In full shade I could see that, or using the a/c only at night when they typically draw very little power. I didn't realize that was your planned use-case. Everyone can and should build for what they want/need. No harm intended with all this feedback. I just didn't realize that was your plan.I ran it today non stop in my garage at 95 for 9 hours and it stayed right around 350 watts. I have camped many times in over 100 temps and I’m not expect to run the AC and sit in the van. The only way to get the AC to pull close to 600 watts was to put it in boost mode. And that 50 amp draw is still within the 70 amp specs of the converter and even at that for 2 hours and the converter maxed out at 115 with no air flow. Yes the converter is rated for 70 amps continuous and up to 85 for short periods of time.
I only started this because I thought 115 was hot but I’m learning it is not crazy hot. I also figured out it gets that hot pulling 13 amps so I guess that is just the working temp.
LOL!In full shade I could see that, or using the a/c only at night when they typically draw very little power. I didn't realize that was your planned use-case. Everyone can and should build for what they want/need. No harm intended with all this feedback. I just didn't realize that was your plan.
I guess for me the point of a/c is to both to sleep well, and to escape the blazing afternoon heat. Hike all morning/midday, then cool off and watch a movie, read a book, or take a nap in the afternoon with the a/c blowing cold. But everyone's situation is different.
A few other thoughts that may be useful. If not, feel free to ignore:
When it's 100F outside like it's been for the past 2 months here, with the van partially shaded by a nearby tree, my a/c barely draws any power (85-155W) and the compressor runs very low. But when the full sun hits the metal van and windshield for a few hours in the afternoon, power consumption goes up 200-400% to maintain the same temp. I could force the a/c to stay on a low setting, but van temps would rise quickly.
Before insulating, I did some heat tests. After insulating, without any a/c running it just takes longer for these same temps to be reached. Insulation or not, eventually everything becomes heat-soaked. Here's a 100F ambient day with the probe in the shade:
View attachment 177048
Re-parked the van in full sun (was in partial shade), and after just 15 minutes:
View attachment 177049
My Fluke 325 is rated to operate up to 122F, so I stopped the test even though temps were still rising. But before I got out, I used the probe (which can handle much higher temps) to do a quick check of the dash area:
View attachment 177050
My point is 115F case temps while parked in the shade with an a/c running inside the van may not be an issue. But if you park in the sun and go for a hike on a 100F day and the converter case becomes heat soaked at 120F+ , and then it has to dissipate its usual heat when you get back and turn on the a/c, you may have issues, especially if that happens repeatedly.
System designers typically aim for 33-50% overhead to help deal with situations like that. Hence my suggestion of a larger converter, and affirmation of other people's advice to add forced convection cooling.
But that was before I knew you plan was to only run the a/c at night.
Lol. Happy to help. I really wasn't trying to show off, but true that -- they're pricey! I bought it two years at the beginning of the van build, so it's not quite new anymore, and the thermocouple was just a bonus feature. But it sure has come in handy.LOL!
How much of your post is about temperature in your van, and how much is about your new very expensive Fluke clamp meter? 😀
Source: I was recently in the market.
Thanks for the show-and-tell!
Very interesting on the operational temperature. I checked my UNI-T clamp meter and these are the temp specs as I understand them:Lol. Happy to help. I really wasn't trying to show off, but true that -- they're pricey! I bought it two years at the beginning of the van build, so it's not quite new anymore, and the thermocouple was just a bonus feature. But it sure has come in handy.
The main benefit of the Fluke versus Amazon/China brands, Klein, or Milwaukee is they can't operate below 32F. The fluke can get down to 14F, which helps if you ever need to troubleshoot something in the dead of winter. They're accuracy is also as good as you can get for a handheld. I mostly needed it for clamp amp monitoring during the power system build, but it's also been useful for other things around the house. The genuine fluke alligator clip set is also really nice. I have a second cheaper Amazon/China unit (AstroAi TRMS 6000) and their addon alligator clip set is a real step down. You notice it when inserting the probe into the alligator clip head. It misses the inner barrel about half the time on the cheaper set, and never misses on the Fluke set. Everything just feels like quality with Fluke gear.
Yep, same for Fluke regarding temp coefficient of accuracy. The other brands I mentioned don't ensure any level of accuracy below 32F. They just aren't designed for use below that.Very interesting on the operational temperature. I checked my UNI-T clamp meter and these are the temp specs as I understand them:
Accuracy assured temperature condition is 18-28 C (64-82 F); stability within ±1 C.
Outside this (<18 C or >28 C) an additional temperature coefficient error is 0.1 x (specified accuracy)/(C)
AC or DC current accuracy ±(2%+5 counts)
So, as I understand it, at -10 C (14 F), there would be an additional temperature coefficient error of 2.8 x (specified accuracy)
Good to know.
At this point, I don't plan on voluntarily spending time at 14 F, let alone making measurements then, but thank you for bringing it to may attention.
I have a Fluke DMM, and I find the quality of the physical aspects of my new UNI-T clamp meter to be fully on par.
I did do research. UNI-T (actually Uni-Trend Technology) is significant designer and manufacturer of a broad range of electrical test equipment.
My theory was that Fluke had priced themselves so high that they left a huge opening for competent competitors to become a factor in the market. Another important vendor was Triplett, but they seem mostly gone. As for Klein and Milwaukee, electrical measurement tools are not a core business. Their products are likely rebranded white box faire. When studying, I came upon UNI-T.I didn't know about UNI-T before this discussion, but they sound like a solid company. Thanks for the heads up.
Pricing is pretty outrageous, that's for sure. Then again, I bought mine in 2020 when it wasn't quite so bad, and I expect it to last for decades. Sub-32F operation was the factor that pushed me over the edge, and again, I've seen many freezing mornings in the rig, so I know it'll come in handy someday. Even testing outdoor circuits around the house after a big freeze/storm seems likely. It looks like UNI-T also has that 0C/32F lower limit.My theory was that Fluke had priced themselves so high that they left a huge opening for competent competitors to become a factor in the market. Another important vendor was Triplett, but they seem mostly gone. As for Klein and Milwaukee, electrical measurement tools are not a core business. Their products are likely rebranded white box faire. When studying, I came upon UNI-T.
If Fluke isn't careful, they may not be a factor in the market in the future. Maybe they are already there.
I'm not sure about techs, but as an EE, I can't imagine being tied to a particular test tool. For simple devices like DMMs, etc. you should be able to quickly pick up a different brand. For complex test instruments, the designs evolve too fast to expect that you won't have to learn how to use a new device, no matter the brand.Pricing is pretty outrageous, that's for sure. Then again, I bought mine in 2020 when it wasn't quite so bad, and I expect it to last for decades. Sub-32F operation was the factor that pushed me over the edge, and again, I've seen many freezing mornings in the rig, so I know it'll come in handy someday. Even testing outdoor circuits around the house after a big freeze/storm seems likely. It looks like UNI-T also has that 0C/32F lower limit.
But part of the cost is certifications and accreditations. Many education institutions, energy companies, and municipalities only train their employees on Fluke instruments. That's all we used in the electrical engineering program way back when I got my degree, and all the labs still used them when I was working there three years ago. Fluke has like 15 divisions and makes gear for everything from biomed to imaging. They're definitely not going anywhere. The cost of re-certifying to another brand, updating classroom and field training guides, avoiding safety mishaps during the change, etc, is prohibitive, and you have to remember, linemen and other EE's and techs are out there using these devices in dangerous situations everyday, so when it's life and death a few hundred bucks doesn't mean much.
But DIY'ers and hobbyists probably don't need to worry as much about that because they're not doing dangerous work five days a week, so they can probably save some money with a budget brand. I'd just want sub 32F functionality as a minimum. I did look around at the time and would have gladly paid less for an equivalent item with that feature, but every brand I checked only operated above 32F.
EE is a wide field with many sub disciplines. Our power program for instance is nothing at all like our computer engineering program. Engineering Colleges and their constituent departments also vary widely in makeup. An EE department that doesn't have a well funded power program, and hasn't produced any CEOs of major power companies may not have much in common with one that does. Heck, the College had just spent millions upgrading the Innovation Space and fab lab when I left, and the Department was up next on campus for a new building (tens of millions). We probably spent more on toner cartridges than all the Fluke meters combined.I'm not sure about techs, but as an EE, I can't imagine being tied to a particular test tool. For simple devices like DMMs, etc. you should be able to quickly pick up a different brand. For complex test instruments, the designs evolve too fast to expect that you won't have to learn how to use a new device, no matter the brand.
I also saw no brand loyalty in my university setting. In fact we occasionally had different devices at different lab stations. You were expected to learn how to use the tool. You may be speaking of some tech schools which have a narrow industry focus, be even there, I didn't work with techs that had that type of training.
I suspect we are talking past each other. I have no concept of using the term linemen and EE in the same sentence, so you have a very different vision of what the EEs I know do for a living.
As an engineer, for most purposes, a solid mid-level DMM is fine. There are situations, like testing, etc. where traceable calibration is necessary, but even that doesn't require a gold-plated tool.
Fair enough. I was surprised we went from joking about pricey Fluke meters to debating the value of the company itself. But I know what you mean, they are pushing the pricing envelope.I'm sorry, I hadn't intended to bash Fluke. I was mostly trying to explain (to myself) how they could justify such premium prices. You provided some good answers as well as pointing out the breadth of their current product lines.
I also never look down on other disciplines or roles, especially the excellent technicians I have worked with.
I have never worked in power generation and distribution systems, and I'm sure that much of your commentary on safety and certifications relates to that field.
While electrical engineering is a broad subject, I went to university at a time and place where the "focus" was on learning about all of the "sub-disciplines" of electrical engineering. As far as credit hours, it was truly five full years to get a BS. The EE department was the largest department of the largest university in the US at the time. I don't believe that many other students were (or are) as fortunate as I was to get this kind of broad education.
We had a well funded power program, as well as most other EE sub-disciplines. I ended up most interested in—what turned out to be very broad—digital design. The sub-disciplines of logic, computer design, signal theory, control theory, electromagnetics, semiconductor design, as well as some analog design have all played a role in my career.
I used simple and complex test equipment throughout my education and career. But I didn't spend any real time using my power systems training.
I am currently spending time putting together a battery electric power system for my camper van. I find that there is plenty to learn in this field also. I am learning about the chemistry of batteries, the poorly documented subject of charging and discharging LiFePO4 batteries, the use and operation of fuses and circuit breakers, as well as the construction methods for putting it all together.
My initial comment was a tease for all the pictures of your expensive Fluke meter. I really did mean it in fun.
Thanks for sharing.Fair enough. I was surprised we went from joking about pricey Fluke meters to debating the value of the company itself. But I know what you mean, they are pushing the pricing envelope.
I've been down that rabbit hole of battery chemistry research, read dozens of journals, news coverage and investigative reports on major lithium fires in auto, power plants (APS) and industrial settings, and spent months testing sample cells with an iCharger X6. Here's what I concluded:
NMC (what Tesla has mostly used, although their angling away from Cobalt now), is a higher risk option for RV and marine applications. You get greater energy density, which is a major plus, but there are just too many ways an NMC bank can overheat or be accidentally charged when frozen. In my view, NMC is only suitable for use with dedicated heating/cooling system built by a reputable turn-key vendor.
For that reason, LFP really is the safest option, and you can still get reasonably good energy density by opting for prismatic cells. Cylindrical cells only have one real advantage of greater heat dissipation. Some will argue they are safer, but the sealed ends can and do blow out in extreme situations. Prismatic cells expand/contract during normal operation, so in my view, it's not wise to build DIY batteries using them. But many quality vendors sell batteries with dedicated cases that conform to the cell bundle which limits excessive expansion. Some of the highest end battery makers (Lithionics being the biggest) focus almost entirely on prismatic packs. Upstart retail vendors like Battleborn sell bulky, massively overpriced batteries using cylindrical cells and spend tons marketing them. Don't believe the hype. It's a total waste. They buy their cells from the same place everyone else does -- China.
Once you settle on LFP cylindrical cells from a respected vendor with a dedicated conforming case, then it gets easier. SOK seems popular these days, and their latest packs do offer CAN communication, which can be useful for certain things like signaling an impending BMS cutoff so an alternator regulator (Wakespeed) for a dedicated second alternator can shut down. But Al over at Wakespeed is particular about which vendors he's willing to support, and I'm not sure if he's included SOK yet. Plus, if your bank is 24V (ideal voltage for most people) you don't really need that function. A simple Sterling load dump protection device can serve as a backup for that rare circumstance, if it even occurs.
I personally went with Carl over at electric car parts company. He's been around for ages, and has a good reputation. He mostly sells BestGo batteries, and other high quality cells. I got to know his contact at BestGo quite well. He sent me sample cells which proved to be top of the line, with excellent IR (internal resistance, ~0.5mohm), full stated capacity (and then some), and incredibly low parasitic discharge. I've had the two cells in the hot/cold garage for 2 years now, and only had to add a few Ah once last year to keep them at 50% SOC. The battery packs themselves have a really nice aluminum case with IP67 rating and very good vibration tolerance. I expect them to last for years.
I had them design a custom "Pro" pack for me with extra features, but in hindsight I'm not sure it was worth it. I haven't used any of them yet, but maybe someday when the batteries get old they'll help with troubleshooting. They keep price down by going with aluminum busbars instead of copper, so there's a tad more heat generation during high-amp charge/discharge, but the reality is with a large bank (20kWh) there's just not a lot of opportunities to push the C rates. Most of the time when traveling it's just my solar array, which even with 1110W, when you spread it across four batteries and account for panel heating and the sun's angle, it's only driving them around 0.05C, so they only rise 1-2F during charging. If I add my inverter/charger in the mix, maybe 3-4F increase. Not enough to even blink at.
Regarding LFP charging best practices, I recommend this: when actively using the bank, charge to 95% (around 14.2/28.4/56.8 for 12/24/48 banks), and discharge to 15% (use coulomb counting on a quality monitor for discharge, not voltage, which will vary widely due to varying loads). That gives you 80% capacity and a good lifespan. If possible, narrow that range if you find you don't need that much power (e.g. during the winter). I cycle between 40 and 70% when the weather is good in the winter and I have too much solar output. When at home, store around 40% and keep the temps under 85F. Storing lithium at 100F and 100% SOC will rapidly shorten its lifespan.
For monitoring, it's hard to beat Victron's BMV 712 because it gives you a relay that can trigger on SOC or many other data points. Otherwise Victron's smart shunt will do. Both let you monitor in the app, which is really nice. And if you're willing to shell out, the overpriced Cerbo GX lets you monitor from the cloud. It's expensive, but worth it. I use it all the time.
Lastly, RV tank heater pads with their imperfect thermostats removed (described in my build thread) are a simple inexpensive way to keep the batteries warm if you plan to use the RV during very cold winter outings. I paired them with my MPPT's temp-controlled relay and a timer relay, so I can use time slots (night, early morning) and temp (<50F) to warm them (>60F) as needed.
Ahh yes, good catch. I meant "settle on prismatic cells." Fixed it. Thanks.Thanks for sharing.
Much of what you suggest as set points I have read elsewhere. A thought on working on saving lifespan though: we may be at a point where the practical useful lifespan exceeds the the lifespan due to larger charge/discharge cycles.
Can you explain the effects of an alternator losing its load (such as a BMS shutting off)? In my case, I am charging from CCP2, which Ford's system switches on and off with a relay. Maybe the system "prepares" the smart alternator for the change in load.
You likely have a typo: the SOK batteries consist on prismatic cells, not cylindrical cells.