Do you have files for the dust shoe you’re using?
That’s cute, but I didn’t do all of this for 10mm DOC. I said I was testing.
This is why I did all of this: https://youtu.be/r_J7ueCLIt0
18mm DOC 4k feed.
With plenty of headroom for more speed.
2 Likes
And cause I really want to drive my point home, here’s 18mm DOC 6k feed: https://youtu.be/tDCA-OLRmSE
1 Like
That’s arrogant.
But good on you for doing 6000. Now do some hardwood. 
2 Likes
Following along and learning, cheers for the edutainment!
Nice seeing what’s possible with $$ to $$$ motors/drivers/gears/belts/etc… 48v sounds fun if/when a future JackPot supports, not sure how, 48-60V TMC5160 uses SPI, maybe 36V TMC2225 uart is more doable with less major revisions, but not worth it given 1.4A limit? Until then…
Reliable bang for buck gains using existing components would be very interesting. Sounds like 2:1 reduction mod on X would be $10-15 in parts depending on whether self sourced or bought as a complete kit from V1E… Curious how much this mod helps increase feedrate (and/or headroom) by ?%
Bang for buck for cnc
Bang for buck options for drivers for 48v on cnc will be externals rather than 5160’s. there is little benefit to the options afforded by SPI or UART with TMC drivers isn’t that is application. Best deal will likely $10aud dm556 off AliExpress. Bang for buck for steppers is definitely the 17HS19-2004S at 5 for $80AUD. 2gt 3:1 reduction gears can be had for $6 to $8 a set from AliExpress,
I’ll be testing this combo once parts arrive and will post results
I will not be the person to suggest cheap amazing power supplies because I don’t want that on my head. I’ll be using a second hand flex power x7(industrial grade brand and model, modular supplies 5v,12v,2x 24v and 148v rails through discrete modules) got off eBay US and was cheaper for me to get two of those shipped than two meanwell 48v domestically. Not a recommendation as these are big and I’ll have to get creative with mounting to put it on the gantry. But they go for about $30US today.
2 Likes
1 Like
So whole 2 axis upgrade for drivers, power supply and gear reduction could be had for $80usd - if you already had decent stepper motors. Not cheap enough to go in the lowrider BOM. But if we see consistent gains on multiple machines, may be a viable upgrade for those wanting to eek a bit more out of their machines
1 Like
You could do this with the stock motors. Your gear reduction isn’t what made this possible. The stock motors and drivers at 24v are way more than enough for what the actual machine can do. The limiting factor for the LR4/3 is the flex in the beam. Not at all trying to be rude but you haven’t done anything with these new motors that can’t be done already with the stock motors. And I bet if you really measure your 18mm doc test there you will see the flex in the measurements. Fast and deep is great. But not worth anything if your cuts are not straight and accurate
3 Likes
One more reply about this and then I’m moving on.
I put 10+ full sheets of MDF and ply through my stock LR4. I know what it can and can’t do stock. I’ve made cabinets and kitchen fronts with it and had successes but also failures.
One of the things I’ve noticed for example is that if you can’t achieve the required FPT for your bit, that bit will get dull quite quickly.
So that brings me to the reason for these upgrades. I want plenty of headroom with fast rapids. I’m building a lot of built in closets, cabinets, … for our house and reliability and speed are key.
Yesterday I was able to process 3 sheets of MDF in less than 2 hours to make drawer carcasses. These had tons of pre drilled holes and pockets.
And to get to your point about the gantry flex, we know it does that. It was literally the first thing I told MrFish. It’s next on the list of things to get an upgrade.
I get that a lot of people you deal with end up doing upgrades or attempting to do so without knowing what they’re doing, but between TheMrFish and myself we have 4 CNC machines and a shit ton of 3D printer builds and upgrades.
6 Likes
Yeah, move on to hardwood from MDF. 
2 Likes
Alrighty, this is sounding a little hostile. We all love pushing our machines in different ways, lets keep it nice. We all value different things in terms of speed and price, let’s celebrate when someone breaks some records! 
Erfan, if you don’t mind, I would appreciate it if you call your changes “Modifications” instead of “upgrades”. The reason is while you value the changes you got for the money you spent, I am not sure if it is that important to others. Newcomers without 5 machines experience might feel your changes are necessary when reading these replies.
10 Likes
This is a very good point Ryan, I didn’t think of it that way!
Will make sure to not call them upgrades from now on 
5 Likes
Agree. I’m here for data points not arguments Collaboration is great. And I’m happy to take criticism when it’s constructive and well reasoned(the reasoning part is very important).
Modification of the machine is in no way criticism of the machine on my part. In fact I think I’ve made it very clear how enamoured I am with some of the design choices.
I see some responses that are very protective of any changes to the machines that divert from stock. I don’t know if that’s a theme inherent to the community or not, but I’d just ask for those feeling that way - if that is your reaction ask why? Is it because you are worried about people with little experience getting out of their depth and wasting money and not seeing the benefits and having a bad experience? I totally respect that.
I don’t think I fall in to that category. And I am happy to explain why. Building machines is what I do for fun and have for many years.
When I do modifications it’s often to achieve multiple goals. If all low riders performed at their best every time and were resilient to things not being perfect then there would be nothing to look at.
However with a machine this finely tuned for budget there will always be elements that ride close to the edge. In the case of lowrider the steppers are definitely being run close to the edge. An example I can give is one I gave above - operator error ran a 4mm single flute through a brad nail at 4mm doc and it skipped steps. Another example is I ran it through a knot in coarse grained 3 layer hardwood ply and it skipped steps. There may be other thing in the system that caused this - I already know the way I have my stepper drivers set up is sub-optimal.
What I want to do is make my machine resilient to things not being perfect. Feeds and speeds a little out of spec - FPT too high or low, surface speeds out of optimal - and for the motion system to be able to handle that.
I’ll refer back to my Fourbie mod for Voron 2.4. It will make just about any 2.4 print consistently at 13-14k acceleration for outer walls. An extremely well tuned and built and maintained boron 2.4 300 CAN do that, but everything must be perfect. A well tuned fourbie CAN do 20k but everything must be perfect. The goal isn’t to do 20k - the goal was to do 14k when the belts are frayed, A and B belts are not well tensioned and equal, the input shaper tune is off. If it can still do those speeds when those things are true then we have widened the window of acceptable settings for any given scenario.
If I can widen the window of acceptable settings on my lowrider and make it more resilient to belt issues or bad bearings - dull end mills or inconsistent stock - and importantly my dumb mistakes then I can spend less time getting a project from conception to reality. I’d like to run the machine fast and not have to jump through a small window of acceptable requirements to do that where possible. And I’m willing to invest some time and a little money to iteratively try and widen that window.
However I’m not here to step on or fracture a community I’ve just joined, and if sharing this journey of exploration here upsets people or somehow damages the community then I’m more than happy to take my findings, thought process and musings back to other discord communities and chats with good friends - no harm no foul.
5 Likes
I whole heartedly agree that beam stiffness and core nod issues will be the next thing to address. The FEA work someone here did on the LR3 beam is very instructive. Definitely going to look to replace my strut plates with some 80x6mm alu flat bar as I seem to be able to get it for $17/m
What to do about the tendency of the core or the strut to nod under load I’m not sure - will do further analysis after installing the stiffer strut plates and see if we can figure out what can be done!
1 Like
This was supposed to read “cheap Amazon power supplies” - I’m using a cheap Amazon power supply right now, on my lowrider and it works but I also think it’s a bad idea. I’m a big proponent of high quality power supplies and good clean power.
It‘s completely fine to test and tune, and I really think it‘s a neat idea (though I still think it does not really help. I can‘t prove my point though, since I haven‘t had any problems with the stock LR and am not planning to do your mod). My problem was with your friend being smug about it and talking down on people who only recently helped him to get his LR to run when he wasn‘t able to get his spindle going and used a wrong ESP32. So it‘s not really surprising I can‘t take his findings for granted. Everything else has been said by Jonathan already.
Well, I can say that Erfan is one of the most capable and diligent builders I’ve ever come across, with incredible troubleshooting skills and working with him on a number of projects has been an honour. I wouldn’t take much from that specific issue, I don’t know many who have successfully got flexihals working on Linux cnc or successfully helped others troubleshoot rs485 issues and printer controller issue remotely. He moves fast which is likely why he reached out here - to people with mnowldge of this specific controller.
Ultimately I think we are looking at yet another example of text communication being flawed. Context and tone don’t always come across as intended and are often interpreted in the worst way possible E.g. the twitter effect.
As far as other building this mod, I’m not even sure yet if I will release the files. While I think I can already see benefit on my end I’ll want multiple data points that show potential, and possibly more people to test. I’m not keen on just putting it out there if people are wasting their time/money.
I’m aware of the effect where people see the new shiny thing and think they need it to be successful, and I’m not super keen to contribute to that.
At the very least I want to confirm in several configurations(e.g not just the premium config Erfan is running) there is clearly no significant disadvantage.
I’m actually pretty sure my assumptions about the minimum torque levels required to perform rapid movements in a system with gear reduction were wrong and that in fact the machine will perform far better than my maths would suggest. Erfans results suggest this, though he is using what might be the best possible motor for this setup.
3 Likes
My parts to go 3:1 and on x and y were supposed to be here tomorrow but the incoming cyclone has slowed things. Assuming there isn’t damage to deal with(should be fine…), hopefully I’ll have time to print install and test next weekend or the weekend after.
2 Likes