-
Hello ES! We could use some help to get us past the finish line on building the new knowledgebase for the forum.
Can you donate? Please see our fundraising page. Thank you!
DIY Pulse Arc welding copper directly to cells
- Thread starter devmonkey
- Start date
I'm gradually working towards a short run of kits and/or finished units. It is a lot of work though.
troubled81
New here
- Joined
- Aug 22, 2018
- Messages
- 4
very interested in this!!! and am truly impressed with the product you have developed. Please Message me when you are looking for someone to test. I have this 1400hp liquid temp controlled monster of pack to build.
Attachments
brianthelion
New here
I'm keen to try one as soon as you have some ready 
I made two interesting discoveries today. I have noticed that when welding over and over on the same small area on the knife blades i've been using to test on the electrode picks up contamination and eventually stops firing. When this happened I was stopping, removing the electrode and regrinding the tip with a dremel diamond disk. It also happens when welding cells but at a much slower rate as the welds are much more spread out so generally you are welding over fresh copper.
Over welding the same small area again and again:
Done a lot of test welds, this is 400 blades.
The first discovery was that switching to a 1mm 1.5% lanth electrode (down from 1.6mm) dramatically reduced the rate the electrode got contaminated, this maybe because it gets a bit hotter and burns off the contamination (copper oxide) a bit more?
The second discovery was that rather than shape the end of the electrode to a truncated cone the 1mm electrodes worked best with a straight cut end, this means that when the tip does need to be redressed it can be done by just touching it off on a dremel diamond disk WITHOUT REMOVING THE ELECTRODE FROM THE WELDER, literally takes 1 second then you can get back to welding. On cells I think this new electrode will need to be redressed every few hundred welds which given each cell only needs 3 welds per terminal isn't much of a problem.
1.6mm tungsten ground to truncated cone:
New 1mm electrode just ground flat, much easier:
Redressed using a diamond wheel in a dremel, about $1:
Over welding the same small area again and again:
Done a lot of test welds, this is 400 blades.
The first discovery was that switching to a 1mm 1.5% lanth electrode (down from 1.6mm) dramatically reduced the rate the electrode got contaminated, this maybe because it gets a bit hotter and burns off the contamination (copper oxide) a bit more?
The second discovery was that rather than shape the end of the electrode to a truncated cone the 1mm electrodes worked best with a straight cut end, this means that when the tip does need to be redressed it can be done by just touching it off on a dremel diamond disk WITHOUT REMOVING THE ELECTRODE FROM THE WELDER, literally takes 1 second then you can get back to welding. On cells I think this new electrode will need to be redressed every few hundred welds which given each cell only needs 3 welds per terminal isn't much of a problem.
1.6mm tungsten ground to truncated cone:
New 1mm electrode just ground flat, much easier:
Redressed using a diamond wheel in a dremel, about $1:
Last edited:
dhruvpatel00764
New here
any update on this machine? I'm following you. thanks.
Thanks for the continued interest.
Much progress has been made and following more testing by myself and some local testers I am making a final but significant design change. The issue with the current design is the kinematics of the needle actuation (the welder needs to know exactly where the needle is every microsecond) require some level of tuning. This would lead to variation between units and potentially over time as the system aged which would be a pain for both you and me. I have redesigned the entire actuation system (again) to eliminate this problem, which also required power stage miniaturisation to co-locate it in the weld head.
New PCBs are being finalised and will go to fab next week. I will update here once that testing is complete.
Didn't see this till now. What is lanth?
A-DamW
Regular
Lanthanated is an alloy of tungsten for welding electrodes.
I have a product development question I need a steer on before moving forwards, hopefully you guys can help answer it for me.
Background
The currently proposed final welder consists of a compact head unit which you hold in you hand. This head unit contains the electrode actuator and position sensing, power stage for forming the arc current, gas control and mcu. Connections to the head unit are power (6S battery), gas, and USB channelled down a single sleeve.
The welding process is configured with 20+ parameters that control everything from electrode pressure, speed, forming current curves, retraction distance, gas timings and more obvious things like total weld energy.
These parameters are stored in a recipe called a weld "profile". This allows us to setup different profiles for different jobs, for example welding 0.3mm copper to cells will use a different profile from 0.2mm copper or 0.2mm nickel. Copper to copper or brass to brass or any mixture would require different recipes. The current weld profile is stored in an EEPROM in the weld head so it persists between restarts.
How I expect the welder to be used is we dial in a profile for a particular job then don't really change it again until a different type of job comes along, for example 99% of the time I am welding 0.3mm copper to tabless cells.
User interfaces
As designed the head unit has no display, and one or two buttons that are used to arm/disarm the unit, release the electrode (for when you need to change it) and purge the air, it has a couple of status LEDs that indicates armed status and will also indicate errors (such as under voltage or over temperature). This very limited user interface is all that is needed whilst actually using the welder for building packs.
It is difficult to include a richer user interface on the head unit, due to size and also that OLEDs / LCDs don't like being millimetres from the very large current spikes that occur during the weld.
However we need a rich user interface for configuring weld profiles, viewing detailed weld statistics and graphs for when we are tuning in new materials or types of weld.
Proposal
The USB connection is there to allow firmware updates, it could also be used for setting up weld profiles if connected to an app.
The app would work on any PC (windows/mac/linux) or Android device and would not require installation (it would use WebUSB so at most will require you to install Chrome/Edge/Opera but most people will already have these installed)? This is a similar experience to using a VESC where you do some detailed setup in the app then just use the configured ESC. This would also make it easy for us to share weld profiles.
If you are only ever going to weld the same materials then you would only have to do this once anyway, however exposing the detailed configuration will hopefully lead to experimentation and discovering new uses for the welder.
Summary
Configure and fine tune the welder by plugging it into a device over USB and using the app. When satisfied or bored of looking at detailed results you can just unplug the welder and continue to weld using the last configuration stored.
I'd like you thoughts on whether this is an acceptable approach?
Background
The currently proposed final welder consists of a compact head unit which you hold in you hand. This head unit contains the electrode actuator and position sensing, power stage for forming the arc current, gas control and mcu. Connections to the head unit are power (6S battery), gas, and USB channelled down a single sleeve.
The welding process is configured with 20+ parameters that control everything from electrode pressure, speed, forming current curves, retraction distance, gas timings and more obvious things like total weld energy.
These parameters are stored in a recipe called a weld "profile". This allows us to setup different profiles for different jobs, for example welding 0.3mm copper to cells will use a different profile from 0.2mm copper or 0.2mm nickel. Copper to copper or brass to brass or any mixture would require different recipes. The current weld profile is stored in an EEPROM in the weld head so it persists between restarts.
How I expect the welder to be used is we dial in a profile for a particular job then don't really change it again until a different type of job comes along, for example 99% of the time I am welding 0.3mm copper to tabless cells.
User interfaces
As designed the head unit has no display, and one or two buttons that are used to arm/disarm the unit, release the electrode (for when you need to change it) and purge the air, it has a couple of status LEDs that indicates armed status and will also indicate errors (such as under voltage or over temperature). This very limited user interface is all that is needed whilst actually using the welder for building packs.
It is difficult to include a richer user interface on the head unit, due to size and also that OLEDs / LCDs don't like being millimetres from the very large current spikes that occur during the weld.
However we need a rich user interface for configuring weld profiles, viewing detailed weld statistics and graphs for when we are tuning in new materials or types of weld.
Proposal
The USB connection is there to allow firmware updates, it could also be used for setting up weld profiles if connected to an app.
The app would work on any PC (windows/mac/linux) or Android device and would not require installation (it would use WebUSB so at most will require you to install Chrome/Edge/Opera but most people will already have these installed)? This is a similar experience to using a VESC where you do some detailed setup in the app then just use the configured ESC. This would also make it easy for us to share weld profiles.
If you are only ever going to weld the same materials then you would only have to do this once anyway, however exposing the detailed configuration will hopefully lead to experimentation and discovering new uses for the welder.
Summary
Configure and fine tune the welder by plugging it into a device over USB and using the app. When satisfied or bored of looking at detailed results you can just unplug the welder and continue to weld using the last configuration stored.
I'd like you thoughts on whether this is an acceptable approach?
Last edited:
scianiac
Regular
- Joined
- Oct 24, 2018
- Messages
- 999
I'm sure many will say they want hardware controls and no apps and stuff and while that is nice and all it does add a lot of cost, complexity, etc. Would it be possible to use the app interface you describe but also make the controls future proof in some way, that is if in the future the app stops working the settings can still be changed. For instance if the settings on the device are stored in a text file that can just be manually edited by plugging it in via USB and the device also shows up as a storage device that exposes it's storage.
troubled81
New here
- Joined
- Aug 22, 2018
- Messages
- 4
I don't plan on switching up a bunch, so this works for me.
Or just open source the app
Frank
Regular
Your proposal would work for me!
dhruvpatel00764
New here
this would be nice to see for somebody like many of us looking for diy.
This is fantastic! Wow! How best to get a hold of you? I need to weld 0.2mm copper to 0.2mm copper for a project I'm working on.. I'm willing to buy anything needed to do so? Let me know if you are in any possition to help me out.. I have alot for trade and alot more to offer.. Thank you
Capekayak90
New here
This thread actually prompted me to make and account so I could be part of the conversation and hopefully purchase one of these when its available at some point.
I meant to comment on the interface development but time got away from me. However I would like to say that for myself and I would think at least some others, I would like to eventually develop my own attachments for more freeform welding so having access to at least some of the protocols and programming directly would really help. I'm personally not really a fan of mobile apps to control tools and would rather modify the coding more directly but I could definitely still work with just modifying an open source app. PS: I totally recognize that I'm probably more of a niche case on here and not asking you to cater to me just thought I'd share some input
I meant to comment on the interface development but time got away from me. However I would like to say that for myself and I would think at least some others, I would like to eventually develop my own attachments for more freeform welding so having access to at least some of the protocols and programming directly would really help. I'm personally not really a fan of mobile apps to control tools and would rather modify the coding more directly but I could definitely still work with just modifying an open source app. PS: I totally recognize that I'm probably more of a niche case on here and not asking you to cater to me just thought I'd share some input
mybatteryguys
New here
What happened to this project? Any updates? I am working on somthing similiar with a laser I put together and some custom weld patterns.
Project is still continuing, latest design works great and based on the feedback it can be used completely standalone - I added a small display and rotary encoder to the base unit, I have one further round of PCB protos to order once I get the displays in my hands to validate. I'm still trying to work out how to best make it available so people can buy it or buy the parts to build their own. It currently consists of,
1. Weld head pcb (Pre-programmed)
2. Base unit pcb (Pre-programmed)
3. Custom micro stepper motor
4. Gas valve
5. A set of custom CNC brass parts for the weld head
6. Cables, tubes, O-rings, bearings, electrodes, gas and electrical connectors
7. 3D printed enclosure and weld head parts
I need to order the custom steppers and CNC parts in batches of at least 100 so I will soon be putting up a list people can add themselves to if they want a kit. There maybe some soldering of through hole components (XT90/XT60 connectors, display and encoder) required on the base unit pcb. All SMD components will be pre-soldered.
Longer term if it proves as useful to people as it has been to me I would look to partner or license the system for a proper production run.
1. Weld head pcb (Pre-programmed)
2. Base unit pcb (Pre-programmed)
3. Custom micro stepper motor
4. Gas valve
5. A set of custom CNC brass parts for the weld head
6. Cables, tubes, O-rings, bearings, electrodes, gas and electrical connectors
7. 3D printed enclosure and weld head parts
I need to order the custom steppers and CNC parts in batches of at least 100 so I will soon be putting up a list people can add themselves to if they want a kit. There maybe some soldering of through hole components (XT90/XT60 connectors, display and encoder) required on the base unit pcb. All SMD components will be pre-soldered.
Longer term if it proves as useful to people as it has been to me I would look to partner or license the system for a proper production run.
Last edited:
Capekayak90
New here
Heck yes!! This is so exciting! I will almost assuredly be asking to be put on the list.
PS: I could be shooting myself in the foot here as I'm not exactly flush for funds myself currently but I've seen group buys done on forums many times in the past and they often end up being a logistical nightmare. Might I suggest streamlining a way to make a deposit even if its not the full cost to cover the stepper and CNC order I have found that once people have invested some amount of money into it they are far more likely to follow up later and it also just makes the project more real.
A happy-ish medium could be to cover the deposit of the custom stepper while also getting access to a copy of the CAD files for the weld head as someone can always send them to PCBway or SendCutSend and get a one off part made if they need to. Just and idea
Capekayak90
New here
Also I'm not sure if I missed it in the updates but what are the specs on the stepper motor? # of steps? Rough Dimensions? Etc? I've actually been looking for tiny good quality steppers for a while to use for things like camera lens focus/filter controls and whatnot where speed is needed but higher torque (relative to a stepper motor) isnt. Most of the time when they get small they get really cheaply made, only have a few stator poles (poopy for positioning) or they use an inline gearbox. I may be actually be interested in picking up a few of the steppers for my own projects, especially if I there is room to mount an encoder on one end. If you don't have time for my shenanigans though do not worry about it haha
Similar threads
