Desoldering Wick: Width, Tube vs Ribbon, and ROL0
Open a fresh roll of desoldering braid and the label gives you three things: a size number, a width, and a colour. It is tempting to assume the number and the width are the same fact written twice. I actually put the Chemtronics 2018 catalog pages 18-19, five official TDS sheets, the product pages on their website, Hakko's WICK spec page, and six live Amazon listings side by side, field by field, and the number, the inch value and the millimetre value contradict each other across three documents from the same manufacturer. The only field all three agree on is the one that looks least like a specification: the colour.
Let me be explicit about money first. The Amazon links below are affiliate links, and I earn a commission if you buy through them. Every width, tolerance, flux code and unit price in this article comes from a manufacturer document, a TDS, or a live product page, so you can check each one yourself. The commission has nothing to do with the numbers.
And let me be explicit about the boundary. This round I did not solder a single joint with any of these braids, and I did not measure braid thickness. Anything about feel, wicking speed or residue is quoted from manufacturer copy and labelled as such. Anything I calculated is labelled a conversion or an estimate. That distinction matters here, because measured data is exactly what this category lacks.
TL;DR: three rules if you just want to buy correctly.
One, buy by colour and by inch, not by millimetre. The millimetre column differs by up to 20% across three documents from the same vendor (table below). The colour code (White/Gray, Yellow, Green, Blue, Brown, Red) is the field that survives all three documents, and it carries across brands.
Two, check whether the braid is a flattened tube or a single layer. Most wick is a flattened tube of braided copper. The lead-free series is a single-layer ribbon: roughly half the copper cross-section at the same width, less capacity, but it heats faster. Use the ribbon on lead-free joints and the tube for clearing large pads.
Three, read the four-character flux code. ROL0 is rosin, low activity, zero halides. REL0 is synthetic resin, low activity, zero halides. The words "No Clean" printed on the bag are not a classification, and they map to different codes at different brands.
Six fields on one roll, and only one of them agrees
Lay a roll of braid out against its datasheet and you find six separate things: a size number (1 to 6), an inch value, a millimetre value, a colour, a part number, and a length. They are not six ways of writing one number.
For the same 80 series (Soder-Wick rosin), three official documents give these widths:
| Size | Colour | 2018 catalog | TDS (Rev J) | Website product page |
|---|---|---|---|---|
| 1 | White/Gray | .030" / 0.8 mm | .030" / 0.76 mm | 0.03" / 0.8 mm |
| 2 | Yellow | .060" / 1.5 mm | .060" / 1.27 mm | 0.05" / 1.3 mm |
| 3 | Green | .080" / 2.0 mm | .080" / 1.78 mm | 0.07" / 1.8 mm |
| 4 | Blue | .110" / 2.8 mm | .110" / 2.54 mm | 0.10" / 2.5 mm |
| 5 | Brown | .145" / 3.7 mm | .145" / 3.30 mm | 0.130" / 3.3 mm |
| 6 | Red | .210" / 5.3 mm | .210" / 5.08 mm | 0.200" / 5.1 mm |
(All three documents retrieved 2026-11-10; sources at the end.)
Row two is enough. The same 80-2-5 is 1.5 mm in the catalog, 1.27 mm in the TDS, and 1.3 mm on the website. The gap between the high and low value is 18.1%. Row three: 2.0 versus 1.78, a 12.4% gap. Row five: 3.7 versus 3.3, 12.1%.
The colour column is identical in all three documents. That colour scheme also carries to another brand: MG Chemicals sells the 426-NS as #4 Blue and the 427 as #5 Brown, exactly matching the table above. That is why I said buy by colour. It is the one field on the roll that no unit conversion has touched.
The TDS millimetre column was computed from someone else's inch value
The TDS column deserves a closer look. It prints .060" as the width and 1.27 mm as the metric width. But .060 inch is exactly 1.524 mm, and 1.27 mm is what .050 inch converts to.
In other words, the TDS took its inch values from the catalog and its millimetre values from the narrower inch values on the website, and stitched two sources into one table. Recomputing from its own inch column gives +20.0% for size 2, +14.2% for size 3, +10.0% for size 4, +11.6% for size 5 and +5.0% for size 6. Five of the six rows disagree with their own inch figure.
The tolerance is not consistent either. The TDS appends ±0.25 mm to every row; the website prints ±0.01 mm. That is a factor of 25. A 0.76 mm braid with ±0.25 mm tolerance is ±33%, so "1 mm wide" is not a statement you can make about desoldering braid. It is a nominal value only.
Practically, I would use the TDS tolerance: if you order a "2.0 mm" braid, plan for a real range of roughly 1.75 to 2.28 mm. That spread is wide enough for the same braid to either just cover a pad or overhang it, which is why the official usage instruction is not a number: choose a braid width equal to or slightly larger than the pad or connection.
Numbering: three different "size 2" braids at 0.76, 1.27 and 1.90 mm
The second trap is the word "size". Inside one brand, two product lines do not share a numbering scale.
Chem-Wik 2-5L: .030 inch, .76 mm in the TDS. In the catalog's selection table it is size 1, because the "2" in that part number is a different code entirely (2, 5, 7, 10 map to .030, .050, .075, .100 inch).
Soder-Wick 80-2-5: here the 2 really is the size number, .060 inch, printed as 1.5 mm in the catalog and 1.27 mm in the TDS.
Soder-Wick lead-free (40 series) size 2: .075 inch, 1.9 mm.
All three are called size 2, and they measure 0.76, 1.27 (or 1.5) and 1.90 mm. The narrowest and widest differ by a factor of 1.5. If you reorder "the size 2 that worked last time" but switch series, you get a different object.
The lead-free size 4 repeats the problem: .095 inch / 2.4 mm in the TDS, .110 inch / 2.8 mm in the 2018 catalog, and .110 inch in the Amazon title for 40-4-5. One part number, two values, 15.8% apart.
Capacity is set by construction, not by width
Width governs coverage. How much solder the braid holds is set by its cross-section. The manufacturer buries the key fact in an FAQ:
Most wick, including Soder-Wick No Clean, is a flattened tube of braided copper. Soder-Wick Lead-Free is a ribbon, a single layer, which lets it heat up more quickly and reduces thermal stress on sensitive components at lead-free temperatures. (Chemtronics FAQ, paraphrased)
The lead-free TDS states the same trade-off from the other side: the single-layer weave is "lighter in mass than any other desoldering braid", which is what allows lead-free removal at lower temperatures.
This is a structural difference, not a dimensional one. As a geometric approximation, at equal width and equal strand thickness, a flattened tube has a top wall and a bottom wall, while a ribbon has one layer, so the ribbon carries roughly half the copper. That is my estimate, not a manufacturer figure. What you buy with that half is thermal mass: less of it, so faster heating, which on a joint above 217°C is often worth more than the capacity you gave up.
Worth noting: none of the four vendors publishes a capacity number. Hakko's WICK page gives width, length, ESD bobbin and flux classification. Chemtronics gives phrases like "high capacity" and "as much as 50% faster than conventional". MG gives "flux percentage <5%". Solder uptake, as a specification, is entirely adjectives. Cross-brand capacity comparison therefore has no common method; you can only infer within one manufacturer's series from width and construction.
One temperature figure is concrete. Step 3 of the TDS usage instructions sets the iron to 600-750°F, which is 316-399°C. That window matches the temperature guidance in the September soldering iron comparison, except that article was about the iron and this one is about how much heat the braid can absorb.
Flux: ROL0 and REL0 are the spec, "No Clean" is not
The line on the bag is marketing. The four-character code behind it is the specification. An IPC J-STD-004 designator packs three facts: the first two letters are the base (RO rosin, RE synthetic resin, OR organic, IN inorganic), the third letter is activity (L low, M medium, H high), and the fourth character is halide content (0 for below 0.05%, 1 for halides present). Independent technical references describe the scheme identically.
Three vendors answer the same question three different ways:
Hakko is clean about it on the WICK spec page: flux type No clean (halogen-free), classification ROL0 per IPC J-STD-004C, halide content 0.00%, copper corrosion test PASS (JIS Z 3197), copper mirror test with no breakthrough (IPC-TM-650), insulation resistance 1x10^8 ohm or higher. Its "No Clean" is in fact a rosin-based ROL0.
Chemtronics says in its FAQ that Soder-Wick and Chem-Wik use natural gum rosin classified ROL0 under section 3.2 of J-STD-004, while the No-Clean and Lead-Free lines use synthetic non-colophony flux classified REL0.
But its own lead-free TDS says Type ROL0. Same company, FAQ says REL0, TDS says ROL0. MG goes further: its comparison table marks all five braid types, including No Clean and Lead-Free, as "R (Rosin) ROL0".
So "No Clean" carries no comparable information on braid. Hakko's No Clean is ROL0. Chemtronics claims REL0 for its No Clean. MG marks its No Clean as ROL0. What you can actually compare is the code and whether residue must be removed.
That residue difference is operational. The Chemtronics rosin TDS says plainly that flux residues must be cleaned after the rework process. The selling point of the lead-free and No Clean lines is that residue is non-conductive and non-corrosive and does not have to be cleaned. Pick the wrong line and you have added a board-washing step.
One more trap: RMA is not ROL0. The MG 427 title says RMA Flux (mildly activated rosin) while its product table prints ROL0. RMA historically means slightly more active than plain rosin and normally lands in the halide-bearing grade under J-STD-004. If a listing says RMA or RA and gives no code, plan to clean afterwards.
Length and units: even the vendor's own conversions disagree
On the same website, 500 feet appears as 152.4 m for one part (5-500L) and 152.5 m for another (50-2-500). Five hundred feet is exactly 152.4 m, so the second is a typo.
The smaller conversions drift too: 5 ft printed as 1.5 m (actual 1.524), 25 ft as 7.5 m (actual 7.62), 50 ft as 15.0 m in one table and 15.2 m in another (actual 15.24). The errors sit between 1% and 2%, which is harmless in use, but it tells you the table was typed by hand rather than computed.
Retail listings are worse. The Amazon title for the Hakko FR551-2515-J reads "0.1 inch (2.5 mm) x 3.9 ft (1.5 m)", while the bullet on the same page reads "Length: 4.9 ft (1.5 m)". 1.5 m is 4.921 ft, so 3.9 is wrong by 20.7% and 4.9 is right. 2.5 mm is 0.0984 inch, so rounding to 0.1 inch is fine.
Unit errors fail differently from width errors. A width error means the braid does not cover the pad. A length error means the roll you thought was two metres is actually 1.2.
Price has a hidden layer too. Catalog prices are case prices. The rosin 80-2-5 case is 25 bobbins for $137.75, or $5.51 per bobbin, or $3.62 per metre. The same part number sells as a single bobbin on Amazon for $9.87, a 79.1% premium. A case price is not a realistic option for most people, but knowing the gap stops you from treating $9.87 as the cost of the braid itself.
Sorting the official 25 ft spool prices per metre makes the structure visible: the narrowest .030 inch braid is $3.159 per metre, the widest .210 inch is $4.273. Width grows 7.0x while price per metre grows 35%. Per millimetre of width per metre, the narrow braid costs 5.2x what the wide one does. Narrow braid is expensive because of processing, not copper. That is why the fine sizes always cost more per gram.
Two-year shelf life, judged by the colour of the copper
This is one of the few figures in the category with a hard limit. All four TDS sheets say 2 years, and the website FAQ says 2 years, adding that this assumes average storage conditions.
The check is official as well: brighter copper is fresher, oxidised copper darkens like a dirty penny, a tightly wound roll lasts longer than a loose bobbin on a bench, and if you suspect a roll you should unroll about 6 inches, cut it off and test the rest. The failure mode is not "does not work". It is "reacts more slowly and needs longer contact", which pushes people to raise the iron temperature or press harder, and that is what actually damages the board.
So checking the manufacturing date before you buy is worth more than comparing two widths.
Six braids you can buy today, and who each one suits
All six were purchasable from the page when retrieved on 2026-11-10, and prices are what the page displayed at that moment (third-party seller prices move; check the page). I give width in inches plus a converted millimetre value, because the millimetre column is not reliable.
HAKKO FR551-2515-J: the one that publishes the full flux classification
2.5 mm (0.098 inch) x 1.5 m, halogen-free flux, officially ROL0 per IPC J-STD-004C, halide content 0.00%. The 1.5 m roll ships on an ESD-safe bobbin; the official table shows no such column for the 25 m rolls, so the packaging attribute changes with length. $17.15 when retrieved, which is $11.43 per metre and the highest per-metre cost of the six.
Real drawbacks: 1.5 m is the only length in this width (the 25 m rolls cover only 2, 2.5 and 3 mm, with no 5 mm option), so heavy rework empties it fast; the Amazon title's 3.9 ft is wrong and misleads on length; there is no 5 mm version for large pads. Suits: lead-free boards, low residue, and anyone who wants the classification documented.
Chemtronics 80-1-5: the fine-pitch size
.030 inch (0.76 mm) x 5 ft, rosin flux, ESD-safe bobbin. $12.99 when retrieved, $8.52 per metre.
Real drawbacks: only useful on pads at the 0.5 mm pitch scale; clearing a large pad with it is slow; the rosin TDS requires cleaning the residue; the official table lists size 1 only in 5, 10 and 25 ft, so there is no 100 or 500 ft spool and the unit cost never drops; stock showed 15 rolls left.
Chemtronics 80-2-5: the general-purpose size, and the one with three widths
.060 inch x 5 ft, rosin flux. $9.87 when retrieved, the lowest single-roll price of the six, at $6.48 per metre. Its width is 1.5 mm per the catalog, 1.27 mm per the TDS and 1.3 mm per the website, so decide which document you trust before you buy.
Real drawbacks: 5 ft goes fast, two boards and it is done; rosin means washing the board; the case price works out to $5.51 per roll, so the single-roll premium is 79% and long-term use is poor value; the 18% gap between TDS and catalog widths means you cannot use it as a datum in a fixture or automated setup.
Chemtronics 40-4-5: the single-layer lead-free ribbon
.110 inch x 5 ft, lead-free series, halogen-free no-clean flux. $12.99 when retrieved, $8.52 per metre. This is the single-layer construction in that series: low thermal mass, fast heating, positioned by the vendor for lead-free rework.
Real drawbacks: at equal width a single layer holds roughly half the solder of a flattened tube, so large pads need a wider size; the width is .095 inch (2.4 mm) in the TDS but .110 inch (2.8 mm) in the catalog and the Amazon title, a 15.8% gap; the seller is third-party (CONCORDE ELECTRONICS); and a single layer is less forgiving if you drag it across pads.
MG Chemicals 426-NS: large pads, ROL0 stated, flux percentage given
0.1 inch (2.54 mm) x 5 ft, No Clean, marked ROL0 in the official comparison table with flux percentage below 5%, ESD bobbin. $13.68 when retrieved, $8.98 per metre.
Real drawbacks: the title gives only inches and no millimetre value, and at 2.54 mm it is 10% narrower than the Chemtronics #4 (.110 inch / 2.79 mm), so the "number 4" label does not transfer between brands; several third-party sellers list it between $13.04 and $13.68; and it comes as a 5 ft roll, so longer lengths mean a different model.
MG Chemicals 427: large pads and terminals, but plan to clean
0.125 inch (3.17 mm) x 5 ft, RMA rosin. $15.96 when retrieved, $10.47 per metre. The second widest of the six, suited to terminals, large pads and shield-can grounding.
Real drawbacks: RMA is mildly activated rosin, and the listing says RMA while the table says ROL0, so treat it as needing a clean; 3.17 mm pulls too much solder from a small pad and can empty it; stock showed 3 rolls left, so restocking is unreliable.
If you work in an ESD-controlled area, the bobbin material matters too. The ESD mat article covers the three resistance numbers for a bench surface, which is the same judgement applied one layer down. The matching problem in solder wire is covered in the solder wire spec article, and fume extraction is in the fume absorber piece.
A ten-minute check
Ignore the marketing copy on the bag and do three things.
One, measure the pad. Use calipers on the pad or connection you need to clear. The official rule is equal to or slightly larger, so a 1.2 mm measurement means the 1.27 to 1.5 mm size, not the 2.5.
Two, match the colour, then confirm the inch value. Colour survives all three documents: White/Gray is the finest, then Yellow, Green, Blue and Brown, with Red the widest. Note the colour, then read the inch column. Treat millimetres as reference only.
Three, compute the price per metre and read the date. Price per metre is price divided by feet times 0.3048. At the same width, a 5 ft roll and a 25 ft spool can differ by 2x per metre, and the two-year shelf life runs from the manufacturing date, not from the day you open it.
By job, four branches:
- 0.5 mm pitch, 0402 and smaller: 0.76 to 1.3 mm, narrow, accept that you will change it often
- Through-hole, 0805 to SOP: 1.27 to 2.0 mm, general purpose, flattened tube
- Large pads, terminals, shields: 2.8 to 3.7 mm, tube, one pass
- Lead-free joints where heat is a risk: single-layer lead-free ribbon, width chosen by pad, not widened for capacity
If you cannot decide, a general-purpose braid (1.3 to 1.5 mm, flattened tube, rosin or REL0) is hard to get badly wrong. It covers most hand rework, while narrow and wide sizes each only help at the extremes.
FAQ
Q: What happens if I buy a braid that is too wide?
A: It pulls too much solder at once, especially on a small pad. Empty the pad and you have to add solder back, which is an extra round trip. The official rule is equal to or slightly larger than the pad, not as wide as possible. A wider braid also has more thermal mass, so at the same iron temperature you need longer contact.
Q: Does "No Clean" mean I never have to wash the board?
A: Read the code, not the words. ROL0 and REL0 are both low activity with zero halides and usually can be left alone, but the rosin TDS explicitly requires cleaning after rework even though it is ROL0. Anything labelled RMA or RA should be treated as needing a clean.
Q: Can I reuse a piece of braid?
A: No. The official FAQ puts it plainly: once the braid reaches soldering temperature the flux has been used up, so that section will not draw more solder and only acts as a heat sink. Cut off the used part and work from the end of the roll.
Q: How does this split with a desoldering pump or an electric desoldering station?
A: Braid is flat and sits against a planar pad, so it suits drag-soldering and clearing residual solder. A pump has suction and suits through-holes. They are complementary. I am writing this from category knowledge only, without a measured comparison, so check the official specs for selection.
Q: Can old braid be saved?
A: The official test is colour: bright copper is fresh, dark copper is oxidised. Unroll about 6 inches, cut it off and test the rest. If it reacts noticeably slower, do not compensate with a hotter iron, because that moves the thermal stress onto the board.
What this article did and did not do
What I actually did was a document-level comparison: Chemtronics 2018 catalog pages 18-19, five TDS sheets (rosin Rev J, no-clean Rev H, lead-free Rev D, unfluxed Rev H, Chem-Wik Rev E), the product pages, Hakko's WICK spec page and six Amazon listings, field by field, with every disagreement converted into a percentage.
What I did not do is any hardware testing. I did not measure cross-section thickness, I did not run a wicking speed comparison, and I did not test ionic contamination of residue. That is why there is no measured solder-uptake figure anywhere in this article and why the capacity section is explicitly labelled an estimate. Anyone quoting "brand X picks up 30% more solder" without a test method is probably quoting copy. This category does not even have an agreed way to measure capacity.
Bottom line
The braid spec table looks as if it holds one number. It holds four that bite: width contaminated by conversion, numbering that does not cross series, capacity set by construction, and flux that must be read as a code rather than an adjective. The most stable field on the roll is the least technical one, the colour.
Before you buy: measure the pad, match the colour, check the date. Get those three right and you never have to argue about whether size 2 is 1.27 or 1.5 mm.
---
Sources (all retrieved 2026-11-10): Chemtronics 2018 product catalog pages 18-19; Chemtronics TDS_SWRosin Rev J, TDS_SWNoClean Rev H, TDS_SWPBFree Rev D, TDS_SWUnflux Rev H, TDS_Cwik Rev E; chemtronics.com product and FAQ pages (soder-wick-rosin, soder-wick-no-clean, soder-wick-lead-free, chem-wik-rosin, desoldering-braid-wick); hakko.com English WICK specification page; six amazon.com listings (B0DWM8T2NY, B00FS7AUDW, B01I7Q2ULA, B004IOMM0U, B00424MWHO, B008O9VQ5M).
👉 Join Xiaomi MiMo Platform: Leading AI model platform with cost-effective inference
👉 Join Aliyun AI: Top AI products with exclusive coupons for business innovation
📌 This article was AI-assisted generated and human-reviewed | TechPassive — An AI-driven content testing site focused on real tool reviews
🔗 Recommended Tools
These are carefully selected tools. Using our affiliate links supports us to keep producing quality content:
As an Amazon Associate, I earn from qualifying purchases at no extra cost to you.