ASK THE EXPERTS: Microhydro Flow

Measuring flow, one of the two key components of determining available hydro power.

My family is considering installing a microhydro-electric system to power our remote cabin. We plan to use a 3-inch penstock (approximately 900 feet long) with 85 feet of head. Can a theoretical flow rate be calculated from these figures? The upper end will be submerged in a pond/creek and the water returned to the same creek.

We would prefer a batteryless AC system. This is a joint project with a neighbor. His part of the deal is installing the penstock; ours is selecting a hydro turbine. I keep telling him that the penstock has to come first so we can measure the pressure and flow. He wants me to “just buy a turbine” and says we’ll make it work. Can these figures be estimated given the information above, or should we install the line first and measure directly?

Rob Day • via email

Measuring head and flow are critical design jobs that should be done before any equipment is specified or purchased. Deciding on the penstock diameter without solid head and flow measurements could lead to wasted money or lost power. While both these measurements are more easily done with an installed pipeline—a pressure gauge for head, and a bucket and stopwatch for flow—you really need to know them in advance to choose the right pipe and size. Buy a pipe that’s too small and you could lose most or all of your power to friction loss; buy one that’s too large and you will waste money.

Once you have determined head, flow, and pipe length, consult friction-loss tables for the pipe type you choose. Next, you need to decide if the pressure (head) loss is acceptable. People typically accept a head loss between 15% and 25%. I personally favor lower losses, since the pipe cost is a one-time purchase, and the head loss is forever.

So instead of starting by specifying a pipe diameter, get some data and examine possible scenarios: “With 3-inch pipe, we’ll spend $X,XXX and lose YY% of the potential from friction loss. With 4-inch pipe, we’ll spend $X,XXX but only lose XX% of the potential.”

See our many microhydro articles for more on this and other aspects of system design. For instructions on how to measure your hydro resource, see

If you’re providing electricity for a conventional house or two, I’m doubtful, with that head and pipe size, that you’d get the power necessary for a stand-alone AC system. Your system would need to produce more than 2 kW of power to provide for the loads of a typical house. Stand-alone AC systems are limited in peak power, which equates to how many loads you can run simultaneously. Again, do the math first, and decide what system configuration is most appropriate.

Ian Woofenden • Home Power senior editor

Comments (3)

kenbell48's picture

I have a very similar system to what you describe wanting to install. 90 feet of head and a 2000 foot penstock. But I am using 6" PVC pipe so there is no head loss. I am using two Harris brushless alternators to generate 40 amps (20 each). I have been doing this for over 30 years and was lucky in that I built a diversion, installed the pipe, alternators, inverters and batteries and the whole thing worked properly with no adjustments. You won't be able to go batteryless off grid because you need the momentary amps they provide to start inductive motors. I run two alternators (one can put out enough power) so I have redundancy and can run them each at a lower output which means less heat generation. At 20 amps the case is about 100F degrees and the interior (stator) is about 140F. They are rated for 100 amps but I have run an individual alternator at 60 amps and it was hot enough to smell. I'm using 4 nozzle manifolds and a total of 120 gallons per minute. I have had these running for over 10 years with no maintenance and know of one with over 12 years and no problems. They have bronze runners so you have to have a settling tank at your inlet to eliminate sediment or buy a new $300 runner each year. Since I put in a 30' long 4' deep 6' wide tank at the inlet there has been no discernible wear. The tank has a 6" pipe for a drain so sediment flushing is kind of like flushing a toilet, just pull the plug. Make sure you have a vent near any pipe inlet or you will collapse the pipe if the inlet ever gets plugged.

Malcolm.drake's picture

A really important issue with hydropower is you don’t want to design your system around ONE flow measurement.

Virtually all waterways vary in flow rate from season to season, and even hour to hour.

You need to make LOTS of measurements, ideally over a time frame of at least a year.

If you plot your gpm/cfs over time, you can get an estimate of how long a time period you’re likely to have different flows from week to week or month to month.

Based on these data, you will be able to judge whether it’s worth a larger pipe diameter (and turbine, for that matter) in order to harvest power for time periods where the flow is higher, but less frequent.

I also suggest you make sure your penstock's ENTRY is designed to make sure you don’t limit the flow through the inflow to a level that’s less than the rate your penstock is otherwise able to carry at your desired friction loss rate.

Be aware that, while hydro is awesome, in that it produces power all night! And on rainy days, you will be required to keep your penstock's intake clear ion debris. In your case, this means walking 900 feet and up 85 feet.

Maybe a longer or shorter walk, depending on your home's location, really I’ve to your inlet location.

Have fun! I’m very envious!

by the way, there are innumerable pipe friction loss tables online, which will give quite accurate loss rates, typically in PSI/100feet of pipe. The good sites actually have different tables for different types of pipe, e.g schedule 40, schedule 80 pvc; galvanized steel (for those with more money than god) copper, pokypip. Etc

Irvin Kanode's picture

Couldn't they come up with a flow estimate by installing say 10' of pipe at their planned spot. They could test various depths, spots and in different seasons. That might change the starting location, thus the amount of pipe they'll need to order.

I agree the first step needs to be calculating the amount of power they want to generate!

Show or Hide All Comments