But I do own an Tadiran Wave INV 40/3 installed 2016, and then I had some issues with it. In hindsight, there was some real kind of problem a few years ago, and the rest of the problems were because the AC guy that came to fix the original problem replaced a 100k thermistor with a 5ok one. So the microcontroller detected something was wrong every now and then, and stopped the show (preferably on a hot summer day). He used to come, change the thermistor again, and the machine worked nicely until the next time. Why this helped on the first occasion and on those that followed is beyond me. Eventually I acquired and replaced the thermistor myself, with rated value given in the installation manual, and I hope that’s the end of it.

By the way, a 50k thermistor instead of a 100k thermistor makes the controller think the temperature is 16°C higher than it actually is. So it’s not wrong enough to make prevent the machine from kicking off, but bad enough to dislike it enough to stop after a few months. Or a year.

I have no training in the field of air conditioning or anything of that field, and my memories from the Thermodynamics course I took (and somehow passed with a fair grade) is that the world is turning into a mess. Being a human, I’m always at the risk of writing rubbish, but on this post I feel pretty free to do so. If I got something wrong, kindly correct me below. If you blew your compressor because anything I wrote below, blame yourself for taking advice from someone who has no clue.

But the story with the thermistor is yet another piece of evidence that if something is important, better be involved in the gory details. Not necessarily those written below.

The jots

• The name of the game is the refrigerant’s boiling temperature, which changes with the pressure. The compressor pushes coolant towards the condenser’s coil (“radiator”), which travels downwards and cools down there, turning into liquid somewhere in the middle of it. As the refrigerant cools down all the way through the coil, it reaches the end of the coil at a temperature that is lower than the boiling point (but the pressure remains more or less uniform). The difference between the measured temperature at the boiling point calculated from the pressure is called subcooling, and gives an indication of the level of the refrigerant inside the coil. That is, where in the coil it turned from gas to fluid.
• By the same coin, inside the evaporator, the fluid refrigerant (actually, mixed with some gas due to spark evaporation at the metering device) travels upwards and heats up by the inside air it cools. Somewhere in the middle the fluid reaches its boiling point, and continues as gas. Once again, the difference between the measured temperature at the end of the coil and the boiling point indicated by the pressure is called superheat, and indicates where in the coil this took place. This too gives an indication of the refrigerant fill.
• The measurement of temperature and pressure is typically done at the outdoor unit’s connection to the pipes. There might be differences in pressure and temperature from those actually relevant at the evaporator coil, but this is good enough for figuring out the overall situation.
• The cooling gas is R410A. According to the installer’s guide for my AC (page 34), the equilibrium temperature is 4°C with pressure 116 PSIG on the evaporator, and 50°C with 429 PSIG at the condenser (AC unit working as cooler). These figures are possible to derive, because there are typical temperatures of the refrigerant at the end of both coils: The environment’s temperature at the end of condenser’s coil, and some 15°C at the evaporator’s (?). This can however be deadly wrong if either coil (or both) don’t function properly, most likely due to poor ventilation (dirty filters, obstructions, poor ventilation path indoors etc.).
• The transition from high pressure to low occurs at the metering device, which is basically a narrow pipe. TXV / TEV valves are commonly in use. In a typical in/outdoor unit setting, the metering device is indoors, next to the coil. This means that the refrigerant flows in the pipe towards the indoor unit as a compressed fluid — this is referred to as the “liquid line” and is therefore the narrower and hotter pipe. Narrower, because the flow is slower and the refrigerant is expensive. In the other direction, the refrigerant flows as a gas, in the wider, cooler pipe. It’s a bit confusing that the hotter pipe is the one containing liquid, but it’s all a matter of pressures.
• Note that in both coils, the higher part is hotter: In the condenser because hot refrigerant is pushed from above, and it cools down as it travels downwards. In the evaporator, the fluid arrives from below, and travels upwards as it heats up. Consequently, in both coils, a physically higher transition point between fluid and gas inside the coil means a higher pressure — because the temperature there is higher. This is how the fluid levels even out between the condenser and evaporator: In steady state, the compressor generates a constant pressure difference and flow on the gas side, which works against the metering valve and differences in altitude. By conservation of mass, there must be a certain amount of fluid in the system to accommodate the existing amount of refrigerant (to complete the amount that is in gas form). If the fluid level rises on one coil, it has to drop on the other. But since a rising fluid level also means a rise in pressure, it causes more pressure against the metering device as well as the compressor. So a rise in the fluid level makes the coil push harder fluid downwards, and gas upwards making refrigerant to travel more towards the other coil, and the fluid level to drop on the current one.
• Another way to look at it: The pressure difference imposed by the compressor dictates a difference between the temperatures of the liquid-to-fluid transition points of both coils. Well more or less, because the pressure to boiling point relation isn’t linear. So given the amount of refrigerant in the system, the fluid is distributed in the segment from the condenser’s coils bottom, through the liquid line and the metering device to the evaporator coil’s bottom. The levels of the fluids stabilize so as to satisfy the temperature difference imposed by the compressor.
• However if there is too little refrigerant (a leak…), then this temperature difference relation can’t be satisfied. The result, say people out there, is bubble noise in the pipes and accumulation of ice at the outdoor unit’s wider pipe.
• The metering valve is designed to create a pressure gap with fluids, so it the pressure difference is smaller when gas runs through it. The compressor drops the pressure in the (wider) gas pipe, causing the temperature to drop well below zero, and ice accumulates.
• The viscosity of R410A is more than ten times larger as a fluid than as a gas at 10°C, so if it reaches the metering device as a gas, it flows through much easier. Hence if there isn’t enough refrigerant in the system — that is, the segment from the condeser coil’s bottom to the metering device isn’t filled with continuous liquid — gas will travel quickly through the metering device, causing a pressure drop in the liquid line, which even more gas to evaporate in the liquid line (creating bubbles that are often audible).
• Note however that accumulation of ice could be just the result of poor ventilation of the indoor unit, for example due to dust in the filters: If the gas isn’t warmed up enough in the indoor unit, its overall temperature cycle drops, and it reaches the outdoor unit very cool.
• How do you tell? Well, air condition guys ask about those filters every time for a reason. In addition, inverter AC systems have temperature sensors before, in the middle of and after the coil in both indoor and outdoor units. This way, the microcontroller can tell what the fluid level is on both sides. If it’s too low, odds are that it will correctly detect that situation and issue an error code on the user panel, and more accurately with a blinking LED on the controller PCB, either the one indoor or outdoor. It’s a good idea to download the installer’s manual for the specific AC model, if you’re brave enough to open your AC on a hot summer’s day.
• “Heat pump” is just the same of an AC that can also do heating.
• In English, HVAC is a good searching keyword for air conditioning.

Videos

• Youtube videos to watch: An overview, the cooling cycle on a training model, how to charge a system (“fill gas”) with some theory, and a more hands-on video on charging a system. This channel seems to be helpful.

FPGA and Linux and all that hi-tech stuff is nice, but nothing compares to the self pride of getting a simple plumbing job done right. So this time it was about installing a pressure gauge under a bathroom sink, between the water outlet for the faucet’s cold water and the faucet itself.

No need to tell me the reading is valid only if the taps are all closed. Anyhow, this is a simple task if you happen to have a Tee fitting that happens to match the stuff that is supposed to connect to it. If not, go find adapters. Or another Tee. If you’re a plumber, you probably have a large box with stuff to try around. As I’m not (the pipelines I usually deal with are digital register pipelines), the trick is to define the exact parts needed, and find them on AliExpress or Ebay. Or maybe even at the hardware store. The latter option turned out pretty difficult, as these parts are cheap, the motivation to help is accordingly, and if I can’t define what I need exactly, it’s a lost battle. So I ordered the stuff from AliExpress eventually. And I got it right.

So here are the notes to myself for the next time I’ll need to do something similar.

The standards

Spoiler: In Israel, everything follows BSP. A former British colony, after all.

Pluming is a local thing, performed by local people, depending on their local hardware stores, with a “give me that thingy” kind of communication. It’s therefore quite difficult to find exact definitions for plumbing fittings. It goes by “see if it fits”.

For threaded fittings, terms like ½” and ¾” are often used, but they refer to nothing measured on the piece of metal itself. These figures used to tell the inner diameter of the pipe itself, but that doesn’t work anymore. So if you want to measure a fitting and tell what it’s called in the market, you need to go to the standard tables.

And here comes the real fun. There are mainly two standards for pipe sizes, which detail the dimensions for the pipes and threads. It seems like the most common ones in Israel (and Europe) follow the British Standard (BSP), but there’s also American National Standard Pipe Thread standards (Often referred to as Nominal Pipe Size, NPS or National Pipe Taper / Thread, NPT).

Sometimes IPS (Iron Pipe Size) is mentioned, but it usually means NPS.

The two standards are incompatible, despite similar terminology and measures. In particular, the thread pitch doesn’t match between the two standards. But there’s also the thread form: American goes with Sellers, which has sharp edges, and British with Whitworth thread form, which is a sine wave shape. Not that I can tell the difference just by looking. So if you try to mix British with American fittings, screwing will probably be difficult, and it won’t hold pressure well, if at all.

Either way, for historical reasons, the inch number used in these standard matches none of the measured diameters of the pipe: Neither the inner or outer. The OD, outer diameter, is the easiest to measure, and should be compared with the standard.

So the main headache is BSP vs. NTP (or NPS, IPS, MIP, FIP and all other abbreviations meaning “American”).

And then we have this thing with DN sizes. For example, DN15 means ½”, and DN20 means ¾”. Even though these were coined for the American standard (and listed on Wikipedia’s page for NPS) it seems like they’re also used in context of BSP. So if a product is listed with a DN number, it probably means nothing on which BSP vs. NTP.

Actual measurements

This is what I measured on my own stuff.

• The tap for my washing machine is a ¾” according to the manual, I measured 0.97″ outer diameter (1.05″ per standard). Apparently washing machines go BSP.
• A typical shower head has a ½” fitting, not clear if American or British (for this size, it seems like NPS and BSP are roughly the same).
• My supply stop valves (those wall taps for bathroom faucets) are 3/8″ (measured 0.64″ outer diameter). In Israel, these wall-mounted angle valves are called “NIL taps (ברז ניל)”, which most likely refers to the German company NIL, and therefore conforms to BSP.
• The pressure gauge is an ¼” (measured outer diameter 0.5″).

Iron and brass

It’s pretty well known, that if brass fittings are used on iron pipes, or if these two metals are mixed in any other way, the iron will corrode rapidly (within a few years), as they work together as a battery. So the material is crucial.

In Israel, all valves, taps and faucets are made of chrome or nickel plated brass, and is therefore OK for use with brass Tees and adapters.

“Teflon” tape (or PTFE)

When there’s no rubber ring sealing, teflon tape is applied on the thread. It works better when there’s a hard end to the screwing, as the force of the end works on the thread and the teflon applied to it. But it can work well otherwise.

It’s 20 rounds around, or it won’t seal. Apply evenly on the thread with slight tension. The direction is as for the turn direction while fitting, i.e. the circular motion will tighten the “teflon” even more (and not unwind it). Don’t cover the few first threads (the end of the pipe) for easier fitting. If the fitting torque is easy all the way, it’s not going to seal.