and documentation package |
If a supplier cannot produce a test report on your material and will only demonstrate with their own reference sample, treat that as a data point about how the project will go after the purchase order.
6. Seven Procurement Risks and How to Control Them
| Risk | Likely result | Control |
|---|---|---|
| Selecting by chamber size alone | Poor uniformity or insufficient throughput | Test with real parts, real fixtures and real loading density |
| Selecting by rated power alone | Material damage or excessive etching | Establish the minimum effective process window |
| Ignoring the vacuum system | Slow cycles, contamination, frequent maintenance | Check pump-down time, leak rate, service cost and consumables |
| Validating only initial contact angle | Bond or coating failure weeks later | Add strength, ageing and reliability testing |
| Ignoring surface ageing | Activation lost before the next process step | Define and enforce maximum hold time after treatment |
| No recipe access control | Undocumented operator changes, batch variation | Recipe locking, user levels, audit trail |
| Testing on a different configuration | Performance loss when scaling up | Validate the exact machine, fixture and loading density you will buy |
7. Total Cost of Ownership Beats Purchase Price
Two quotes that differ by 30% on capital cost can easily invert once running costs are counted. Build the comparison on these lines:
- Equipment price, installation and commissioning
- Vacuum pump and RF source maintenance, plus spare parts
- Process gas consumption per cycle × cycles per year
- Fixture replacement and chamber cleaning labour
- Calibration and preventive maintenance downtime
- Operator training and documentation upkeep
- Scrap and yield loss from an unstable process
- Technical support responsiveness when production stops
A machine with better process control and traceability frequently wins on lifetime cost even at a higher purchase price — because the expensive line items are downtime and yield, not the chamber.
8. An Eight-Step Procurement Workflow
Define the job
Material, contamination type, part dimensions, volume and the downstream process. Write it down.
Set targets
Measurable cleaning and adhesion criteria on representative samples — contact angle, bond strength, coverage.
Test suppliers equally
Same samples, same targets, comparable conditions. Compare reports, not brochures.
Compare performance
Cleaning result, uniformity, cycle time, gas and energy use, maintenance load.
Run a trial
A small production run exposes repeatability problems that a demo never shows.
Freeze the URS
User Requirement Specification with process conditions, acceptance criteria and service obligations.
Contract it
Put the process window, acceptance numbers and response times into the purchase contract.
FAT → SAT
Factory test, installation, site acceptance, training, then ramp-up with monitoring.
9. Quick Recommendations by Priority
| If your priority is… | Then specify… |
|---|---|
| Cleanliness and repeatability | Calibrated gas delivery, dry vacuum system, recipe management and traceability |
| Throughput | Effective loading capacity, automation, load/unload time, multi-chamber options |
| Heat-sensitive materials | Fine power and pressure control, temperature monitoring, low-damage process capability |
| Cost sensitivity | Total cost of ownership including pumps, gas, fixtures, maintenance and yield loss |
| Bonding and coating | Final interface strength and long-term reliability as the primary acceptance criteria |
10. How We Handle This Before a Quote
We do not quote a plasma system from a part photo. Our sequence is: review the workpiece and the contamination, run sample tests in a chamber with your actual material, report contact angle and bond data against an untreated control, and only then propose a configuration — batch, inline, atmospheric or custom.
That sometimes means recommending a smaller machine than the one in the original enquiry. It is a better outcome than a system that fails acceptance.
Send Samples Before You Send a Purchase Order
We will treat your parts, measure the result against an untreated control, and send you the data with a configuration proposal.
Request a Sample Test