Mobile coffee cart design question - hybrid power supply

megscoffeecart · August 2014

Hello! I'm a newbie and at a complete impasse with my design. Please forgive me if there is a more appropriate place for this request but I came across a few similar mobile food threads here that were very informative, but also more complex than both my needs and current knowledge base. Basically, I'm afraid that I'm overlooking major components/factors and would love an expert opinion.

I am designing a propane + battery power system for running a bike coffee cart. It is a compact, front-loaded cargo bike frame with a cabinet to be built on top. My goal is to operate with as small an environmental footprint as possible. However, I cannot incorporate solar panels due to the size and weight limits of the cart frame (500lbs max, 38"x26" platform) and the location at which I'm vending does not allow generators. I will have access to shore power each night to charge batteries and it's possible that the site at which I operate will have shore power (the vending program is a bit funky and I won't know for sure until well after the cart build has to be completed).

My plan is to use AGM deep-cycle batteries for some of the appliances and use an LPG-fuelled tankless water heater for the hand-washing sink to reduce the battery-powered load as much as possible. I'll intermittently draw for 7-8 hrs a day. The only appliance requiring steady power is the water tower. The water pump and coffee grinder are on-demand usage so I can make it a convention to never run them together (unlikely anyways) and gain a little leeway on the watts total .

Power needs are:
FloJet water pump - 12VDC - 53 watts - 1.9 amp @ 10psi (3.5amps max)
Coffee grinder - 110VAC - 154 watts - 1.4 Amp (http://www.baratza.com/flat-burr-grinders/vario-w-grinder/)
Water tower - 120VAC - 1800 watts - 14 Amps (http://www.bunn.com/pdfs/specsheets/e1.pdf)
Total = 2003 watts/ 17.5-20 amps

Does that mean that I need a battery that supplies 17-20.5 amps for 7-8 hrs?

If so, would something like this battery work with an inverter: http://korrlighting.com.au/battery-160ah-agm.html? (12V 160ah)

Is there any advantage to splitting power into 2 batteries - one house battery to bear the steady, high draw of the tower and one starter battery to handle the short bursts of the water pump + grinder?

I recognize that the water tower is blowing out the power needs, but I have to have a steady supply of 200 degree water and have not found a more efficient supply than this type of tower. Open to suggestions if anyone has one!

Thanks!
Meg

bill von novak · August 2014

Re: Mobile coffee cart design question - hybrid power supply

megscoffeecart wrote: »

I recognize that the water tower is blowing out the power needs, but I have to have a steady supply of 200 degree water and have not found a more efficient supply than this type of tower.

Hmm, there are plenty of propane on-demand heaters; I would use one of them. If temperature variations worry you add a pump to recirculate water. This will not be doable with batteries otherwise. (As a reference, you'd need 14 kilowatt-hours worth of batteries, which means 20 GC2 sized batteries, which means about 700 pounds of batteries.)

I'd make the switch to propane then get a small inverter for the grinder. You'll have to spend a day with a timer to figure out how often the grinder and pump are actually needed to get good power numbers. Conservatively this would be under a kilowatt-hour. You might want to look at lithium ion since this is a small amount of energy and lithium ion is much lighter than lead acid. In fact you might want to consider an off the shelf electric bike conversion kit - the motor will help move all that weight and the battery will provide additional power for the other loads, and all the charging, battery protection etc will already be done for you.

BB. · August 2014

Re: Mobile coffee cart design question - hybrid power supply

Welcome to the forum Meg.

My first suggestion... Do not work in "Amps" and Amp*Hours. It is confusing the issues for you.

When working our your total power needs, do it in Watts and Watt*Hours. Watts is a "complete" discrption of the power you are using. Amps is a partial descrption and you need to keep track of what Voltage those amps are at.

For example, your water pump:

12 volts * 1.9 amps = 22.8 Watts

Your hot water tank:

120 Volts * 14 amps = 1,680 Watts

Your total power usage of 2,003 Watts (peak power):

2,003 watts / 120 VAC = 16.8 amps @ 120 VAC

If you where to run that on a 12 volt battery bank:

2,003 Watts * 1/0.85 AC inverter eff * 1/12 volt nominal = 196 Amps @ 12 volt

That is around 2-3x what a typical US car engine takes to crank... And if you have every had an engine problem, you can crank for about 5-10 minutes and then the battery goes dead.

What is really missing is the Hours of electrical use here... 2,000 Watts * 8 hour day--That uses 2x more electricity than my entire suburban home uses in a day with a family of 4.

Realistically, your loads (including the hot water tank), are going to be much less than 8 hours per day running at rated power... But until you have better information, we won't know what the "real" numbers are.

In the end, propane and/or a well insulated water tank (filled with pre-heated hot water at the beginning of day?) is going to be your friend here. At best, the amount of energy that your battery bank would supply (that you can fit in your mobile cart) will supply your lights, water pump, and your coffee grinder.

You have several ways to proceed here... In general, the best is for you to measure you actual loads and design the battery bank to support them. Use a Kill-a-Watt type meter for your AC loads and a DC Watt*Hour meter for the DC loads.

You can also approach it from another side--How much battery bank can you reasonably carry-Then figure out how much energy budget you have with that bank. Next, figure out out what you can drive with that battery bank (and your budget for appliances--Conservation will be your friend here--It is almost always cheaper to spend on conservation vs trying to generate power--especially from a battery bank).

With batteries--You probably a great candidate for:

GBS LiFePo4 20 and 40ah batts

or similar batteries... They are incredibly light weight, can supply lots of surge current, can be recharged quickly, and have relatively high energy density (small battery, lots of stored power compared to a lead acid battery). AGMs would be nice for your needs--But they are only a "little bit" better than typical flooded cell lead acid batteries regarding energy storage/production. AGMs are 2x the price of flooded cell... LiFePo4 batteries are several times more expensive again than AGM--But with a longer life and smaller/lighter battery pack--It may well be worth it for you.

-Bill

PNjunction · August 2014

Re: Mobile coffee cart design question - hybrid power supply

B.B. nailed it. You've got to get a handle on how much power you actually draw first, and not rely solely on the spec sheets.

For instance, I doubt the water tower runs 1800 watts continuously. That is probably the peak power used when the heater elements come online, and the ones I've been around like that don't run the heater continously, only on short demand, or when the temp gauge indicates that the existing water has fallen below a certain temperature level.

The Kill-A-Watt meter would allow you to get a handle on that. Simulate it at at work/home and measure how much power is used over your planned period of operation with frequent cup filling activities.

As for the GBS (or other Lifepo4 batteries) - they are a great candidate, BUT even when you factor in the advantages, and while upfront costs are coming down all the time, a mistake in your power budget will STILL be a very costly mistake. Before you go lifepo4, you may want to do real-world testing or initial operations with your original thought of deep-cycle agm's. Odyssey / Lifeline would be a good choice for huge initial surge currents with less voltage drag - but even those are costly. If those work out well, THEN perhaps consider lifepo4, but for just starting out you need to tread carefully.

The Kill-A-Watt is worth it's weight in gold as everything hinges on how much power you realistically draw.

megscoffeecart · August 2014

Re: Mobile coffee cart design question - hybrid power supply

Thanks Bill and Bill for your responses! I knew it couldn't be so simple as having one serious battery but the volt->watt->amp relationship is one I keep screwing up in calculations.

It seems my best best would be to use a propane burner and a kettle for getting the high-temp hot water I need and focus on powering the grinder & water pump with a battery. I think you are both right about using a timer to figure out my actual load and working backwards.

I hadn't looked at LiFePo4 batteries at all but they do look like a wonderfully lighter/smaller option. Can you walk me through how to calculate what I'd need in a LiFePo4 battery pack? The site I'm looking at has 3.2V cells that are distributed individually or combined into 12.8V packs. So after I time my load with a meter, how do I work backward to design the battery pack? Or, work inversely as you suggest BB Bill, if I can fit & afford an 12.8V 40ah LiFePo4 battery, how can I tell what I'm able to drive off of that? Forgive me for the rudimentary approach, this is all new territory for me.

Also, since I'm combining AC and DC appliances with dramatically different power profiles, would it be more energy efficient to have 2 batteries, one for the grinder and one for the water pump?

megscoffeecart · August 2014

Re: Mobile coffee cart design question - hybrid power supply

Thanks PNjunction! You're definitely right about the tower - 1800w is their max output. The specs say somewhere closer to 1300 for running but it may well vary in real life. Speaking of, I hear you loud and clear about the real world testing. It's time to start setting up operations an seeing how they perform against the spec #'s I'm using for calculations.

PNjunction · August 2014

Re: Mobile coffee cart design question - hybrid power supply

Wow - 1300w average?

To put it into perspective, with lifepo4 if you ran a 12v bank of 400ah Calb's driving inverters, that would allow you about 3.5hours of operation daily just for the water tower, that would cost about $3K, or $6K for 7 hours daily operation if you doubled up to 800ah.

Making a mistake in your power budget with lifepo4 could stop you in your tracks - although even with high-quality agm's this wouldn't be inconsequential either.

Blackcherry04 · August 2014

Re: Mobile coffee cart design question - hybrid power supply

It's hard to design it for you not knowing all the details. You can get a 6 gal aluminum propane RV water heater. I'd get the Bunn thing if you like it and rip the heater and tank out or you can make one to your own liking. You'll never be able to power 1300 - 1800 watts on batteries realistically. Pre - heat at night and that would limit the propane requirement for the day. You could use a 5 gallon ( or what you think you'll use for the day ) storage tank to keep the hot water filled in a displacement system, otherwise you'll have to pump hot water or use a small air compressor to push it.

Just some thoughts.

BB. · August 2014

Re: Mobile coffee cart design question - hybrid power supply

You ideally want to start with the loads and measure them... Then work towards designing a battery bank to support those loads.

Just to give you a very rough idea of what 6 volt 220 AH (Golf cart) lead acid/AGM battery bank would look like:

1,300 Watt average load * 8 hour day = 10,400 Watt*Hours per shift (or 10.4 kWH as your power company bills power)
10,400 Watt*Hours * 1/0.85 AC inverter efficiency * 1/0.50 maximum bank discharge * 1/12 volt battery bank = 2,039 AH @ 12 volts
2,039 AH / 220 AH per pair of golf cart batteries = 9.2 ~ 9 pairs of 6 volt batteries = ~18x 6 volt 220 AH batteries.

Trojan T105 6 Volt, 225 AH Deep Cycle Battery

$150 per battery * 18 batteries = $2,700 worth of batteries
18 batteries * 62 lbs per battery = 1,116 lbs of batteries

The battery bank could be a 1/3'rd smaller if you use industrial deep cycle batteries (floor sweepers, etc.). In any case, your battery bank will probably last ~3-5 years if you are lucky.

That same amount of power could be obtained from a Honda eu2000i genset (~50 lbs for $1,000) and ~2 gallons of gasoline. Just to give you an appreciation for our much maligned fossil fuel economy.

-Bill

bill von novak · August 2014

Re: Mobile coffee cart design question - hybrid power supply

megscoffeecart wrote: »

I hadn't looked at LiFePo4 batteries at all but they do look like a wonderfully lighter/smaller option. Can you walk me through how to calculate what I'd need in a LiFePo4 battery pack? The site I'm looking at has 3.2V cells that are distributed individually or combined into 12.8V packs. So after I time my load with a meter, how do I work backward to design the battery pack? Or, work inversely as you suggest BB Bill, if I can fit & afford an 12.8V 40ah LiFePo4 battery, how can I tell what I'm able to drive off of that? Forgive me for the rudimentary approach, this is all new territory for me.

First off go with a discrete battery instead of cells. Here are a few options:

http://www.amazon.com/Lithium-Phosphate-LiFePO4-Rechargeable-Battery/dp/B00E4WTSWQ
http://www.batteryspace.com/custom-lifepo4-26650-battery-12-8v-29-7ah-380wh-4a-rate-with-30a-pcm.aspx

You work with watt-hours. A 12 volt, 30 amp hour battery contains about 360 watt-hours. A good rule of thumb is to plan for using no more than 2/3 of that (to give yourself some margin.) Lithium ions let you discharge pretty deeply without ill effects, but the 2/3 will keep you from unexpected power loss.

So let's say that each of your devices runs for 2 hours total out of 8 hours. That's 400 watt-hours. With an inverter you get some losses so that means you need 440 watt-hours. So two 30 amp hour batteries will give you 720 watt-hours, which is well above your need for 440 watt-hours a day.

Or let's say they only run for one hour. Now you can live with one 30 amp hour battery.

Also, since I'm combining AC and DC appliances with dramatically different power profiles, would it be more energy efficient to have 2 batteries, one for the grinder and one for the water pump?

No, not worth it. Batteries are expensive so having one serve both loads is going to be more economical overall.

zoneblue · August 2014

Re: Mobile coffee cart design question - hybrid power supply

A battery powered bicycle coffee cart, you say? Without even looking at your numbers, (not sure you need to), id say this project is a non starter. A coffee cart uses quite a bit of energy and supplying it by battery (and maybe even propane) is going to be space and weight prohibitive. Id be reaching for a shore power cord for a setup like that.

But hey what the heck, lets poke some numbers at it....

Lets say:
- that a viable coffee cart requires, um, what the heck, 1000 cups of coffee per day, and;
- that each cup needs, 1/4 litre x 90*C x 4200J/kgC /3600s ... or 26Wh per cup of energy to heat the water,
- then double it for system derating, pumping, grinding, brewing, and sundry
- so thats 50kWh per days operation
- the energy density of lead acid is around 50 Wh/kg, so thats 1000kg of lead
- the energy desnity of LiFePO4 is around 120 Wh/kg, so thats about 400kg of lithium
- the energy density of propane is about 5000Wh/kg, (+allow a third for the tank), or about 9kg of propane.

Rough numbers you understand....

mike95490 · August 2014

Re: Mobile coffee cart design question - hybrid power supply

You have to do most of that heating with propane, you can't carry enough batteries to go all day on electric. What about getting a Dewar (50L thermos jug) to hold preheated water for the day, and then you are just raising hot water to boiling, not going from cold to hot.

You could develop your own water heater, with thermostat, SSR, and a 24V electric water heater, fill the heater with hot water, plug it in overnight, and you are set in the AM
(do you have to carry your water in the cart, or are you plumbed into a hose ?)

Do you have shade to work in ? 60lbs of solar & rack would give you 200w of PV on a sunny day, and shade to stand in. Foul weather, when you will sell more flavored hot water, you have to rely on stored power. But having a solar shade, a couple of electric meters, and you could capture the green crowd.

PNjunction · August 2014

Re: Mobile coffee cart design question - hybrid power supply

megscoffeecart wrote: »

.. Or, work inversely as you suggest BB Bill, if I can fit & afford an 12.8V 40ah LiFePo4 battery, how can I tell what I'm able to drive off of that? Forgive me for the rudimentary approach, this is all new territory for me...

Um, we really need a lifepo4 section since this tends to hijack threads ..

I'm glad you put it in those words of "driving". Here we are using lifepo4 for storage, or house-bank use, NOT for driving electric vehicles, RC model airplanes, wheelchairs and the like with HUGE surge current capability, and regenerative feedback. THAT kind of use calls for cylindrical cells, such as A123, Headway and the like.

The cylindrical ones that Bill pointed out are usually pawned off on the unknowledgeable consumer looking for a "drop in replacement", and are paying through the nose for an EV / RC modeling, motive-power application, NOT a storage application like we use. The internal cells usually go to the lowest bidder and can have a questionable heritage. You don't know unless you rip them apart. Part of the internal bms system is to keep lower-quality cells in line with each other. Some of the 12v look-alikes have many, many small cylindrical cells inside to build up capacity, which adds to the problem.

A more appropriate choice would be the "prismatic" type which does not contain cylindrical cells, and in a 12v application only has 4 cells in series, which are much easier to keep in balance, are of generally higher quality, and on their own can easily do 1C charge/discharge, with 3C bursts. For the 40ah battery we are using as an example, that means 40a (or 480 watts continuous). Having negligible Peukert and the ability to go to 80% DOD with full performance at 2K cycles even at that repetetive DOD makes them attractive. They are NOT the universal panacea some may think of them, as one should always fit the battery to the application, be it flooded or agm where those make more sense.

http://www.batteryspace.com/lifepo4-prismatic-battery-12-8v-40ah-512wh-10c-rate---un38-3-passed-3-2vx4-dgr.aspx

*** Please note that the charging voltages listed by this retailer are too high and too low for normal operations!! ***

I use this along with a 20ah version, each of which has NO BALANCING circuits. Other than a high voltage and low voltage disconnect, they are not needed in a relatively low-current, low voltage application, where balance is maintained once verified. Only the EV and RC modeling crowds want to foist them upon us, instead of using proper charge procedures - which means not running them wall-to-wall when charging or discharging.

Much of the disinformation comes from outside the house-bank or storage application field, complicating a very simple setup. It can be well-intentioned, but in many cases they have ZERO personal experience other than EV/RC modeling motive power use - or just simply repeating what they have read. Others have a financial interest in selling balancing circuits to the house-power crowd, or confusing the multitide of various li-ion chemistry characteristics, of which lifepo4 being the least energy-dense is the safest to use.

megscoffeecart · August 2014

Re: Mobile coffee cart design question - hybrid power supply

Thanks to you all for the answers & patience! I've got my marching orders to do some testing on the actual outputs of the various appliances and what my average daily usage will be.

My updated plan is still to use a hybrid of battery and propane but to allocate both the hand sink water heater and the coffee water heating to propane (eliminating the water tower and using a 2-burner range + kettles) and get the appropriate AGM or LiFePo4 battery for the grinder and pump. I've seen other mobile food folks use propane for getting the hot-hot water but I had been looking for a way to achieve the same ends with a different solution.

Bill Von, as you rightly point out, it'll likely be between 1-2 hrs of actual usage across the 7-8hr day since they're each used in 30-90 second intervals. The 1-2 12V 30Ah batteries may be sufficient for just those 2 appliances if my testing is consistent w/ your estimates.

Zoneblue, I hear ya! It seems impossible but there are a few people running them across the country with some mix of electric & propane. I haven't been able to get details from those folks on their set-ups, hence the convoluted line of questioning here.

megscoffeecart · August 2014

Re: Mobile coffee cart design question - hybrid power supply

mike95490 wrote: »

Do you have shade to work in ? 60lbs of solar & rack would give you 200w of PV on a sunny day, and shade to stand in. Foul weather, when you will sell more flavored hot water, you have to rely on stored power. But having a solar shade, a couple of electric meters, and you could capture the green crowd.

I love this idea but my bike frame has a 500lb max and only 38x26 platform so I don't think I can draw enough energy to do much with that limited space and can't devote 60lbs to just a portion of the power. I have found some small solar powered elements to incorporate like string LED lights and cell/tablet charger to "capture the green crowd" which is a great comment since it's certainly a part of the cart's identity.

I will be carting all water on the bike, which is a limitation. However, I'll start from a commissary kitchen, so I can start with fairly warm water in an insulated vessel like you suggest, so coming to 200-205 degrees won't be as big of an energy/time obstacle.

Blackcherry04 · August 2014

Re: Mobile coffee cart design question - hybrid power supply

Here you go, hide this puppy in your cart and pipe the water up to your tower.

http://www.walmart.com/ip/8187723?wmlspartner=wlpa&adid=22222222227009265342&wl0=&wl1=g&wl2=c&wl3=41873436070&wl4=&wl5=pla&wl6=39976845430&veh=sem

westbranch · August 2014

Re: Mobile coffee cart design question - hybrid power supply

megscoffeecart wrote: »

but my bike frame has a 500lb max and only 38x26 platform

Are you able to add a small, light towable trailer? Use that for some of the weight.

http://www.shop.ca/c/homcom-bicycle-cargo-trailer-black-red-5664-0005R-24242653?gclid=CJ6xz--I_b8CFUEbaQodViEA2g

A high end one, http://www.mec.ca/product/4015-395/bob-yak-bicycle-trailer/

OR get one custom made...

megscoffeecart · August 2014

Re: Mobile coffee cart design question - hybrid power supply

westbranch wrote: »

Are you able to add a small, light towable trailer? Use that for some of the weight.

I'd been thinking about this option quite a bit - it'd surely help to increase the weight capacity, but adding that much weight may push it outside of my biking abilities. Because it's a front-loaded and heavy duty trike, it's not an electric-assist candidate: it'd need essentially a motorcycle brake to be able to stop the front weight once it gets cranking. I've reached out to the cart manufacturer for their input since they've made innumerable customizations on these and might have help.

Thanks for the great idea!

The coleman water heater looks nice but won't get hot enough (near/+ 200 deg) for my purposes. I'm planning now to use 2 propane burners and stainless steel kettles to create the correct temp in smaller batches. It's more labor intensive, but it's critical to get the very hot water because I'm making pour over coffee to order.

westbranch · August 2014

Re: Mobile coffee cart design question - hybrid power supply

Be sure to ask about lareg rotor disk brakes for your unnit...

eg http://www.ebay.ca/itm/Complete-Bike-Disc-brake-conversion-kit-for-all-bike-front-rear-free-tools-/190710018226

or

http://www.bikemag.com/blog/braketherapy/

bill von novak · August 2014

Re: Mobile coffee cart design question - hybrid power supply

megscoffeecart wrote: »

Because it's a front-loaded and heavy duty trike, it's not an electric-assist candidate: it'd need essentially a motorcycle brake to be able to stop the front weight once it gets cranking.

Don't quite understand this. You can get a disk brake that will stop a 1000 lb motorcycle from 90mph; stopping a 500lb coffee cart doing 10mph is very easy in comparison. And all that hardware is off-the-shelf, including the motor assist.

Here's an MTB brake that I can pretty much guarantee will stop you:
http://www.ebay.com/itm/Gatorbrake-8-Piston-Hydraulic-Disc-Brakes-Rotors-F-R-/370728396895
Here's a motor assist for a standard wheel:
http://www.ebikes.ca/shop/ebike-parts/motors/mtc100r26.html
If it's a trike that requires stub axles these will probably work:
http://www.ebikes.ca/shop/ebike-parts/motors/m3540ss.html

Mobile coffee cart design question - hybrid power supply

Comments

Categories