4 ms·
Short answer: For home use - no. Price is to high, subscription fee is show stopper for me. Long answer: Home use Why will be home owner or tenant bothered
by stakent 17y ago
Short answer:
For home use - no. Price is to high, subscription fee is show stopper for me.
Long answer:
Home use
Why will be home owner or tenant bothered by power use?
First group - people bothered by energy cost. The calculation is simple: the cost of hardware and accumulated subscription fees has to be recouped by energy cost savings in one year.
Yearly subscriptions fee $120, hardware cost, say, $200 gives $320 cost of the first year use of the service.
Energy cost $0.1 per kWh so we have to save at 3200 kWh per year to break out our investment.
3200 kWh per year means, if power consumption is constant, reducing energy draw by about 0.37 kW.
Doable, but is it easy?
Replace incandescent light bulbs by more energy efficient compact fluorescent light bulbs or led light bulbs. Done.
Is cost sensitive home owner the target group? He can be if the price is right. Read - low enough. So - mass production, lower cost per hardware unit, lot of subscriptions sold on ongoing basis. Well, there is one problem. Up front spend on hardware batch. And second one. How to reach these cost sensitive home owners? Marketing. Which means more money up front. Which means more risk. By the way, that means competing on price what is not the best idea for someone starting with limited funds.
Second group - people bothered by Earth future, global warming.
Each saved kilowatt hour means less about 2.3 lb (1 kg) of CO2 emitted to atmosphere. Saving 3 kWh daily during a year means about 1000 kg less CO2 emitted to atmosphere.
How reach people bothered by Earth future? It looks little bit easier. Is it easy enough?
Third group - early adopters, geeks looking for new gadget.
The new cool gadget has a little drawback. It is clipped on the electrical wires, presumably in the hidden electrical box. How to show it to other geeks? There is website with fancy graphs, twitter integration, facebook integration and, of course, iPhone application. And some game mechanics, a little competition: who has smaller energy consumption, who is greener so can be awarded by, say, more images of green leaves. Or bigger tree image. Or blue ribbon. Or ...
This group looks easier to target. It has another nice feature. Virtually all of them can be called linkerati, which leads to some nice side effects. Links, SEO, boost in visibility.
Industrial and commercial use
One 10 kW motor working all year long uses 86400 kWh. Let's assume energy costs $0.05 per kWh. Yearly cost: $4320.
Completely different game.
Industrial environment is harsh. Electromagnetic noise level is really high despite all countermeasures. Current in wires is hardly sinusoidal because of all theses nonlinear power receivers (eg. frequency converters).
What savings are possible in industrial environment?
Use less energy. Use GridSpy for ad hoc power monitoring during machine tunning. Possible for air conditioning, refrigeration, making compressed air, pumping.
Use cheaper energy - prices vary during the day and week. Locating low hanging fruit.
Do not pay for use more power than contracted. Add an ability to alarm someone if there is to much power drawn, allow to change the formula for power calculation. Example: power calculated as 15 minutes average or an 1 hour average.
Do not pay for too low power factor. Power factor monitoring and alarm someone.
Another beast is data reflecting production put on the internet. I know, passwords, vendor commitment to security etc. but ... Btw. this concern exists for home use too.
Despite that all energy conservation is an interesting field.
- gridspy 17y agoWow. Thanks for the through reply. The price is high, as you mention - once we have an established cashflow with industrial clients and proven technology it will be possible to gradually migrate into the much more cost sensitive residential sector. To enter that sector from the get-go seems like a huge mistake to me. Sure, the subscription fee is a showstopper. I'd like to have several levels of subscription: Free: Live data online, 2 days moderate resolution (sample per minute) / 1 week low resolution (sample per 15 minutes) history. Simple embeddable widgets / twitter usage notifications, etc. $6 /mo ($10 NZD): As free, 1 year moderate resolution, 10 years low resolution. Add Webhooks, SMS, email, alarms, budgets, etc. Premium: (this is for industrial, price TBD) 1 month ultra-high resolution (1 second samples), unlimited storage for other data. Most home users would be satisfied by that free option. As for your customers: First Group (price sensitive domestic): I agree. I'm not even attempting to reach this market right now. Second Group (CO^2 sensitive): I think that these users will be most interested in fostering power saving competitions within and between different businesses. Most of them have probably optimised the hell out of their domestic power usage. They might be interested in leasing our system for a month to validate their power saving measures. If they have off-grid power, our solar integration might interest them. Third Group (Early adopters): The device itself isn't that sexy. As you mention, it is the online dashboard that makes it good. To reach this market segment, it is important that we have great integration with a wide range of other tech toys. I'd also like to make it really easy to hack together neat things using our live data via webhooks and the API. I'm looking forward to the linkerati aspect. Hopefully we can foster a situation like Twitter - people linking proudly and prominently to their GridSpy Dashboard to prove their Green/Geek cred. Industrial Environment: Good point about the non-linearities in industrial circuits. Gridspy could be used to hunt down the poor power characteristics that cause industrials to be penalised by the power supplier. Because of our realtime data acquisition, we could highlight the power factor of motors during startup, running or shutdown. This would help electricians locate and fix key power issues. As you mention we could be used to help cut power usage during certain times of day. We could also perform load-shedding - turn off AC while many motors spinning up for instance. The peak power usage during the day needs to be kept down to reduce power cost. Finally alarms and budgets could be used to attract attention to important issues. All in all I'm pretty excited about our opportunities in the industrial sector. I think that the Residential sector is too disinterested and price sensitive to enjoy GridSpy.