I noticed AndyMark has come out with a new compressor as an alternate to the Viair 00090.
The primary claims about it seem to be that it has a higher 1.1 CFM output (legal limit per 2016 R79), and that it has a higher rated duty cycle (15% vs 9% on the Viair).
A quick glance at some of the specifications makes me doubt the value here though.
~13 sec - Claimed time to fill (1) 574ml tank 0-120 psi
Based on the listed specifications, the new one has a little better duty cycle - which may make it a little more reliable. Otherwise it weighs nearly a pound more, takes up more space, and has ~30% less volume output at 100psi - where it matters for topping off during the match. Even the stated fill times for a tank are worse than the Viair.
I like the look of the larger heatsink, but between the weight penalty and the lower ouptut at pressure, I’m not seeing the advantage.
I’ve had both in my hands–the difference in size isn’t that significant (especially compared to the 100%-duty-cycle Viair pump 4901 used in 2015…), though obviously a pound is a pound.
We just got the 1.1 pump page up today, and it’s entirely possible that we had a typo because that is a head-scratcher. I’ve got emails in to investigate the concerns you bring up (which are absolutely fair ones!).
I was excited about it, and then my student Ruben pointed out the same thing.
We charted the performance data for comparison, the cross-over point is at 30psi. I don’t think our system pressure has ever been that low without leaks involved.
(Having trouble uploading the image…)
2502 used a 1.1 CFM pump this year (Stronghold) and without it we would not have been as functional as we were. We sucked through a lot of air with our robot, and the extra .22 cfm from the .88 cfm that we tried before saved us a lot of hassle, and a number of issues we were having were avoidable.
so IMO yes, they are that much better
But when they say they get hot, they get really $@#$@#$@#$@# hot. (we melted some pneumatic tubing)
On this Andymark page, you can find specs and pricing on the most common FRC legal compressors out there. Teams can send in recommendations to Andymark about one that they use that is not featured on the site, and Andymark can add to the page or even sell it if they wish. There would be options to compare fill times for given sets of volumes across multiple compressors. Want more information? You can click on any compressor listed and see pricing, recommended vendors, and performance sheets. While we are at it, how about some nice looking performance graphs for visuals too.
Agreed, we have a 98c in the shop and the sticker shows the 12v rating is legal, but the website data at 13.8v showing its over the 1.1. I would love to see FRC increase the cfm rating or at least give us a limit at 13.8v so we can look at the other Viairs.
Perhaps if AM did testing on the Viair product line at 12v we could site that to justify compressor selection?
There’s actually pictures of it here on CD I think if you browse the photos section a while. We did something similar to this on our 2011 bot, but we just mounted one of the 120mm fans next to the compressor so that air blew across it. The improvement in heat dissipation was actually pretty drastic: Test Results
That said, even if an external fan can help keep a compressor tolerably cool, if you’re running a low duty cycle compressor almost constantly throughout a match, you’re still using it beyond it’s designed limits. I would also suspect that you will still loose efficiency over prolonged use this way due to internal heating that isn’t adequately dissipated by external fans.
You can actually think about it a bit like how we think of motors. The 775pro motor is a great, strong little motor, but too much heat (generally generated from stalling the motor or extended use) will cause it too loose efficiency and burn out, and while blowing a fan across it might help, it won’t stop the problem. Alternatively, a CIM motor can run continuously even after substantial heat build-up (even in stall condition) without a fan with little to no damage due to its increased thermal mass.
Essentially <15% duty cycle compressors are 775pros and ~100% duty cycle compressors are CIMs.
We just put 4 bolts and zip tied around them. Sounds like a kludge , but it works. Best to have the fan on top. That’s the hottest part. This would be a good CAD - 3d printing project to make a clip on fan shroud. Post to thingiverse.
Just got off the phone with andymark… they are testing the 1.1 and the older compressor today side by side type test @ 12v so teams will have better numbers to base their decision
The problem with the claims of flow rates is that they typically ignore the critical element of temperature. We do not always operate at ambient temp. Our robot compressors are usually too hot to touch for back-to-back matches, meaning it is tougher to compress air than they typical duty cycle compressor. I suspect that the AndyMark compressor, with a longer duty cycle, may do better in a typical FRC environment given its better duty cycle, but I’d like to see some numbers.
This how we do our final pneumatics tests to know if we need another tank, and it seems to work for that purpose. I don’t know what it’ll tell us about different compressors though:
Fill up 4 tanks -> 5 minute break -> run testbed for 2 minutes -> Record Pressure
-> 2 minute break -> fill up 4 tanks -> 2 minute break -> run testbed for 2 minutes -> Record Pressure
-> 2 minute break -> fill up 4 tanks -> 2 minute break -> run testbed for 2 minutes -> Record Pressure
We’ve updated both the Viair and 1.1 Pump pages with our real-world testing.
This testing was done in our shop, with a regulated 12V power supply. Same tubing, same tanks, same gauge, same everything-as-much-as-we-can-humanly-measure.
Viair Pump:
Fill Rate for Number of Clippard tanks (574 ml) to fill from 0 to 110 psi:
1 Tank = 34 Seconds
2 Tanks = 64 Seconds
3 Tanks = 103 Seconds
4 Tanks = 140 Seconds
5 Tanks = 176 Seconds
1.1 Pump:
Fill Rate for Number of Clippard tanks (574 ml) to fill from 0 to 110 psi:
1 Tank = 25 seconds
2 Tanks = 45 seconds
3 Tanks = 72 seconds
4 Tanks = 95 seconds
5 Tanks = 122 seconds
We’re working on converting to CFM for each 10-PSI breakpoint, but I timed both 5-tank tests and the 1.1 Pump was ahead on each one of them. Hopefully, this answers the sixty-four-dollar question.
