30 second delays in internet on AT&T U-verse 5268AC FXN modem

My wife and I were running into inexplicable “delays” in our AT&T internet service over WiFi. The speed of the internet would be fine when it worked (speedtests showed good lag/upload/download, etc…) but sometimes the entire internet would “pause” and not respond for 20-30 seconds at a time. Usually not enough time for a connection to time out, but websites would be stuck loading for a long time, or Google Web Apps wold have a “loading….” message for half a minute before recovering (or failing to recover, making us try again with an edit to a document or calendar item…).

After much gnashing of teeth, network profiling, and dark vodoo, we traced the problem down to our devices auto-switching between the 2.4 Ghz and 5 Ghz wifi networks from the router [a Pace DSL modem Model 5268AC FXN ].   The problems happened most frequently when we had about 50% wifi signal strength to the 5G radio, and apparently our devices would see the stronger signal strength on the 2.4 router and switch over to it, then decide to switch back, and so forth.

The root cause of the problem is that the AT&T Uverse DSL gateway / wifi router has both networks with the same SSID (Name) and password, so our devices felt that they were “the same” network, just on different frequencies, and would switch between them frequently.   I have no idea why this would cause a delay of TCP/IP traffic, as a change in the physical/data link layer shouldn’t affect the Network/Transport layers (at least, not for 30 seconds).   Perhaps when using a different brand/model of Wifi Router devices can auto-switch between 2.4 and 5g seamlessly. ( Or perhaps not, our previous cable modem from Spectrum / BrightHouse named the two networks differently (with a 2 and 5 suffix) so that once you connected to a particular network frequency, you stuck with it, but at least we didn’t see this type of issue. )

In any case, the solution was simple. For testing purposes, we fixed the BSSID (mac address of the router) in our client devices to the 2.4 Ghz network, so it would not switch to the 5 Ghz radio. This fixed the problem.   Renaming the 5 Ghz network name to something different from the 2.4 Ghz network on the router would also have the same effect for all devices (for example, using myNetwork2.4 and myNetwork5 as the names).

Dual hose portable AC window adapter & Whynter ARC-14SH review

I found a good deal on a used Whynter ARC-14SH dual hose portable air conditioner and have installed it in the storage room off my garage. If I leave the door open it can help cool the rest of the garage, or if I close the door, it can condition the air in just the storage room quite easily.

I wanted something a little more permanent and secure than the included plastic window adapter kit, so I cut a piece of 1/2″ plywood to fit inside some convenient pre-existing slots in my single hung window, attached the hose end-plates to it and painted it with exterior paint.   The board fits into the window just inside the existing screen, so I don’t have to worry about bugs getting into the inside of my AC unit. Here is a video montage of building the window adapter.

 

I was considering a window mount unit, but this portable unit gives me the flexibility to mount it inside the garage later (venting out the ceiling to the Fascia) or wheel it into the house to use to cool a single room if the main AC goes out.   Obviously, having outside and hot exhaust air cycle through two hoses inside the conditioned space is slightly less efficient than a window unit, but it’s much better than a single hose portable AC unit, which will cause outside air to slip into your building as it exhausts it’s waste heated air. It also presents a cleaner look on the outside of the window. (It does take up more floor space inside the room however….)

The EER rating for this unit is 11.2 according to the Home Depot website.   A comparably priced 14,000 BTU Energy Star window air conditioner from GE has an EER rating of 11.8, so from an efficiency standpoint the portable dual hose model isn’t terrible.

The unit draws a maximum of 10.5 amps, and appears to hover around 1050 watts when the compressor is running and 45-65 watts to just run the fan, depending upon what speed the fan is set to. I usually leave the fan on low if only cooling the storage room, and turn it to high when cooling the rest of the garage. The compressor is at least as noisy as the fan in “high” mode, so don’t expect “low” mode to be quiet AND condition the air at the same time, although you can run the unit in “fan only” mode if you just want to circulate air.   The unit is a bit noisy (56 dBA). This is not bad for a workshop, but could be an issue in a bedroom or media room. (In comparison, a nice mini-split ductless AC unit usually runs closer to 34 dbA.)

In AC mode it uses the collected water to evaporatively cool the hot side, and exhausts the humidity with the rejected heat, so in my experiance doesn’t need to be otherwise drained. (This hot moist exhaust air is another good reason to paint the entire adapter board with exterior paint.)

If you run it in “dehumidifier” mode (where it attempts to remove water vapor without putting too much energy into cooling) you are supposed to vet the exhaust air back into your conditioned space to prevent “cooling” from happening. (Living in Florida, this probably isn’t an issue for me….). But, you also need to remove the collected water. Because this is primarily an AC unit, it doesn’t have a large water reservoir, so to use it effectively as a dehumidifier, you will NEED to rig up a permanent drain hose of some type. And, because the drain is located about 2″ above the floor, you may need a pump system unless have have a conveniently located floor drain nearby.   I haven’t tested the heating mode, but according to the manual it has one that will work with outside air down to 45 °F. It also has some timer modes to turn on or off after a set number of hours which I also haven’t used yet.

Amazon Dash Wand

Amazon is running a promotion where you can buy an Amazon Dash Wand with Alexa for $20, and get a $20 credit on your account when you register it. (So if you are a prime member with free shipping, you get a dash wand for $1.30 in taxes.) It’s basically a handheld wifi enabled barcode scanner with Alexa voice input designed to get you to buy more stuff from Amazon.

The wand arrives in a small blue and black box, and is half white and half black. The black end shrinks to a rubber ring so that you can hang it up on the included sticky hook, and it has magnets hidden inside so you can stick it to your fridge.

 

To install the two AAA batteries (included in the box) the quickstart guide says “Open the Amazon Wand by pulling the two halves apart” when it should really say “Get two strong guys to play tug-of-war with your wand until the two sides pop apart”.

It requires you to have the Amazon app on your phone to pair the Wand with a (2.4 Ghz only) Wifi Network, which also links it to your account. After that, you can use the wand by pressing the button. A red light shines out the end, and if you point it at a barcode, the item will magically appear in your Amazon Shopping cart. (No wonder they are practically giving them away…)   You can also press and hold the button to talk with Alexa, to, for example, add an item to your cart that doesn’t have a barcode by voice.

Amazon music is not supported on the device (I suspect playing music would run the AAA batteries down too quickly, plus the single speaker isn’t exactly high quality), but some other Alexa skills are, so you can check on the weather or play colossal cave (although you have to push and hold the button every time you want to issue a command).   Home control commands (such as HUE lighting) is supported, so this could make a good secondary control device for a smart house. Messaging with Alexa is NOT supported. (Which is a pity, as it’s ALMOST small enough to wear like a   combadge.)

For a teardown, see this link.

Practicalities of On-board solar charging for small EV’s

I’ve been running the numbers on building a small 1-2 person “motorcycle” (3 wheeled) electric vehicle, and was considering adding two 330 watt solar panels to act as the hood and roof/sunshade, which would provide shade for the driver and charging from the sun.

The drive motor I was looking at runs at 96 volts and 95 amps to drive a 325 lb vehicle (with 170lb rider) at 60+ mph. Twelve Nissan leaf modules would provide 96-100 volts at 60 ah for a total storage capacity of 5.7 kWh (giving around a 45 mile range at 60mph, probably close to a 60 mile range at 35mph, an efficiency of   between 83-111 Wh/mile).   This battery pack would weigh 100 lbs, plus BMS/mounting hardware and wiring.

Weight Considerations

Two 330 watt solar panels mounted on the roof/hood would also weigh 100 lbs.This could conceivably be 30% or more of your vehicles weight budget.

With around 6 hours of good solar exposure a day, they would probably provide around 600 watts per hour, or 3.6 kWh of charge (a gain in driving range of between 32-44 per day). They could fully charge my hypothetical 5.7 kWh battery pack in two days.

More batteries?

The alternate way to spend this weight budget is to double the battery pack size. This would give a 11.5 kWh battery pack, giving 90-120 mile range from a single charge. A side benefit is that the extra 100lb of weight could be placed low to the ground, instead of up high on the roof of the vehicle, greatly improving performance on corners.   (Also, the aerodynamic effects upon handling and range of adding a horizontal sail to the top of your vehicle must be considered….)

In my opinion, if you are regularly returning to a home charger, it is more practical to use extra weight allowance for batteries, as opposed to solar panels. Solar panels make the most sense when the vehicle is designed for non-round-trip applications, such as with an RV/Camper or road trip vehicle.

Bigger/Faster charger?

For an “in-town” vehicle, where J1772 (level 2) chargers are readily available, adding a high speed on-board charger (6.6kWh) would allow you to refill a small battery pack in under an hour, and would add less weight than commercial solar panels or extra batteries.   Having an extra 15 lbs of charger instead of an extra 100 lbs of solar panels or batteries would lower your rolling resistance and increase your range and acceleration.

Specialized solar panels

Alternate solar panels (smaller RV style, or thin film flexible solar panels) would weigh slightly less, but the weight savings is not as impressive as you may think. A 330 watt “house style” panel weights 50 lbs, or 0.15 lb per watt. A 100 watt RV panel weights 15 lbs, or the same 0.15 lb per watt. A 72 watt PowerOak flexible panel weights 6.2 lbs, or 0.086 lbs per watt. This is a weight savings of almost 50%, but unfortunately they are much less efficient, so would need more surface area, something in short supply on a motorcycle class EV, plus they cost much more on a per-watt basis.

Custom Alternatives

If you wanted to take the time to fabricate your own solar panels out of individual cells as part of a fiberglass layup, you could conceivably make them weigh less and fit the contour of your vehicle better, possibly integrating them into your vehicles body.   But if they are integrated into the skin of your vehicle you have to worry about solar heat gain. I think it would be better to have them mounted as a “shade” or “2nd skin” just above your vehicles main body with airflow channels between the two.

Cost Considerations

100 lb of 330 watt solar panels (two) cost around $500, while a 100 lb Li-Ion battery pack would cost about $1200-1500 (unless salvaged from a surplus battery pack). So the solar panels could cost less than a larger battery, but would require more work to integrate into the vehicle. A 1.5 to 2kWh charger would be fully adequate for a vehicle with a 5.7 kWh battery pack. You could even have only 110V charging (1kWh) and save the expense and complication of a J1772 inlet, while still being able to recharge a fully used battery pack in six hours. A minimal charger like the ELCON PFC1500 would cost $575.   An Elcon PFC 5000 ( TCCH-84-50 ) could charge at 5 kW, giving a small EV an almost “QuickCharge” charging speeds for around $2000 with J1772 inlet/adapter.

Modular Vehicle

One option would be to mount several solar panels on a trailer (possibly with a 2nd battery pack, and even extra motors) to be used only on longer “road-trips”. It is possible that the trailer could have room to hold 4×8 sheet goods, and/or a sleeping compartment under the solar panels for road trips. If the solar panels could swing up, it could be used for transporting larger furniture or appliances. (Consideration would have to be given to adding a lower gear ratio to the tow vehicle, or including extra motors on the trailer itself for heavier loads.)

LED Headlight power savings for Electric Vehicles

Upgrading the headlights on an EV from incandescent bulbs to LED’s will save some electricity, but it’s such a small amount of electricity compared to what the motor uses to move the EV that it’s probably not worth the effort purely from a range perspective.

My original OEM halogen headlights (together, at high beam) take 130 watts, while the motor takes around 12,000 watts just cruising down the road. So the headlights account for only 1% of the total energy usage.

One burnt out, so I replaced them with LED units that take less power.
Driving around with my high beams on, I’m saving a maximum of 78 watts by replacing both my headlights. 78 / 12000 = just over a 1/2 of one percent energy savings. So that gives me an extra quarter mile of range.

There are other benefits to LED headlights when compared to the OEM halogens. First, they produce more light, which is a safety advantage when driving at night. (I was never happy with the light output from the original headlights, and feel happier driving at night with the LED units.)

If you like the “cool white” color temperature of LED bulbs, you can say it’s an improvement in appearance. (They do make the headlights look more modern, as most newer cars have “cool white” headlights.)

Because the headlights are one of the largest consumers of power on the 12 volt accessory bus, swapping them out would allow you to significantly reduce the size/capacity of your DC-to-DC converter used to keep your 12 volt accessory battery charged.

 

The real power savings come from the addition of DRL, which only take 2.2 watts, compared to driving around with my low beams on.

G-board uses more power than it should, shortening your battery life

The new Google Keyboard application (G-Board / Gboard) works fine, and has a few nice features, such as the ability to search for an animated gif to send when you want to blow through a lot of bandwidth.

You might notice that in the screenshots above, the Gboard power usage is higher than my email client, and even YouTube.   When you take more power than streaming videos, you know you are a power hog.

However, all of these new features come at the expensive of battery life.   As a simple keyboard application, it shouldn’t be using more power than any of the other applications on my phone.   I use the keyboard when sending text messages or email, and for a few search bars, etc…

In my opinion, a keyboard app should never appear in the top power consumers on your phone.

 

Updating XML Google Maps plugin for newer versions of PHP

If you are using the XML Google Maps wordpress plugin (version 1.14.1 by Patrick Matusz), it has not been updated for a while, and won’t work quite right with newer versions of PHP.

If you are getting the following error message twice at the top of every page in your wordpress blog:
Warning: mysql_get_server_info(): No such file or directory in … on line 10

Change line 10 from:

 

if ( version_compare(mysql_get_server_info(), ‘4.1.0’, ‘>=’) ) {
to:
if ( version_compare($mysqli->server_info, ‘4.1.0’, ‘>=’) ) {

And that will make things work without the error messages.

Curtis 1231c upgrade: Binning Gate Drive Resistors

I am upgrading the power board of my Curtis 1231c DC PWM motor controller. It uses 18 MOSFETs to switch the power, and each MOSFET had a 47 ohm resistor on it’s gate input. The point of such a high resistance was to slow down the switching of the MOSFET’s so that they would all share the current somewhat equally and no single MOSFET would turn completely on before all of the others had a chance to start shouldering the load.

Continue reading

Comparing Brighthouse Networks (Spectrum) with AT&T U-verse Internet

I’ve moved my SamKnows FCC broadband testing unit from one location to another in the middle of the month. The first location had one of the best home Internet plans available from BrightHouse networks (now Spectrum after the Charter purchase), the “Lighting 200” plan. The new location has the best available AT&T U-Verse plan, limited by the distance from a central office and older wires. AT&T did set me up with a bonded pair (using 2 phone lines, or 4 wires to provide better service).   The U-Verse plan is much slower, on both downlink and uplink, but is perfectly serviceable. The graphs below do show the stark contrast in available bandwidth.

I am going to miss the 24 up / 200 down of BrightHouse, but the 5 up / 30 down on U-Verse is still enough bandwidth for most needs.

The quality of the service has also gone down slightly, with slightly higher rates of packet loss and slightly higher latencies.

The extra 12 ms of latency isn’t much to worry about, but it is a definite change. The jump from almost no packet loss to 0.5 to 1 percent packet loss is annoying, and you can see that even on the worst day Spectrum / BrightHouse   is better than AT&T U-Verse on an average day.