Is there any significant benefit or harm for putting 2+ pumps in the same loop? Or would it be better to create two separate loops. This is purely a theoretical question, as i have no specific pumps or hardware in mind but would assume at least 2 radiators, 2 pumps (going in the same direction of course), and a gpu and cpu but only 1 reservoir. I realize if you go above ~50psi (probably lower for some things) a user can run the risk of breaking/blowing connections, aside from that any other possible harm?
I'd have two separate loops. The thought of coolant entering the second pump under pressure seems wrong for some reason. I'm sure it would also have negligible performance increase. In fact, something nagging at the back of my mind says it would be detrimental, although I can't remember why just now.
iirc - because the pumps would be pumping at the same rate, it creates no performance increase - the only reason you would need 2+ pumps in a loop is if the loop is really long (e.g. 5 meters? I dunno - something not normal ) In such a scenario, you would have one at one side of the loop, and the other at the other end (to balance the load). There was a post about this recently iirc - give it a quick search if you fancy the details
It's a pig to get water to the second pump and prime it. That's one experiment I won't try again. Apart from maintaining the pressure in a long loop some have them for redundancy but if you use the BIOS thermal shutdown options or Core Temps overheat protection I'm not sure why you'd need it.
Alternativley, you could put the two pumps in paralell, which i'm pretty sure would double the flow rate... ie: Code: /-->--Pump-->--\ >-->- ->--> \-->--Pump-->--/
In applications where you're limited by space and can't go for a dual loop, two pumps in series double the head pressure while flowrate stays constant. This is from what I've read, and it makes sense.
I believe running two pumps in serial increases pressure while running two pumps in parallel increases flow. Ninja edit: Found an old post of mine with this graph I posted in another thread. You should find it useful.
The 1 major advantage of going parallel is that if a pump fails the other 1 will keep the flow going so minimizing any chance of a melt down
You could say the same for series, only time that wouldn't be true is if the pump seized/died in such a way that the flow was extremely restricted or blocked. I think poor maintenance and organism growth are probably far greater causes of blockage than failing components. Well aside from the failing user component.
That top puts the two pumps in series, and adds pressure like the graph I linked to illustrated. Even though it results in some great performance numbers, I'd rather run two pumps separately and link them with tubing instead of a top like that. The two sharp 90 degree corners in between the first and second pump will give performance that's just a little bit worse than running them in series with tubing, and a top like that is probably going to be expensive. Of course, if two pumps with stock tops in series can't outperform that dual top, buying the dual top would probably be a better value than hooking up two pumps with their own aftermarket tops.
The EK DDC top fixes that - inlet is from on top of the first pump, and goes sideways and slightly upwards to the top of the second pump. I can't find a review, but it's one of the best dual-pump tops out there. Here's a picture:
I was more refering to the text as Skinnee explains the benefits of two pumps quite well The XSPC top is actually pretty reasonable - £20 at most places. Given the relationship between flow rate, pressure and restiction, in restrictive loops, you should see an increase in flow rate too as well as pressure.
