Classic Retrofit A/C Experience & Guide

[Duc Hunter]

NOTE SOME PEOPLE ARE SEEING SOME OF THE PICTURES UPSIDE-DOWN. Let me know if that is the case. On my systems they all seem to be correct.

Disclaimers

This is a long one, sorry. It just kept growing and growing. I am by no means an expert, just wanted to pass along what I learned.

I paid for my Classic RetroFit (CR) A/C System and Alternator myself. I have no affiliation with CR, other than being a customer.

I live in the Florida, USA. I am tried to use Celsius as much as possible as my outside air temp gauge, my evap temp gauge, and the systems ECU all read in Celsius. I also tried to convert to Fahrenheit as much as I could but to be honest, I got tired of converting everything.

Yes, I repeat some things throughout this write up. That is mostly on purpose, because I feel they need to be emphasized, or are pertinent in several different areas.

BLUF (Bottom Line Up Front

I loved the idea of the CR A/C before I bought it and knew my particular case would be a tough installation. I also knew my climate would be about the toughest there is for an A/C system given the heat and humidity we have here. Throughout my install CR have been very supportive and helpful, even when I screwed things up myself. I did that at times, royally to be sure. They were wonderfully patient and positive.

So, what do I think of the system? I have an A/C system that works very well! That is even though I am in a miserably difficult climate for an A/C system, and in a car that is almost as bad as a stripped-out race car for and A/C system. I believe that if it works the way it does for me in my car, it would be even more amazing in a car like my 1989 3.2 Carrera (nice factory interior, never been a stripped out race car, has all its sound deadening, etc), let alone how it would work in places that are less humid, or hot.

In short, on any day but the hottest days here the car is very comfortable, I turn off the footwell/crotch vents, and turn down the blower rather quickly. On the hottest days, the car cools off after being heat soaked in the sun and is plenty comfortable to drive. In fact I have been daily’ing the car all this summer, a first in any of my air-cooled cars.

My Goals for this writeup

To provide people looking into, or who have the Classic Retrofit A/C system some guidance on getting the install & settings right, and the benefit of my testing and experience.

To provide the information in a way that will help others avoid some basic pitfalls that I ran into during my installation & setup, and that can enable people to “rapidly” test their newly installed system, and arrive at really good settings that are adjusted to their personal installation. Installation technique has a big effect on where the settings need to be for comfort.

To provide a good understanding of how the system works, where performance can be gained through the settings or install, and where it cannot.

To provide good data to CR and others from one of the worst climates in the world for A/C, and CR’s development. Keep in mind that CR has no way to simulate humidity in their environmental chamber, as it gets drier the hotter the temp gets. Therefor they were taking a bit of an educated guess (and a good one) at the system’s settings and capabilities in humid climates. Because of this I wanted to do as much as I could to gather data for them in my heat chamber, called Florida. Just the other day it was 101F air temp and over 40% humidity here. Having lived in CA for 15 years, and spending time every year in the Nevada desert in the summer, I can say my climate is far tougher on man and A/C than a 120+ degree desert day. Based on my climate and experience I came to the ECU settings and testing procedures below. Jonny and the folks at CR have been very appreciative and supportive of my testing.

To try to give back a little bit to this community that has given me so much over the years. This has ranged from actual advice, to entertainment, and inspiration. I also wanted to give back to the historical record that this board is, and how that history of information has been so invaluable in my Porsche Journey to date.

My A/C Setting Theory

In a heat-soaked car I want to cool the car as fast as possible and get it to my desired temp ASAP. Many modern A/C systems (like my 2012 BMW X5) start to back off well above the set temp, and therefor never really get to the set temp. In my X5 I have to lower the temp below where I want it, then once the car gets close to how I want it to feel I move the temp setting back up to let the car back off and keep the car cool. This is rather annoying in a hot and humid climate. My desire was to have the system “cold crash” the car as fast as possible, and not back off until right before the set temp, while still giving the system enough resolution to back off without over cooling the car. I think I have achieved that, though come cooler fall weather I may need to make some minor adjustments. Time will tell.

In a hot climate I want to avoid having the compressor start cycling on/off to maintain temp. In hot climates and/or humid climates this causes larger temperature swings than I find comfortable. This is especially true if the car is heat soaked (just started after being parked in the sun), and you are sitting at a stoplight. With CR’s electric A/C it should be possible to have the compressor always running and let the ECU control it such that the cabin is comfortable. I built a computer model of the PID controller that is this systems ECU. This model enabled me to play with settings in a virtual environment so I could arrive at settings I thought would work. With these setting the system will run at max cooling power until it is within about 1.75C (2.8F) of the Cabin Set Temperature (CTS). In my testing this works well in the hot summer.

A/C Things to Know.

I have proven this system has the capacity to handle very hot and very humid climates. When an A/C system runs out of capacity the condenser output temperature rises well above the ambient air temp. Conversely in a properly running system the output fitting on the condenser is near ambient at all times, once the system has stabilized. CR and I occasionally talked about the possibility of adding a second condenser to the system to help with my really hot & humid climate. To-date though my condenser output has never been more than 3C above ambient (using my IR Gun), and that includes on 100+F days with 40%+ humidity.

Humidity is a killer on all A/C systems, as I keep saying. Dry desert air at 120F degrees is far easier on an A/C system than 92F and 56% humidity, which is a “feels like” of 102. If you live in dry climate you are far better off. If you are in a humid climate, I am about as humid as possible, and the system works very well for me.

Below is a chart from Porsche. This is the designed performance of the 993 A/C system, based on outside air temperature. This is outside air temp vs center vent temp, after the car has cooled off the cabin. Why do they use this metric? As I mentioned above, the condenser is a freon/air heat exchanger. As a result, if this system had 9 of its condensers daisy-chained (instead of the 1 it has) they would only ever get the output down to ambient. Therefor the hotter the ambient, the hotter the output of the condenser, which means you are losing efficiency. The best way to gain performance would be a larger evaporator, and this one seem about as big as can be fitted in our cars very limited space. I am getting far better performance than this chart says a 993 would, with my blower on high and all my vents open. Just the other day my outside air temp gauge (VDO I put in) read 42C, and my center vent temp was 10.4 (been driving for 30 minutes). Based on this chart a properly running 993 system would not have seen any lower than 13C vent temps, if you extrapolate the chart out to 42C.


Below is an image from Honda regarding the performance of one of their A/C systems. Their test wants all of the vents closed except the center dash vent, and the cars blower on low. My CR System has been outperforming this as well, once the car cools off. That is also with my system on blower on high and all of the vents open. Lowering the blower speed and closing off vents will lower the open vents temp dramatically. This is because the air moves more slowly over the evaporator, which gives the evaporator more time to cool the cabin air, which of course gives lower vent temps.


Good A/C is about turning over the air in the cabin to cool it fast FAR MORE than a cold vent temp. To that end I have learned the car cools off fastest, and you are cooler faster, if you have all of the vents open including the footwell/crotch vents. The center vent temp will be higher this way by a few degrees C vs closing off the crotch vents, but the car will take longer to cool off and you will feel warmer if you close them, as you will have less air blowing on you. Opening the crotch vents also reduces the air volume coming from the dash vents. Do not be deceived into thinking cold air falls, the more air out of the dash the better. Trust me! Been down that path and back. When you see a vent temp of 18.3C (65F), you might think the system sucks. In my heat-soaked car, as it cools off, I see those temps. When I do though the intake air is often above 25C (77F). Remember the system can only take a certain amount of heat out of the air every time it passes over the evaporator. My X5 can run vent temps of mid 50’s F and the cabin can cool down to low 70’s no problem. Again, vent temp is a smaller part of a much bigger equation.

[Duc Hunter]

These A/C systems will see similar low pressures to a traditional belt driven compressor system. They will not see the same high sides though, at all. The chart below is a typical R134 ambient temp-to-pressure chart. I have never seen my high side pressure much above 170PSI, and my low sides, once the system has stabilized, are slightly below these. Also remember the evaporator’s temp is typically about what your low side pressure is. Notice the PSI of 50-55PSI at 38C/100F outside air temp? This means any R134 systems evaporator is “supposed to be” about 50-55F, or 10-12.8C, in that weather. This means, in theory, no R134 system can get near 4.4C/40F degree vent temps in 38C/100F outside temps. R134 is not as efficient/cold as R12 was, and yet R134 cars still cool off just fine. Just keep this in mind.
AC Pressure Chart | R-134a Temperature Pressure Chart

3M Crystalline tint is a key to helping these cars cool off. It can be clear, maintaining a stock look, and still block a substantial amount of heat. Without window tint the greenhouse in an old 911, especially the rear window, is just too hot for any A/C to overcome. This goes double if you have a dark car, like my black. You can have the windshield tinted too with this tint!! Even in CA (I had it for years). Here in Florida windshield tinting is illegal. 3M Crystalline dealers will still tint the windshields in the clear though. All my cars have had this done and I have never been given a ticket for it, even though I have been stopped. The 90% clear (CR90) 3M Crystalline Clear Tint blocks 34% of the solar energy and 90% of the infrared (heat) energy vs no tint. That is massive. Their next lightest is 70% clear (CR70), and that blocks 50% of the solar energy, and 97% of the infrared (heat). This is what I am running on my 3.2 Carrera all the way around, and on the windshield of my 3.6 Rat Rod. It is great at night, and the heat rejection improvement was massive. Again, now one ever notices. CR70 is so good even my mother, who HATES window tint on cars, had her Audi done all the way around. She does not notice the tint at all, and the difference it made in her silver 2017 A6 was again massive.

My car has the smaller 911 SC style side vents, and I added the 911 SC style center vent. As I understand it the center vents on the SC AND the Carrera are both fed by the same size tube 60mm, and that is the limiting factor in terms of airflow for the center vents. The same goes for the side vents, both cars use the same size feed tube to those vents (40mm). This is why I chose to go with the earlier style vents in my car, to keep it more “period.” I was confident I was not going to give away performance. I will say though the Carrera side vents are far better at directing the air where you want it. This was not enough for me to choose to cut my dash apart more though.

My Systems Performance

Living in Florida we have massive heat and humidity. In one test where it was 40-42C on my outside airtime gauge throughout the test, I got into a heat-soaked car where the evap temp was 41C (105.8F), the center vent temp was 51.4C(124.5F) when I started the system. Can we say STEAMY? 5 minutes in the evap was 9.5C(49.1F) and the vent was 15.5C(59.9F) and the car was comfortable (not cold, but comfortable). 10 minutes in the evap was 6.8C(44.2F) and the center vent was 13.5C(56.3F) and the car was comfortable. 15 minutes in the evap was 5.5C(41.9F), the center vent was 10.6C(51.1F), and the car was very comfortable. The whole time the system was on high blower speed and all of the vents were open. This outperforms the 993 noted above, outperforms the Honda noted above (remember they want low blower speed and only the center vent open which causes even lower vent temperatures). This also outperforms my 3.2 Carrera with the RetroAir system that I was previously very impressed with. Don't forget, it also out performs what R134 "should" be capable of doing. Both of the local mechanics I work with here say this is by far the best A/C they have felt in a 911 pre-996 and are eager to start selling and installing the systems. This is in a car with no roof insulation all, black paint, not door insulation, and very thin RS carpet. I am very satisfied. That said, when it is 95+ and 40%+ humidity and the car is heat soaked it will take 15-20 min before I think of turning down the blower etc. That is simply function of the heat and humidity, not a fault of the system. I am comfortable quickly though. These old cars have nowhere near the heat insulation in the interior of a modern car like my X5, and those cars also have more room for a larger evaporator etc. Given the design requirements of our old 911’s this system is amazing.

Installation Suggestions (random order)

PRESURE TEST YOUR SYSTEM! Do not rely on vacuum testing it. Here in the USA most A/C shops look at you like you are an alien when you ask them to pressure test your system with nitrogen. No one here does that it seems, like they do in Europe. That said, the ONLY WAY to see if you have leaky fitting is to pressure test. Here in the US they want to vacuum test-system, and if it hold above 30in of vacuum for a period of time (overnight in my case) “you're fine.” Wrong! Vacuum sucks. Duh right? It sucks fittings TOGETHER. Pressure pushes, and it pushed 2 of my fittings apart to show a leak the I finally pressure tested like Jonny had been begging me to do. That little “mistake” meant I was chasing my tail for over a week, complaining to Jonny about how my system was not cutting it in Florida etc. etc…..OPS! 2 other people I have talked with did the same thing, and when I convinced them to pressure test…they found leaks. My system see’s over 170PSI at times in use. So pressure testing to that pressure should be sufficient.

The power switch is bright at night when the system is on. I mounted mine under my speedo where my car originally had the mechanical odometers reset knob. I can see it day or night, but at night it is not pointed right at me. I tested it pointed at me and it was too bright (brighter than the high beam indicator when the system is on, and larger too). So that is something to consider. Mounting it in the center console may not be an issue. Also, if you are good with wiring, you could wire up a separate switch and light to the wires provided.

Do not try to over think the install. I.E. I made my own air intake tubes, thinking the system could inhale more air that way and perform better. The space in the frunk is VERY tight for all of the tube, blower, T’s, mixing vales etc. In the end I deleted my separate intakes and went to what I describe blow (the way CR wants it done). It's a very well thought out install as written by CR.

Note that as of firmware 1.13 the evap temp can be read through Putty. That said the system does not look at evap temp to prevent freezing yet. Jonny & Co assure me this is coming soon. Why is this important? Well if the evap temp goes much below 0C(32F) ice can build up on the evap. This will reduce airflow, which will cause more ice…and the system will freeze up. When this happens the cabin temp will rise, until the ice melts (meaning you stop for lunch and leave the car in the hot sun). How do I know? The RetroAir system in my 3.2 can easily freeze up that cars evaporator, and when it happens it’s not good. When CR implements freeze protection in the software, as I understand it, if the evap gets to the ETC (Evap Temp Cutout) the compressor will back off to hold the evap right above the ETC. In dry climates you could probably run the ETC well below freezing and never have an issue (like when I lived in CA). In my climate though, where there is humidity, my limit would be about 33F to be safe. Yes, that could be advantage dry climates, but that in no way means this system would not work in humid climates. In reality, during the day in the summer here, I am not seeing evap temps below freezing on drives less than 30 min. At night though I am.

Also with firmware 1.13 or newer testing is easier. You will be able to read the evap temp in the scrolling feed in Putty. This makes testing easier as all the data you might want is right there for you to see.

[Duc Hunter]

Please install the footwell/crotch vents. I know the tubing may look like an afterthought but DO IT! I and a few others are working on some 3D printed vents that will replace the tubes Jonny & Co supply, provide the same function, and accommodate the cabin air return as well. Until then though USE the ones provided!! They make a massive difference in how you feel and how fast the car cools off. Don’t try smaller ones that look better, been there and done that too. I cannot stress enough to use them. I did not want to and have learned just how amazing they are.

For max cooling completely block off your fresh air inlet under the windshield. I made a second cover for the intake vent. In cooler weather I will run with just the center of the 3 sections open for fresh air as CR suggests. However, for my summer's heat I block off the entire thing. Leaving it open is like driving with your window half way down and expecting your car to cool off. I have taken a few 6+ hour road trips with it completely blocked off, and there is never a desire to crack a window for fresh air. So help your system out in the summer, block it off. Especially if you are in a very hot and/or very humid climate. Once your system works well with it blocked off, you can experiment with opening the center and see the change in performance.

In hot weather unhook the heater tubes from the T’s and cap the tubes. Leave the T’s open in the cabin. I have found the blower can move slightly more air by doing this. It means no heat, so it’s not ideal in cold weather, but more air is always better. The bottom of the T is the suction side of the A/C system. Also, it is important to cap the heater tubes as hot air from the engine bay, or just from the heated body the tubes go through in the door sill, can leak into your cabin. I found a PVC cap at Home Depot that fit perfectly in the flexible tubes that come up from the door sill, with a hose clamp. Takes 2 min to change come winter when I want heat.

Keep in mind that the hotter your frunk gets the harder your system will have to work. I got some silver bubble wrap insulation from Home Depot and covered my system to separate it from the trunk, the best I could. This is the same material the WRC cars use inside to keep the heat out, and can be seen during their onboards up on the roof. Once the system has been running for a little while the whole area from the blower back to the dash is very cool, so keeping the trunk heat away from that area adds efficiency.


As you can see below, with the carpet in the insulation is essentially invisible.

Plug all open holes into your cabin! I found several, but my car had no interior when I bought it. This should be less of a problem for those with complete cars. That said the rubbers deteriorate over time. The ones around the T’s need to be checked during install, as a lot of hot frunk air can get in here. Door seals at the window are another place to check for hot air intrusion. Not just the roll up part of the window, but also the part behind the front quarter window. Mine at the top of the door panel were gone, and once replaced made a big difference. I found it very helpful to put a bright light into the frunk, pointing towards the cabin and look for holes in a dark garage. Also look for holes from the fenders into the frunk. Look for holes from the front wheel wells into the cabin, I found a few missing grommets there. It was truly amazing how much air comes in small holes on the freeway. Lastly it was shocking to me how much hot air can come in through the tunnel. There are rubber grommets in the front and back of the tunnel, that are missing in my car. Even with the carpet in place there is a big hole in the front of the tunnel in the cabin, and the shift coupler access hole in the rear. Hot air can come in these areas and wreak havoc on your systems capability. In my car reaching down by the gas pedal in a cooled off car you could feel warm air while driving down the freeway. Once covered the system worked even better.

Make sure your heater valves in the engine compartment close all the way. If they do not, you will be sucking in hot air from the engine bay all the time. Alternatively, if you don't run heat you can go to Home Depot and get a rubber cap with a hose clamp on it that fits the tube where the heat comes up out of the side sills into the footwells. Believe it or not, even just then hot air resting in the door sills (heater tubes flow through there) can cause heat to come into the cabin. In the summer I remove my heat tubes from the front kick panel area completely and do this. Note this is different than what I describe above (capping the heater tubes). You can do one or the other.

Get some kind of sensor into the evaporator that you can see on your dash temporarily for testing. I used these from amazon, thank to Jonny's suggestion. This is great way to see what the system is doing. In a heat-soaked car, before the air is cold, it is great peace of mind to see the evap temp crashing. You can also see that the evap temp will crash quickly to a point and then stabilize. From this point on, as the cabin cools off, the evap temp (and vent temp) will drop in concert with each other. My system will run a 5-8C differential from my evap temp to the center vent temp (based on where my sensors are mounted). As the system backs off this differential is reduced, and sometimes I even see my evap temp rise. The point being evap temp comes down as the cabin cools off. Watching it helps you learn if the system is working properly or if you have a setting wrong. This is how I saw the system backing off (evap temp rising when I was not cool), which clued me in to the fact that at one point I had set my CTS too high on accident. Once you are done testing this can be removed easily. Alternatively, if you are on software 1.13 you can read evap temp here, but you will need your laptop obviously. If a buddy is helping you test this will be totally sufficient.

Put a digital meat thermometer in the center vent. This is by far the easiest to read vent temp accurately while on the move. This can be removed once testing is complete.

Do I need a CR Alternator? Well, as I put in another thread about this systems.....heard the Meghan Trainor song "I'm all about that bass?" Listen, you will never get it out of your head now! Well for this A/C system it's all about that idle amperage. The stock alternators for 3.2's and 964's struggle to make 30 amps at idle. Ever see your lights dim at night with a blinker on? Well that means the A/C system will have to back WAY off at idle to cool your car. My stock 964 alternator makes 120amps make, and it could keep up on the freeway. In traffic, or around town though it struggled. By contrast my 964 CR alternator makes 100 amps at idle, and the 3.2 makes 75 amp. That means the 3.2 more than doubles how well your A/C can work at idle and in stop/go traffic over stock.

[Duc Hunter]

My environmental Testing

I have spent over 60 hours testing on the roads in Florida. Temps have been as high as 42C (107.6F), as recorded on my VDO dash mounted outside air temp gauge. Humidity has been over 80% at its worst, and never really below 35%. Testing has been going on since late 2018. Additionally, tests have been run at night where temps are below 26C (78.8F), and in winter when temps were in the near 4.4C (40F).

My car is a 1975 911S that is black and had no interior when I got it. Now it still has no insulation in the roof, has a black headliner, an RS interior (carpet glued to the metal, no dynamat or sound insulation at all, and the windows have 3M Crystalline tint on them.




In the picture above, see the holes in the shift tunnel riser? (edit; No you cannot because they are behind the seats.) This is the Rebel Racing Shift Tunnel riser. Keep in mind there is carpet under that tunnel riser with a hole just big enough for the shift linkage to come up through. During the process of getting my system working well I covered those from the backside with silver tape, as there was a ton of hot air coming in through there. This made a big difference in the cars comfort and was done before writing this post. Also, see my pretty RennShift shifter and the holes in the side of it, and how the top is open? Well in my car there is a lot of warm air coming into the cabin there as well. I can feel it as I drive on the freeway (once the car has cooled off). Temporarily putting painters tape over the holes on the side, and putting the little boot the shifter comes with on the top made a big difference in my cars comfort. I discovered this one AFTER writing most of this post too. This gives you an idea of how air leaks need to be found and closed. I was stunned at the difference these two adjustments made.

Settings - What they are, why they are important, and why I set mine where I did.

BMD=12.6
This voltage setting is the primary protection level for the battery, so the A/C ECU can protect the cars battery from over use by the A/C system. If the car goes below BMD for a set time (TBG) the system will switch the compressor to the CLO setting (see below). If you have a standard alternator this will happen often, as standard alternators do not make very much amperage at idle. Do not change this below 12.6, it won’t gain you anything and can cause your battery to be unhappy.

BLO=12
This voltage setting is the secondary protection of the battery. This kicks in when the car has been below BMD (above) for set amount of time (TBM). This is your A/C system recognizing that is needs to shut down the compressor for a set time to let the alternator recharge the battery. You do not want this happening at all for comfort reasons. If this happens the car will get hot quick, and if the car is heat soaked this behavior will prevent it from ever really cooling off. Keep this in mind as you read on below. Do not change this, there nothing to be gained here.

CLO=05
This is the lowest your compressor speed will run before the car reaches your desired set temp (CTS), so as to try to not over cool the car. Raising this just means the compressor could over cool the car when it nears the CTS, and then the compressor will shut off to not over cool the car. IMHO it is better to have the compressor be able to run at a very low power to keep you cool, vs cycling on/off to do so. Leaving this at around 5-10% means the compressor can run all the time, and at a low enough level so as not to over cool you or the car. This is especially true in the spring and fall, when extreme temperatures are not a problem, and you want the system to cool the car comfortably. The ECU will adjust comp speed and amperage for .1C changes, so leave this at 5%-10%. Performance won’t be gained here by setting it higher.

This is also the setting the ECU will switch to if the car has gone below the BMD voltage setting for a period of time. If you do not have a CR Alternator, then you may want to try raising this to a place where your cars alternator can maintain 12.6V or more at idle. This is explained more below. In my car, even at 5% the car will still cool off, though not nearly as well as it does running in normal mode. Going lower than this means the car cools less well at idle in CLO mode.

CTG=110 on a 964 Alternator
Remember that my RC 964 alternator produces 110 amps at idle, and even so at this setting the car stayed above 12.6V. I have a Steve Wong chip, which sets the idle at 990rpm I think, and probably helps. This means the system is not losing performance at a stoplight, or in stop/go traffic, at this setting. With this at 80 and the CHI at 65 I never saw much above 65 amps.
With this set at 110, and the CTG=90 I would never see above 107amp at idle, and the car maintained over 13.6v. As a result, CTG=110 and CHI=90 should be enough for everyone with a 964 alternator. If you have a 3.2 alternator, then you might need slightly lower settings, because that alternator only makes 75 amps at idle and 150 max. The goal is to maximize performance (CHI & CTG) while keeping idle voltage above 12.6V with everything on the car running. Watch your car, and as long as at idle the compressor can back off w/o shutting off and maintain 12.6V you should be fine.

If this setting gets raised high enough that, at idle, the car cannot maintain 12.6V with all its electrics on, the compressor will go into low mode (CLO) to save the battery. In a warm and humid climate this could cause the cabin almost stop cooling off the cabin in a car that is heat soaked. This is not at all bad, unless you’re stuck in traffic for an extended period of time. Below I talk about testing the system and address this again.

If you are running a factory alternator getting these settings right gets challenging. Never set this above the max of your alternator. 3.2 alternators are max 90 amps total, and 964’s are 110amp max I believe. The alternator must run the car too, so I would say never set these above 75 for the 3.2 and 90 for a 964. Also keep mind older alternators, or rebuilt ones, may not meet factory spec’s. Really safe numbers would be 65 for a 3.2 and 85 for a 964. These alternators also require adjusting the CHI down (see below). The goal is the same, and switching to BLO mode is fine at idle, but shutting off is not ideal.

CHI=90 on 964 alternators
This is the compressor speed. If for some reason amp monitoring is lost, then this ALSO prevents the compressor from using too much amperage and draining your cars battery. So this should not be run too high, just in case. As a result, CHI and CTG need to be tuned in concert with each other.

Ironically in testing with CHI=55 the compressor speed can show 55 and amperage draw can be anywhere from 35amps all the way up to 65 amps. Seems strange to me, as I would think more speed would cause more amperage. They seem to have influence on each other, and yet change separately little. Likewise, a CHI=90 can use as much as 107amps of CTG. IMHO this can only be raised to a point where your car, with all its electrical components running, can maintain 12.6v at idle. Raising it higher is asking for problems. In my testing going above 90% doesn't seem to improve performance either.

For stock alternators this needs to be backed off. In my testing CHI=65 never give me above about 55amps of draw on the alternator. As noted above, keep in mind that the alternator has to run the car too. In winter that may mean the A/C, window defoggers, and lights (think rainy night). You need to watch the amperage and voltage in testing.

CTS=19
This is the desired cabin temperature. This is very dependent on where your cabin sensor is mounted. You must test your specific install and see where the CTS needs to be set so that your car is comfortable to you. For some installs I have helped with this could be 15C, others it could be 25C. More on this in the testing section below.

KP=65
This is essentially how hard you drive the compressor until it gets close to your desired cabin temp. Setting it very low means the computer backs off the compressor well before your car gets to its CTS. The 4% my system came set at exhibited this issue. As noted above, I built a digital model of the PID to analyze this behavior virtually. Being at 4% meant the car could not get to CTS in anything but low outside air temps and low humidity, and it started backing off the compressor about 20C above the CTS. With this set to 30% the compressor is going be at full power (based on the CTS and CHI) until about 3.5C above your CTS. In less humid climates this should work fine, but for Florida it did not. Setting it to 65% means the system runs full power until about 1.75C above the CTS. Going above 65% did not improve the cooling much, but in more temperate weather it hurt the systems resolution to back off and maintain cabin temp. Meaning you got too cold as the car reached CTS. 65% should be plenty here, and less in climates that are not as humid. Come fall I may back this down, but for now I am leaving mine at 65%.

[Duc Hunter]

KI=.01 & KD=.01

These 2 settings are a little complicated to explain. Basically, they control how quickly the system backs off the compressor speed and amperage as the car approaches the CTS. One of them controls how hard it tries to back off as it approaches the CTS, and the other is how much it tries to predict/prevent an over shoot in CTS temp. PID controllers are used for all kinds of industrial uses, where they can over shoot easily, but on our case it won’t. Just change these to my settings and don't play with them. There is nothing really to be gained by moving away from my setting, that I can tell. Increasing them back to the 1.12 settings can slow your cars ability to reach CTS. In dry climates, or mild weather this might be desirable. I will have to re-test this when it cools off here (October maybe).

System Comfort Setting Testing

Your settings will need to be set specific to your desired cabin temp AND your install. Mostly, where and how your CTS sensor is mounted influences it's specific reading vs where you personally like the cabin to feel. This takes some testing, and specific testing in not super-hot temps (i.e. at night in the summer should be fine).

Setting Goals

Cool the cabin quickly

Remember this system does not have a knob you can use to warm the car up with if you (or your passenger) get cool. We need a laptop for that.

Find a CTS that is comfortable year-round. This way in the summer the system will just work, but in spring and fall it can be used and not over cool the car. Few things are more annoying than manually switching the system on/off because you are too hot and then too cold. The system should just work, and mine does. I may play with fan settings to blow more/less air on myself as the car is at CTS, and I may open/close the footwell/crotch vents for comfort…. but that is it.

Ensure your settings do not cause the compressor to shut off at a stoplight. If possible, you also want to find settings that do not cause it to get warmer at a stoplight, but this may be tougher in cars without the CR Alternator, if you are in a very humid climate.

Things to Try, or Look For

If you need to warm your car up during testing after its been cooling off, just roll down window and watch the vent temp rise, then close the window. Simple, but it took me a lot of testing to remember I could do this. You can easily get focused in the weeds of your numbers and system and forget simple things.

As you watch Putty you can leave it in regular mode (no need for DIAG mode). In this mode you will see something like the picture below. I go into more detail farther down about some of this, but as an overview;

V is the current voltage. At anytime, if this goes below 12.6V for more than 10 seconds the system will go into BLO and back off. While running on the freeway, all electrics on, you do not want this below 13.6V ever. As I said above, at a stoplight BLO=5% should still be fine.

I is the compressor current draw. CTG is the max this can ever be. This needs to be only as high as required to reach your max CHI setting. If this is too low you won’t reach max CHI, if this is too high, there is nothing to be gained. Mine is set to 110, because in testing I never saw above 107.6amps draw on the compressor when it was pegged at my 90% speed setting.

S is your CTS, your desired cabin temp.

C is your cabin current temp. This is where your system reads its temperature to see if it is at CTS. This is what you will watch this and see what it says when you “feel” cool enough.

E is your evaporator temperature, as measured by your sensor in the evaporator. If you do not temporarily put a dash mounted gauge to read this as I did, then this is where you read it. When I took this picture I was on firmware 1.12, where ever temp is turned off. As a result it always reads zero. Firmware 1.13 turned this back on.

D is you’re your CHI, or compressor’s speed. You will be watching this to see if it is at your CHI setting when the car is hot, and you are on the freeway. There are a few reasons this could not be at your set CHI. If your alternator cannot make enough juice to get the compressor speed up to where you have asked for it to be (I.E. at a stop light and idling) the system will back this down to lower the amperage. If you are running and your amperage is pegged maybe your CTG is too low to support your set CHI. Also, your system could think it’s near your CTS and be backing off the compressor.

In the above image I had just started the system, so you can see the amperage rising as the compressor is running at 90%.

Always use the center dash vent for the vent temp measurements. It is often the warmest and is the one most often used by people. Plus, you and your passenger can read it during testing.

Again, if you are on software 1.13 or above the Putty will give you an evap temp, and you may not want/need to install one as I suggest above. I like mine as much of my testing was done solo, and reading Putty while driving is less than ideal, clearly.

[Duc Hunter]

Testing Procedure

Part 1 CHI, CTG, etc……

Put your system in high/low mode, not climate. This means your system will essentially run on high as long as it has enough amperage. If it does not have enough amperage for high (think at a stoplight) it will switch to your BLO setting. If it still cannot maintain voltage (which at BLOW=5 it should even on a stock alternator) it will switch off.

Set your CTS=5 for this portion of testing. Yes, that means a cabin set temp of 5C or 41F. The idea is you want it to run at full power all the time for this part of the testing.

Take a friend and have your system hooked up to a laptop with putty running so your friend can watch it as you drive, or visa-versa. I found running a USB extension from the Trunk to the passenger footwell worked great, and there are a few grommets over there you can use.

Step 1 – Finding your best MAX Settings. When it’s hot out……on the freeway test your max power settings. Watching Putty, you will see if the car can maintain good voltage when the A/C is running at full power. This requires a hot car. Set your CHI and CTS to 55 to start. You want to see the compressor running at full speed (max CHI), and that you have over 13V. If you go below this your car won't be charging well. You can decrease/increase the CHI and the CTG in increments together to find the max your car can handle on the freeway while keep yourself at 13V or better. Ideally this is also with ALL of your electronics running. Remember from above CHI and CTG are somewhat linked. I don't think anything above CHI=90 is ever needed. In my climate it is not. Also remember from above that a CHI of 90 can use up to 107 amps in my car. If I had set my CTG at 100 then I would see the CHI of 90 sometimes, and others it would be in the low 80’s because the system would hit the limit of the 100 amp CTG first. Basically, you want to see the D readout in putty = to your CHI, while maintaining over 13v, and note the max amperage. Lower your CTG be just above the max amperage you are seeing in Putty running at the set CHI. This means your alternator is driving your A/C system as fast as it can and still able to keep your car charged.

Step 2 – Testing your idle settings. On a hot day heat soak the car and drive it. Find stoplights or sit in your driveway or a parking lot, and see how the car does. If amperage drops below 12.6V for 10 seconds the car will switch to BLO. You will see this in putty as the D will go from your CHI setting to your BLO setting, with my settings this would be from 90 to 5. Does the cabin get warmer at BLO? Does the evap temp rise? If it does not, or if it only slightly does don’t touch the settings. My experience is that even on BLO=5 the heat-soaked car won't get warmer at a stoplight, and a stock alternator can maintain enough voltage to stay above 12.6V. It may not cool as well when really hot but that is ok at this point. Also, once the car has cooled off much at all this will become irrelevant, as the cabin is cooler. This is only really an issue when it is REALLY heat soaked and you just started the system up. Remember raising BLO above 5 means the system will not have the resolution to back off in cooler weather, or near your CTS. In the heat of the day in summer it may seem minor, but it will be a huge benefit most of the rest of the year. If you choose settings that are too aggressive for your alternator, esp. the BLO, your compressor will switch off. This will warm the cabin very quickly and is not good. If this occurs, you will need to back off your BLO. If even at BLO=5 you find the cabin just gets too hot at a light then you may need to back down your CHI and CTG, to prevent the low voltage issues. This would only be for stock alternator cars. Back it off so that the car does not drop out of BLO mode at idle.

Step 3 - Setting your CTS
Here you are trying to determine the cabin temp setting. Do this in shorts and a t-shirt. Drive your car at night, or on a cool day, with the cabin blowers on high and all your vents open. You are trying to get the cabin “too cold.” You need to be watching your cabin temp sensor reading in Putty and noting your actual body temp as the cabin gets cool. As you get cold turn off your footwell/crotch vents (as they REALLY cool your body) and keep going a little. Note at what temp you want to turn the system blower down and are “too cold.” Whatever your cabin temp sensor says in Putty is what you will need to set your CTS too. In my case as my cabin temp fell down to around 19C I was thinking it was too cool, and I was turning the vents off me (on blower low). I did this on about a 30-40 min drive at night. As I noted above, some folks I have talked to needed to set CTS=25, and others were down around 15.

As an example, when my CTS was placed in the driver’s side tube between the blower and the mixing valve, nearest to the blower I needed to set it 14-15C. The cabin was comfy but the sensors placement meant that is was getting cooled by the its proximity to the cold blower. Yes, once the car is cooled off the entire blower looks like a cold beer can put outside on a hot day…covered in sweat (condensation) and cold. As the system runs and cools, all of the tubing gets cold too (think vent temp). This cools the sensor well below the actual cabin temp you feel. When I moved the sensor to a place in the cabin, out of the intake air stream I needed to set it above 26C. Right now it is placed in the bottom of the driver’s side mixing valve and 19C works great for my car. 19C seems very cold on paper, but the crotch blower is right under it, and that tube is cold (again think vent temp). Also, ensure the metal end is in the airstream, and not touching any metal or the tube its placed in. If it is then it will read the temp of the tube, which may be far cooler than the air stream once things cool off. Bottomline, this is why you must make sure you mount your cabin temp sensor properly, and must custom set your CTS based on your systems install.

[Duc Hunter]

Part 2 - Home Stretch

Switch to Climate mode and change your CTS to the temp you decided was “cold enough” above. Go drive again at night and see if the setting you chose above works. Initially I chose 18C, and then I decided on 19C after testing in Climate mode. In fact, setting my cabin temp at 19C meant the compressor would slow to 5% (CLO setting), the vent temps were in the high 30’s/low 40’s F, and I did not want to turn the system off. The CLO being 5% is also key to be able to use this system to aid with defrosting in the winter, which you WILL do. Trust me. If you watch your evap temp you can see it rise as the system backs off to not over cool the car. At times I have seen mine come up from -4C to 6C to maintain cabin temps on a cool night.

With your CTS setting done, based on your unique install and after this testing procedure, the rest of the settings noted here should take care of the systems performance when the cabin is above your set temp.

In Conclusion - Finally right?
With the settings you have arrived at your car will run at max compressor speed/amperage, based on your alternator’s capacity, until it gets within ~1.75C (~2.8F) of your CTS (desired cabin temp). This is essentially as fast as you can cool the car given your equipment.

Once the cabin temp sensor is within ~1.75C of the CTS the system will SLOWLY start backing off the compressor amperage and speed so as to not over shoot the CTS temp. It will also not feel “too cold” where you want to move the vents off your body as the temp reaches your cabin temp, except you will want to close off the crotch blowers before this point. As your cabin reaches “comfortable” the crotch vents become too much as they REALLY cool your body off.

Your car will maintain good charging on the freeway, and can back off as needed at a stoplight, and should not have to resort to turning the compressor off/on (alternator dependent in some cases).

There were many times where I got frustrated during my install process and vented to Jonny questioning the systems performance capabilities. Each time I learned a few things (some of which I have shared here). 90% of the time I had made a mistake though, or missed something, or had an air leak. Only once was it CR’s fault per se. I was one of the early customers who received some improperly machined fittings, a problem long since solved. That was literally the only issue that was not caused by my installation. On the other installs I have helped counseled, the same has been true. Installing an A/C System is not a simple thing. The CR kit makes it very doable, but there are a lot of moving parts. Their instructions are laid out pretty clearly. Between following them and reading this “little” guide I think everyone will find it a very satisfying project. Be methodical though, or as pilots say RTFM!

Again, I am by no means an expert. I just wanted to pass along what I had learned. I will try to update this as I learn more, and hope everyone find it helpful. Thank you again to Jonny & Co for a great product, and great support.

PS – Jonny, winter is coming (October/November for me) so get working on that heat! [/INDENT]

[dyerkes]

Hi Chris,

Great write-up! Please put me down for a pair of the 3D footwell vents when you start producing them.

Regards,
David

[sedouglas72]

Thank you very much for the detailed write up. I'm probably going to go this way with my build.

[MikeD930]

Good job on your report and suggestions! I also have the CR A/C and like it a lot since most of its weight is up front (a condenser is in the driver side rear wheel well). I may be interested in a set of 3D printed footwell vents.

Have you ran your car during winter months with heat exchanger in exhaust headers? If yes then how was the heating? I don't have a heater box on my headers but I did tried using a flex tube along the headers but had no luck getting any heat due to ineffeciency of heat transfer and cold air coming in thru the A/C blower box.

[pete3799]

Great write up.
I appreciate the effort put forth. Very informative.

[Jonny042]

Thanks for taking the time to post this for all of our benefit!

This seals the deal for me. I will, one day, be a customer of Classic Retrofit's AC system.

[lvporschepilot]

This is fab. ty!

[Duc Hunter]

Originally Posted by MikeD930

Good job on your report and suggestions! I also have the CR A/C and like it a lot since most of its weight is up front (a condenser is in the driver side rear wheel well). I may be interested in a set of 3D printed footwell vents.

Have you ran your car during winter months with heat exchanger in exhaust headers? I also live in North Florida, so winter is in the 40’s-50’s during the day, and a few days in the 30’s at night. Pretty mild. If yes then how was the heating? I don't have a heater box on my headers but I did tried using a flex tube along the headers but had no luck getting any heat due to ineffeciency of heat transfer and cold air coming in thru the A/C blower box.

I did run it in the winter with my B&B exhausted on my car. The heat is not bad, but is not as god (obviously) as my heat in my 89. That car has the blower in the engine bay AND the foot well blowers. Remember the CR A/C. Blower moves as much air on blower setting #2 as the factory air system every did, and on #3 it move a lot more. Also the mixing valves, when set too use heat, take all of the “cabin intake” and switch it to sucking through the heat exchangers. Also keeps in mind, these cars never had great heat with headers. The stock cast manifolds provide good heat for sure, but headers never have. Hence CR is working on electric heat.

[MikeD930]

Thanks for the info on the heat. I'm looking forward to see how well electric heat will perform and am curious where it will be mounted. I hope it's not inside the blower box (dread to remove the blower box with A/C already charged)!

[Duc Hunter]

The rumor is, and my best guess is, that the heaters will go in the foot well area where the foot well blowers and later model cars went. This only makes sense as putting in there will add to the blower capability, and is a good place to put the heat, while keeping it out of the way of the air-conditioning system.

I also just want to reiterate to everyone to please check the chart above about r134. I talked to someone today who is complaining about not getting 40° vent a temperature of 100° day. Per the chart on a 100° day our 134 will be between 50 and 55 psi on the low side. As I understand it this means the evaporator is around 50 to 55°F. Therefore there is no way with R134 to get 40° and temperatures in 100° ambient outside air temperature.

[famoroso]

Wow. Nice write up. Kudos on a thorough job.

What do your printed footwell vents look like?

Also, did you do anything with the lower center, “bow tie” vent?

Thx,

[Duc Hunter]

Originally Posted by famoroso

Wow. Nice write up. Kudos on a thorough job.

What do your printed footwell vents look like?

Also, did you do anything with the lower center, “bow tie” vent?

Thx,

Thank you for the KUDOS. I was just trying to give back to everyone here like they have given back to all of us, and to give back to CR as they have been great to work with and have a great product!

Lower center “bow tie” vent? My car did not have one, nor did it have the hole for a bow tie vent. NOw I have though about plumbing something into the cabin to use the now capped 60mm output on the blowers drivers side. The best solution comfort wise would be to move the fuel gauge to where the clock is, and mount up a circular vent in to the hole the fuel gauge currently sits in. This makes for the easiest (and maybe only doable) routing of the feed hose, and could get a lot more air into the cabin and on the driver. That said I am not sure it is really needed yet. I am not sure it could be routed down to here the bow tie vent is front eh factory. IF it could, I would point it straight into the cabin unlike the bow tie vent. Remember this system as 2 x 60mm intakes, and mine is feeding 3X 60mm outputs, and 2 x 40mm outputs.

The 3D printed footwell vents? Right now they are far too un-pretty to even post (and I made my first set wrong). Essentially they cover and mimic the bottom of the mixing valves exactly at their base. The hole closest to the passengers is left open, as this is the systems intake. The hole farthest from the passengers is covered and has a duct that comes up, and turns 90 degrees to blow at the passengers. This tube is 1/2” away from the hole closest to the passengers, to permit good intake flow, and extends well past the intake hole, to prevent the intake from sucking in the air you want on your body. It also only extends down from the underside of the dash a total of 3”, so it does not get in the way of your feet (I a 6’2” and have a size 12 shoe). This seems to flow enough air, and be up and out of the way enough. I am hoping to get some new prototypes built soon that work better than my first set. Using online CAD programs is a bit kludgy though.

[famoroso]

These aren't the footwell vents you're referring to?...

Fresh Air Vents got a new batch

[Discseven]

Exceptional AC posting Chris.

Am curious, and forgive me if I missed it, what controls electric AC's compressor's cycling if there's no temp sensor inside evap?