Aluminum Thread

Hey Folks,

I’m interested in making a few aluminum frames in the coming year. But other than TIGing some 6061 (non-structural) channel a few years ago I’m a bit clueless on “Best Practices” for aluminum frames.

This is what I currently understand (from working with overseas factories)


  1. 6061 needs to be heated and quenched.
  2. once that’s done you have a limited amount of time to align the frame.
  3. when the frame’s aligned it needs to be “aged” at a specific temperature for 12-ish hours.


  1. is different and easier. ( :slight_smile: )
  2. is different and easier.

According to the GoogleBox

  • 7005 can be aged at 200F for 6 hours + 320F for 4 hours (a few variations of this have been seen)
  • 7005 can be aged at room temp for 30-ish days
  • 7005 Unlike 6061, this alloy does age-harden over time, reducing the need for a full frame heat-treatment (Reynold’s website)

6061 is out from a cost standpoint (as of 11/19/23 #LAsucksforHeatTreating)

But 7005 might be an option – what I’m not clear on is this: Does 7005 have to be aged at temperature or am I OK to build a frame, align it and hand it from the rafters for 30 days before I build it?

ALSO, if there’s folks that are building alloy frames, please add to this thread. I’d love for this to be a legit resource for folks.


I know nothing about this but I can’t say I haven’t had thoughs. I saw in one of Phil Vandelay’s videos on YouTube he mentioned he hasn’t done any heat treatment on any of his frames and has not had any issues.


@ben.land101 you’re our resident welding guru. Do you have any insight?

Yeah, I’ve come across info like this too. I think about it in two ways: “What can I get away with for a personal bike.” + “What do I need to do to ensure the best/safest build for a client?”

Getting the whole picture helps me figure out if I want to go down this route.

Hi there,

I was thinking of starting an aluminium thread, too :slightly_smiling_face:…I build frames in 7005, as well as steel. I’m also one admin at the Facebook @Aluminum Frame Builder Group. We often get these sort of questions over at the group, so I’ll try to transfer some of the knowledge here.

The choice to go with 6061 or 7005/7020 is a balance between tube availability, and heat treatment requirements. 6061 is universally available in non-bike tube and billet, but only Fairing sell a few bike tube specs for small builders, and it has huge heat treatment requirements and associated costs. 7005/7020 is relatively rare in the outside world in commercial tube or billet, but Dedacciai and Fairing sell small volumes of bike-specific tubes and billet to small builders. It also has far lower (or even zero) heat treatment requirements.

So here’s a summary on 7005/7020 that I put together:

"7005 is a predominantly Taiwanese/Chinese alloy, 7020 is a European specification. They have near identical chemistry and mechanical properties, and are fully compatible in all manufacturing processes: 7005 Aluminum vs. 7020 Aluminum ::…
Many materials summaries will show 7*** alloys un-weldable, but unlike other 7*** alloys, the copper content in 7005/7020 is low enough to keep hot cracking to a minimum. It welds as easily as 6061, but using 5356 filler.

Unlike 6061, with 7005/7020 tubing there is no need for the initial solution post-weld (high temperature + quench), just the secondary ageing process (lower temperature + time).

Ideally this is done artificially in a controlled manner with an oven. Powdercoat oven, homemade or converted domestic oven, pizza oven etc. will all do the job with low risk and low(ish) energy requirement.
Suggested heat and time specs varies a little between manufacturers, but not to such an extent that different tube sources and 7005/7020 alloys cannot be mixed. For example: 6hrs @ 90C/200F then 4 Hrs @ 150C/350F (Columbus and Fairing), or even simpler, one temp down the middle 12-14 Hrs @ 120C/250F (Dedacciai), followed by air cool to ambient (ie: take it out of the oven :blush:).

However, usefully with 7005/7020 there is a large degree of ageing that can come naturally (but slower) from just room temperature + time (see diagram), which means that the artificial ageing process can be avoided, if absolutely necessary, but results are far less predictable (See also: Liability Insurance, Litigation, Due Diligence, Good Practice etc :wink:). Some properties, particularly yield and fatigue strengths, are never fully returned (see table). 2-3 weeks at room temperature is considered adequate.


‘Room Temperature’ or ‘Ambient’ is normally considered as 18-22C, cooler will age slower, warmer will age faster. It’s all about Heat Volume. Freezing can postpone the process almost indefinitely.

Additional Notes:
-Any crimping or bending of tubes should be done after a full annealing process, and followed by the full 2-stage re-treatment, otherwise it is just ‘pre-crashing’ the frame in terms of yield and fatigue strength.
-Align immediately after welding of possible, certainly within 8 hrs of welding.
-The less common 7*** scandium-bearing alloys (Easton Sc7000 or Dedacciai Aegis etc.) MUST be welded with Sc-specific rods and artificially aged in all cases, in order to take full advantage of the material properties. They are amazing materials, but have have very thin walls with hardly any wriggle room on their safety margins. They can be mixed with 7005/7020 alloy tubes and parts, but with different HT settings: 5h @ 135C/275F then 2h @ 150C/300F (Easton).
-Some of the detrimental micro-crystalline properties around the HAZ of a weld will only be removed with the full solution + quench + artificial age process, if at all. Well designed tubes, weld forms, sequences and structures should be able to minimize their impact and an artificial heat cycle will reduce their influence on frame life."

I’ll dig up the 6061 summary in a bit…

All the best,
Dan Chambers


Thanks for the shout-out, I have a good bit of experience welding aluminum, but I have to admit the metallurgical side is not something I’ve studied enough to give sound advice.
Most of the aluminum I’ve welded over the years is not destined for heat treating, or I was simply the welder. I.e, I didn’t design the thing.

I follow the mindset of ‘lot of extra work, for not a lot of gain’ in the context of aluminum bikes. But of course that’s just my opinion.

But it sounds like @Dan_Chambers knows his stuff. I don’t have any experience with the bicycle specific tubes or alloys.


Here’s what I have on 6061. :slightly_smiling_face:

This from Easton:
"6061, Normal processing.

  1. Weld.
  2. Solution heat treat: 30 minutes at 980F/527C. Quench in water or water/glycol (no option for mist or air), keeping quench liquid at 100F/38C or lower. (15-20 secs?).
  3. Align: You want to do this as soon as the material is quenched, max of 8 hours. The more you wait, the harder aligning becomes, and the greater the residual stress that gets locked in. This can make the material less resistant to fatigue failure.
  4. Delay: At least 72 hours at room temperature.
  5. Artificial age: 8 hours at 350F/177C.

If you need to make significant bends or crimps in 6061 you have to anneal. Only 2 hours at 675F/357C, then reduce temperature to 650F/343C, and soak 30 minutes. Follow by air cool, then form as soon as possible after that. No need to heat treat before welding, as the whole frame will get it afterwards"

Lon Kennedy, Nova:
“The treatments do make a LARGE difference. 6061 loses about HALF of it’s strength where the heat affected zone creates a fully annealed condition. The annealed condition 6061-O has a tensile [strength] (depending on how long held at temp around 800 degrees) of around 18-20KSi (fully T-6 is 42Ksi) To get 6061 back to T-6, you have to heat to 985F hold for 2-4 hours and quench in 22% glycol at 20 degrees per second- this cooling can be accomplished with glycol spray or in very special situations with forced air/gas cooling with inert gas. Bottom line is the rate of cooling must be 20 degrees/sec to get the hardness. Immediately after quench for bike frames typically there will typically be distortion that can be straightened within the first 4 hours (before onset of passive precipitation hardening). T-4 is worth about 32Ksi. To get to T-6 you have to age at about 350F for 8 hours. Back in day both GT and Trek had alignment tables within 20 feet of the heat treatment ovens.”

“Al-Si-Mg 6000 Alloy
This special alloy from the 6000 series supplied in F [T-4], is not self tempering and thus necessitates a full heat treatment of the welded frame. Excellent TIG weldability. Very good resistance to corrosion, without trace of the inter-granular corrosion phenomenon [Stress Corrosion Cracking].
It is necessary to heat treat the frame as follows: 535 C for 35 minutes, cooling in water and additives solution. Artificial ageing at 180° C for 10 hours.
Suggested welding rod: 4043/4145.”

Artificial ageing 6061 at 150-200C will increase mechanical properties back to T6, but unless the whole frame has been solution treated back to T-0 (same as the areas around the welds), then the less heated areas of the frame will be over-aged, with declines in mechanical properties. See paper here:
That said, I have built many lightweight structures, not bikes) without post-weld heat treatment, but took account of the reduced mechanical properties.

All the best,
Dan Chambers


Hey Dan

This is such great info thanks so much.

Just to clarify, when comparing artificial aging vs natural aging, are you saying that the only way to return the yield strength to full is with heat or that it never comes back with both processes?

Again, I’m loving the detailed amount of info here!

Looking at the table above, natural ageing over time will return Tensile strength fully, but not Yield or Elongation. Given how influential Yield and Elongation are to Fatigue strength, I suspect that it is not fully returned either, but don’t have data for that.

With artificial ageing post-weld (bottom row of the table above) all the properties of 7005/7020 are reliably returned to the T6 levels.

Lon from Nova had some more to say on this:

"7005 does not require the heat and quench cycle to regain it’s strength, only the re-ageing process (8 hours at 200 followed by 4 hours @ 320). With 7005 the second 320 stage is vital because it not only initiates precipitation hardening, it pushes the material through a phase change that is essential to increase the fatigue life and to stop stress corrosion cracking (SCC). If this is done to spec, 7005 final tensile strength is 52Ksi- almost 24% stronger and fatigue life is 52% higher than 6061.

For 7005 the first (low temp) part of the treatment creates the onset of precipitates.This is where some controversy lies. Natural aging will get the formation of precipitates, but this takes about 3-6 months at room temp and there is no guarantee of the phase change that is vital to fatigue and SCC resistance. The second elevated 320F temp pushes the material through a low temperature phase change. Easton’s process was always a two stage with the second stage temp elevated. This phase change creates a new solidus phase that is more stable than the one from the first stage of aging. The net result is intergranular stresses are reduced and made consistent. For 7005 this has the effect of increasing fatigue life and reducing stress corrosion cracking. I’m not the expert, but Easton had a guy with a PhD in materials with specialties in Aluminum alloys, and a couple of M.S. Mech Es that specialized in characterization of the alloys they used in all possible conditions. This included data not commercially available on area reduction (cold work) effects on fatigue, corrosion resistance, strength (tensile, yield, total elongation at failure, etc.)."

All the best,.
Dan Chambers


Great! Thanks for further clarifying (I’m going to have to get a bit more versed on this stuff).
Now onto figuring out how to build myself an oven. :smiley:

On the oven, look out on YouTube etc for self-build powdercoating or electric pizza oven guides.

Small alloy wheel refurbishing ovens (for 4 wheels) are a good size for 2-3 frames, usually around 3’ x 3’ x 4’ tall, and turn up as salvage quite often. Most commercial PC ovens are too big, or the right volume but too boxy for a single bike frame.

Depending on your plans, it can be pretty small, even a hacked up domestic oven since the volume and amperage required for a single frame is similar to a domestic oven. Most of the parts are standard McMaster Carr stuff.

-Shroud the element to avoid radiant heating of the frame and promote convection.
-The only additional requirement is a programmable thermostat controller to get the two-stage heat sorted. These are usually $30 Inkbird itc-100vh PID units.
-Since the temps are low, domestic rockwool or K-wool insulation is perfectly adequate, no need for firebrick/kiln insulation or such.
-Not lining the inside with metal will speed up heating cycle and reduce power use by reducing the thermal mass. It also eliminated possible therma bridges to the outside shell. Internal metal lined ovens are for constant heated PC ovens operated by gorillas.

Personally, I rent the overnight use of a local wheel refurbisher’s oven, but am looking to build my own. Here are a couple of self-builds from Andy Gilmour and Ross Shepherd:

All the best,
Dan Chambers