Record the evidence that makes a part repeatable

Generated from content/lms/fabrication-and-composites/02-control-the-workshop-process/05-record-process-evidence-for-repeatability.md; edit the source file, not this page.

Source path: content/lms/fabrication-and-composites/02-control-the-workshop-process/05-record-process-evidence-for-repeatability.md

Course: Fabricate composite race-car parts with workshop discipline

Module: Control the workshop before controlling the laminate

Estimated duration: 55 minutes

A composite part is not repeatable because you remember how you made it. It is repeatable when the next build can be controlled from evidence instead of memory. For an intermediate fabricator, that means you record the actual conditions, measurements, timing, checks, defects, corrections, and cure evidence that explain why the part came out the way it did.

This lesson is about the working record, not about shop cleanliness, material storage, dust control, or drawing the whole workflow before you touch cloth. Those are separate skills in this module. Here you are already at the bench. The mould is in front of you. Fabric has been cut. Resin is about to be mixed. The question is simple: what evidence do you need to capture so the same part can be made again with fewer guesses?

The core principle is this: record the variables that disappear. Before cure, you can see whether the cloth is bridged in a corner, whether the release film is tucked down, whether the bag is moving under vacuum, whether the resin is still workable, and whether the workshop is warm enough for the process you chose. After cure, much of that evidence is gone. You can see the finished surface and the trim line, but you usually cannot see the catalyst measurement, the pot-life margin, the exact bag behavior, the leak repair, the resin quantity consumed, or the point at which the cure was far enough advanced to stop managing it. Your record preserves those disappearing facts.

A good process record has two jobs. First, it lets you repeat a success. Second, it lets you diagnose a failure without guessing. If the first part was well consolidated and the second part was resin-rich, bridged in a corner, or soft at demould, the record should make the difference visible. If the record only says laminated nosecone, cured overnight, trimmed, it has failed. That is a diary entry, not process evidence.

The repeatability chain has eight links. Identify the part and tool. Record the actual materials. Record the environment. Record the mix. Record the time window. Record the lay-up and consolidation checks. Record the vacuum or pressure evidence when those methods are used. Record the cure and inspection evidence. If any one link is missing, the next build becomes a reconstruction exercise.

Start with identity. Give the part a name, a build number, a date, and a mould or tool reference. If there are bolt-on mould sections, flange moulds, or an envelope-style bag because the mould lacks useful flanges, record that before the resin is mixed. If you are making a nosecone, a flat panel, a cover, a duct, or a bracket, the record should distinguish not only the part family but the exact version. The point is not paperwork for its own sake. The point is to know which process produced which physical object.

Then record the equipment that controls measurement. The bonded material emphasizes simple tools: weighing scales for resin, syringes or graduated beakers for catalyst, and a thermometer for workshop temperature. Those tools are not background equipment. They are repeatability instruments. If your resin and catalyst quantities are not measured, the cure record begins with a guess. If your workshop temperature is not recorded, your pot-life and cure-time notes cannot be compared honestly from one day to the next.

For resin work, write down actuals, not intentions. The planned resin quantity matters, but the actual resin weight used matters more. The planned catalyst addition matters, but the actual catalyst volume or mass matters more. The source example describes a likely catalyst addition rate of 1 percent, or 1 ml per 100 g of resin, but you should not turn that into a blind rule. Record the rate you used, the amount you measured, and the temperature at which you used it. A process that works at one temperature may not give you the same working window when the shop is warmer or cooler.

Pot life is a timing variable, so it belongs in the record. A test mix tells you how much time you have at the catalyst rate and temperature you are actually using. That is especially important when the part is large enough that you must decide how many resin mixes to make and how big each mix should be. If you only record the total amount of resin prepared, you miss the control question that mattered during the job: did you have enough working time to wet, position, consolidate, bag, and pull vacuum before the resin began to gel?

Make timing visible. For each mix, record mix start, application start, application finish, and the point when the material was no longer useful. If you are vacuum bagging, also record when bagging started, when final vacuum was applied, and whether final vacuum happened before the resin began to gel. The source material is clear that the time taken to make the laminate matters because final vacuum must be applied before gel. That is a repeatability-critical fact. If the next build is slower, the process may fail even when the material recipe is unchanged.

Record preparation evidence before the wet work starts. Vacuum work punishes missing materials because the bagging stack has to be applied while the resin is still in its working window. Your record should confirm that peel ply, release film, breather or bleeder fabric, sealant tape, bag film, connectors, gauge if used, and pump were ready before lamination began. This is not a broad workflow map. It is a process-control checkpoint for this specific operation. If the team loses ten minutes looking for release film while the resin is advancing toward gel, that lost time belongs in the evidence.

The next evidence category is mould condition. For vacuum bagging, the mould must not have sharp corners or spikes of reinforcing fibre that can puncture or damage the bag. If the mould has wide flanges, record that they were available for sealant tape and bag attachment. If it does not, record that you used an envelope bag. Also record any condition that affected bag fit: tight corners, complex curvature, flange interruptions, repairs, or rough areas that had to be dressed before bagging. These are not cosmetic notes. They explain why the bag either conformed cleanly or fought you.

For lay-up, record the fabric reality, not just the design. The source example for the nosecone begins with the required fabrics already cut and the likely resin quantity calculated. Your record should show the ply count or section count actually used, where strips were added, and whether any areas were treated differently from the main field. If you are using chopped strand mat in sections, record how the sections were placed and how the corners were handled. The source warns against bending mat over tight corners where it can form bridging with a void beneath it. That is exactly the kind of defect that must be recorded if it happens and exactly the kind of checkpoint that should be recorded when it does not.

Consolidation evidence belongs beside lay-up evidence. In wet lay-up, the source process uses a brush or roller to consolidate the layer, wet out the material, stipple, roll, and move to adjacent sections until the main areas and flanges are laminated. In your record, consolidation is not a vague word. Write what you checked: wet-out complete, no dry patches visible, corners butted or overlaid as intended, no obvious void under tight-corner material, flange areas laminated, extra corner strips applied where planned. If a part later shows a dry area or corner weakness, these notes tell you whether the problem was visible during lay-up, created during bagging, or discovered only after cure.

Do not hide deviations. A repeatable process record is allowed to contain problems. In fact, it must contain them. If a fabric section shifted and you slid it into position with gloved hands, record it. If a corner tried to bridge and you corrected it with overlapping strips, record it. If you added resin because a section did not fully wet out, record it. If a later part uses less resin, a different corner treatment, or a faster consolidation pass, the record will explain why the result changed.

Vacuum bagging needs its own evidence because it adds several failure points after the laminate is already wet. The bag material does not stretch, so it must be oversized, with slack taken up in tucks and creases. Record that the bag had enough slack, where the tucks were positioned, and whether the bag was eased into corners before full vacuum was held. If the bag is too tight across a corner, it can bridge. If the release film or bag bridges, pressure does not reach the laminate where you need it. The finished part may look acceptable until the weak spot is loaded or trimmed.

For release film, record coverage and conformity. The source states that release film follows single curvature but not complex curvature, and that complex areas may need overlapping pieces. It also gives a practical edge extension of 20 to 30 mm beyond the mould edges, while keeping it off flanges needed for sealant tape. Those are process details worth recording because they change how the laminate breathes, bleeds, and releases. If release film failed to reach a corner, or if it covered a flange needed for sealant tape, the defect may appear later as a leak, a bridge, or a release problem.

For breather and bleeder materials, record both function and placement. The glossary describes bleeder cloth as a felt-like non-woven material that soaks up excess resin during cure and allows air removal during vacuum. It describes breather fabric as a non-woven material that allows air to be removed from the vacuum bag and may soak up excess resin. That means placement is process evidence. Around a through-bag vacuum connector, the source calls for extra breather layers to prevent resin from being sucked into the connection. If you used that protection, write it down. If you did not and the connector stayed clean anyway, write that too. The next build should not have to rediscover the condition.

Vacuum evidence is more than pump on. Record what the bag did when the pump started. The source tells you to watch the bag move and contract as the air is removed, then position tucks and creases as required. It also says the pump can be turned off and vacuum partially released to reposition the bag or ease it into corners. Those actions are normal process corrections, not embarrassing mistakes. Put them in the record: first pull time, repositioned bag at left rear corner, partial release used, second pull held, no bridges visible. That note is far more useful than a clean but empty statement that vacuum was applied.

Leak evidence is one of the strongest repeatability clues in the whole process. The source gives several checks: the bag should not be pullable away from the back of the laminate, a vacuum gauge can monitor sustained vacuum, vacuum should not decline too rapidly, and quiet hissing may identify leaks. Record the gauge reading if you use one. Record whether the reading held, fell slowly, or fell rapidly. Record where leaks were found and how they were stopped, such as pressing a portion of sealant tape or patching the bag. If the part later has voids or uneven consolidation, a real leak log can separate a material problem from a bagging problem.

Cure evidence starts before demould. Cure is the process that changes resin into a hard and tough substance and causes it to adhere to the reinforcing fibres. Curing time is the period a moulded component must be cured before removal from the mould. Those definitions sound basic, but they matter because cure is not just a clock on the wall. It depends on material, catalyst, temperature, part size, and whether you are using ambient or elevated temperature. Record cure start, cure environment, any elevated temperature used, and the time you decided cure was advanced enough for the next action.

A check sample is one of the most useful pieces of evidence in the bonded corpus. The source recommends making a small check sample to the same lay-up as the component and placing it in the same environment, then using it as an indicator of cure stage. That is a repeatability tool. It lets you inspect cure progress without disturbing the actual part. Your record should give the sample identity, lay-up match, location, time checked, and observed condition. If the part is large or expensive, the small sample may be the safest place to test whether the resin has advanced enough for handling decisions.

When vacuum is used, record pump management through cure. The source advises keeping the pump running until you are certain cure is well advanced. It also warns that once the pump is turned off, it should be disconnected so the bag does not suck oil in from the pump. Those are operational details, but they are also evidence. Record pump on time, pump off time, disconnect time, and the cure evidence that justified stopping. If the next build shuts the pump down earlier and the laminate is less consolidated, the records will show a real process change.

Demould and post-cure handling also need evidence. Record when the bag was removed, whether the bag was saved, and whether breather fabric and release film were discarded. If peel ply is left in place until later bonding or lamination, record that. Peel ply left in place is not clutter; it is a controlled surface condition for future work. If a later bonding operation fails, knowing when peel ply was removed and whether the protected surface was exposed becomes part of the evidence trail.

Inspection evidence is where process evidence becomes useful to someone else. The roll-bar inspection material in the corpus is not a composite lay-up procedure, but it shows the same discipline. It assigns responsibility for proof of alloy steel to the participant, requires an inspection hole to verify wall thickness, and lists visual weld acceptance conditions such as no cracks, fusion, filled craters, and limited undercut. The fabrication lesson is the same: an inspector, instructor, crew chief, or future you needs proof, not confidence. For composites, the proof may be resin and catalyst measurements, workshop temperature, pot-life test, bag vacuum hold, check sample condition, cure time, and visible defect notes.

You should build the record around acceptance questions. Was the material measured? Was the workshop temperature recorded? Was the test mix done at the same catalyst rate and temperature? Was the laminate completed in the working window? Were corners checked for bridging? Was the bag tucked into all corners? Did vacuum hold? Were leaks found and corrected? Was cure verified by a same-lay-up check sample? Was the pump kept running until cure was well advanced? Was peel ply left in place when future bonding was intended? Each yes or no should be backed by an actual note, number, time, or observation.

The strongest record style is simple: planned value, actual value, evidence, decision. Planned value might be resin quantity or number of mixes. Actual value is what happened. Evidence is the measurement or observation. Decision is what you did because of it. For example: planned two 150 g mixes, actual three mixes of 120 g, 110 g, and 80 g because the first test mix showed shorter pot life at the prevailing temperature. That record teaches the next build more than a perfect-looking form full of blanks.

Do not confuse repeatability with rigidity. Repeatability does not mean every part must use the same number of resin mixes, the same bag tuck pattern, or the same cure time regardless of conditions. It means the reason for each choice is visible. If the shop is colder, cure may take longer. If the mould shape changes, the bag may need different slack. If you add matched pressure tooling to control thickness and finish, the process record must add pressure-tool evidence. The source describes pressure moulding as a way to achieve consistent laminate thickness, good consolidation, and finish on both faces when the extra tooling time and material are justified. If you use that method, the evidence must show how pressure was applied and why it was appropriate for that component.

This is also why record fields should be stable but not blind. Stable fields let you compare builds. Blind fields encourage box-ticking. A good wet lay-up and vacuum record might include part ID, mould ID, operator, date, material batch or description if available, workshop temperature, resin mass, catalyst amount, mix times, pot-life test result, fabric sections applied, consolidation checks, release film coverage, breather placement, bag slack and tuck notes, vacuum pull time, gauge hold or leak notes, pump-off and disconnect time, check sample observations, demould time, and defects. If a field does not apply to a particular process, mark it not applicable with a reason instead of leaving it blank.

The skill you are practicing is not handwriting. It is process memory. You are building a record that survives cure, cleanup, and the natural human urge to remember only the successful parts. The next time the same component is made, you should be able to sit down before mixing resin and know the time window, mix sizes, bagging risks, corner treatments, vacuum behavior, and cure evidence from the prior build. If you cannot do that, the part is not yet repeatable. It may be familiar, but it is not controlled.

The final standard is practical. A repeatable record lets another competent fabricator make the same part without asking you what you meant. It lets an instructor or engineer review the job and see where the real controls were. It lets you compare a good part and a bad part without relying on mood, memory, or folklore. For an intermediate builder, that is the move from making a part once to owning the process that made it.

Worked example: wet lay-up nosecone process evidence

Use the nosecone example as a process-card exercise. Before lamination, the fabrics are already cut and the likely resin quantity has been calculated. Your first evidence entry is not that the nosecone was laminated. It is the actual plan: mould sections fitted, bolt-on base section installed or left off, flange moulds fitted, fabric sections ready, expected resin quantity, test-mix result, catalyst rate, and workshop temperature.

The critical timing decision is mix size. The source describes using a test mix to determine pot life at the chosen catalyst addition rate and prevailing temperature, then using that information to decide how many mixes of what size are needed. For the record, that means you do not simply write resin mixed. You write mix one, actual resin amount, actual catalyst amount, time mixed, time first applied, and time it stopped being useful. If the part required a second or third mix, each mix gets its own line.

During lay-up, record what happened at corners and flanges. The source process begins with a generous coat of resin, careful positioning of the first CSM section, and sliding it into exact position if needed. It warns that CSM should be butted into corners where there are flanges, and not bent over tight corners where bridging could leave a void beneath. Your record should capture that checkpoint directly: first section positioned, corner butted, no bridge visible, overlap strip planned for corner strength, section consolidated by brush and roller, resin added until wet-out complete.

If a corner does bridge, do not clean up the record to make the job look smoother. Write the bridge location, how it was discovered, how it was corrected, and whether the correction changed the ply stack or resin use. That is how the second nosecone becomes better than the first. The evidence teaches you which corners are naturally difficult, which sections need trimming, and whether the resin window is long enough for the actual hand work required.

Worked example: vacuum-bagged laminate evidence

For a vacuum-bagged part, the record starts before the laminate is wet. The source says bagging materials need to be to hand before lamination and that the final vacuum must be applied before the resin starts to gel. Your evidence card should therefore start with readiness: peel ply ready where later bonding or lamination is intended, release film cut, overlapping release-film pieces prepared for complex curves, breather fabric ready, extra breather layers prepared for the vacuum connector, sealant tape placed or ready, bag oversized enough for tucks and creases, pump tested, gauge ready if used.

Once the laminate is complete, record the stack in order. Peel ply applied to relevant areas. Release film applied, with overlap pieces used in complex curvature. Release film extended beyond the mould edge but kept clear of flanges needed for sealant tape. Breather wrapped around the component. Extra breather layers placed under the connector. These entries are not decorative. They explain how air could leave the bag, how excess resin could be managed, and how the vacuum connection was protected.

When the pump starts, record the dynamic evidence. Did the bag move and contract? Did it pull into the corners? Did any bridge appear? Did you turn the pump off and partially release vacuum to reposition a tuck or ease the bag into a corner? Did the bag become impossible to pull away from the back of the laminate? Did the gauge hold, or did it fall? If you heard hissing, where was it and what stopped it?

The cure record finishes the example. Keep the pump running until cure is well advanced, then record why you judged that condition had been reached. Use a check sample made to the same lay-up and placed in the same environment as your indicator. Record pump off, pump disconnected, check sample condition, and demould time. If the next part cures differently, this evidence tells you whether the change came from vacuum, timing, temperature, or cure management.

Common mistakes

The first common mistake is the recipe-only record. It lists materials and target ratios but omits actual measurements, timing, and observations. Good looks like a record that says what was weighed or measured, when it was mixed, what the shop temperature was, how long the material stayed workable, and what changed during the job.

The second mistake is recording after cleanup. By then, the details that matter have already blurred. Good looks like short entries made at the decision points: test mix complete, mix one applied, first section consolidated, corner bridge corrected, vacuum first pulled, leak found, leak sealed, check sample firm, pump disconnected.

The third mistake is treating vacuum as a binary event. Pump on is not enough. Good looks like a bag-behavior record: bag oversized, tucks positioned, corners checked, no bridges, gauge held or leak corrected, bag could not be pulled away from laminate.

The fourth mistake is ignoring temperature. The corpus includes a thermometer in the basic tool list and ties pot-life testing to prevailing temperature. Good looks like recording workshop temperature beside catalyst amount, pot-life test, mix timing, and cure observations.

The fifth mistake is hiding corrections. Fabric shifted, corners bridge, bags leak, and sealant tape sometimes needs pressure. Good looks like recording the correction plainly because the correction may become part of the repeatable method.

The sixth mistake is skipping the check sample. Without it, you judge cure only from time and hope. Good looks like a small same-lay-up sample kept in the same environment and checked as an indicator of cure stage before you make decisions about vacuum, demould, or handling.

The seventh mistake is treating inspection as someone else's problem. The roll-bar inspection material shows that proof, inspection access, and visible acceptance criteria matter. Good looks like building proof into the fabrication record before anyone asks for it.

Drill: three-build process evidence ladder

Run this drill over your next three composite jobs, or over three practice panels if you are not ready to use customer or race-car parts. Each build takes the normal fabrication time plus roughly ten minutes of record work during the job and ten minutes of review afterward.

Build one is the capture build. Before mixing resin, make a one-page process card with identity, environment, material measurement, mix timing, lay-up checks, vacuum checks if used, cure checks, and inspection notes. During the job, fill it in with actuals. The success criterion is that every major decision has a time, measurement, or observation beside it.

Build two is the repeat build. Before starting, read the first card and mark the three process points most likely to change the result: usually pot life, corner consolidation, and vacuum behavior. Build the second part while recording the same fields. The success criterion is that the second record can explain every intentional difference from the first build.

Build three is the handoff build. Give the first two records to another competent fabricator or to your instructor and ask what they would do before mixing resin. If they can identify the mix window, the risky corners, the bagging sequence, the leak checks, and the cure evidence without interviewing you, the record is working. If they have to ask what you meant, revise the card fields before the next real part.

Calibration cues

You are improving when your records become shorter but more precise. Early records often contain long narrative and few numbers. Better records contain the actual resin amount, catalyst amount, temperature, mix times, bag pull time, gauge behavior, leak locations, check sample observations, and demould time.

You are improving when repeated defects become visible in the records before they become visible in the part. If the same corner repeatedly needs correction, the record will show it. If every warm afternoon shortens your working window, the record will show it. If every successful vacuum-bagged part had a same-lay-up check sample and every questionable one did not, the record will show it.

You are improving when the record changes your next build. A process card that never changes decisions is either perfect or ignored. For most intermediate fabricators, the useful outcome is that the next lay-up has a better mix plan, earlier bagging readiness, more deliberate release-film cuts, cleaner breather placement, or a more disciplined cure check because the previous evidence made the weak point obvious.

Author Review

No quiz questions are attached to this lesson.

Sources

#	Document	Chunk	Pages	Score	Collection
1	Competition Car Composites Simon McBeath	e493d9fa-3b52-2c3b-5bc4-8ddf5343ec5d	144	1	uio_books_raw_v1
2	Competition Car Composites Simon McBeath	69987f7e-a33c-74f1-707c-4cc570632952	146	1	uio_books_raw_v1
3	Competition Car Composites Simon McBeath	e3e08be8-ddb2-aefa-2a56-9a8d5ae9284a	105	1	uio_books_raw_v1
4	Competition Car Composites Simon McBeath	963e162e-7795-f2d8-8672-d4f9f546ac68	43	1	uio_books_raw_v1
5	Competition Car Composites Simon McBeath	a92a57d7-66ad-7c18-c969-cf0c0d4005e9	204	1	uio_books_raw_v1
6	Competition Car Composites Simon McBeath	c8ea927b-ee2f-add5-6a09-0c2ae6daa1eb	133	1	uio_books_raw_v1
7	HPDE Verbatim Master Compilation	a03d609e-dcc5-cb19-1f39-ffff53145161	272	1	uio_books_raw_v1
8	BMW_CCA_RMC_Driving_School_Manual_2023-06	341434edcc05f83a7aa1afe19f6b6bc4	29	1	uio_books_raw_v1