|
Post by majestek12 on Apr 20, 2023 15:45:03 GMT
Hello and thanks in advance for any assistance or knowledge. I'm working to restore a 1947 Nash-Kelvinator Fridge, a family heirloom bequeathed to me by my grandfather. It's an R12 capillary, hermetic, non-defrost system. It ran for years after my grandfather dropped in a 1959 compressor but recently the frost line is inching back towards the evaporator inlet, indicating a potential leak and evap plate temps are increasing. I recharged the unit with a few oz of R12 (legit R12, not 12a, hotshot, etc.) using the frost line method, I know this is not preferred but I didn't have a scale handy and being that it's not the original system or plumbing it would be a bit of a guess anyways. The unit does cool down nicely, going from ambient-empty 65°F to 28 only takes about an hour, which on an old beast like this seems pretty good. Similarly, when undisturbed, average air temp at four points in the chamber will cool at about 1° per minute when half-loaded with drinks, yielding a three to four minute run time. But, logging evap plate temps over multiple 24 hour periods shows a slowly increasing trend after the recharge - pretty sure I gave the charge enough time to settle in and balance out. Condenser coils are clean (were a bit dusty but recently knocked off with a horsehair brush), unit has plenty of airspace around it, and there are no evap or condenser fans to worry about. Thermostat was replaced by my grandfather, it is an air-sensing capillary style stuck to the side of the evaporator plate. I've spent a good amount of time trying to track down the expected leak using a wand (admittedly a cheap one) as well as soap but it seems small enough to not be picked up. Similarly, I can't locate any signs of oil. All the usual culprits such as junctions, service valves, valve stems, etc. have been inspected and sniffed extensively. I've got a sufficient supply of R12 to fully service it multiple times but have not had luck finding anyone near me who is willing to try and identify the leak or do more than what I have done already. When calling around, those who have an understanding of what it is aren't willing to work on it, and those who are willing require the "year, make, model" so they can pack parts before they come Lol. When operating, high side is about 130 and low side is about 5-8, with chamber at temp, normal thermal load, and ambient temp of about 75° F. R12 is not my expertise, but from the limited Google info I can find this is about normal on the antique system. If possible confirmation of this being normal, and not a indicator of a blockage or non-condensables would be appreciated. I've heard partial blockages on such systems can cause low-charge-like symptoms. The evap is aluminum plate and forms the icebox, condenser is steel, and there is a liquid line filter drier of unknown vintage installed. Temperature difference across the drier is minimal, only about 1°F so I don't it is clogged. As far as leak detection, some have suggested using dye but I've never been a fan of the mess it makes or the possibility for it clogging. It's a simple, small system, so I'd like to keep it just freon and oil. I'm getting about down to my wit's end though, and my constitution is waning. Any suggestions on how else to detect the leak? Is 130/7 PSI during operation about right for such a system? I can't get good info on probable head pressures so it's hard to confirm if there's a blockage or I need to dig deeper into a leak. Fridge holds immense value as an heirloom, so I'm exceedingly and humbly thankful for any assistance!
|
|
|
Post by ckfan on Apr 20, 2023 18:27:58 GMT
We have a few members here who can answer your questions better than I can. It certainly sounds like you’ve done your homework though. I am also not familiar enough with R12 to know if those pressures sound right or not.
|
|
|
Post by majestek12 on Apr 20, 2023 20:10:46 GMT
Thanks, I'll keep my fingers crossed, I've reached out on several boards, including HVAC/refrigeration focused ones, and while assistance has been great experienced knowledge a bit lacking.
|
|
|
Post by turbokinetic on Apr 20, 2023 20:25:25 GMT
Hello and thanks in advance for any assistance or knowledge. ..... Fridge holds immense value as an heirloom, so I'm exceedingly and humbly thankful for any assistance! Hi there and glad you're here! You have done exactly what I would do as far as the charging. Because the system has been modified, you cannot use a charge by weight method and used the frost-line method which is absolutely correct. As for the slowly increasing evaporator temps on the cold-plate while using an air-temp sensing thermostat; that is normal. The compressor will run less often while the cabinet is down to temperature, leading to a warmer average evaporator temperature. If the original thermostat design was in use, where the actual temperature of the evaporator was being measured, it would be more consistent. Very small leaks can be extremely hard to locate. You may have porosity somewhere which is leaking an ounce every few years for it to take as long as it did, to leak down. I would monitor it and see if it is indeed losing frost line. It may go for decades. If it leaks down in shorter time, you'll have to troubleshoot it with high pressure nitrogen. The pressures created by R12 at room temperature are around 75 PSI which is hardly enough to find small leaks. I recommend you use the maximum test pressure shown on the system. If nothing is shown, 175 PSI should be safe but not above that. Note that if the leak is in a trap area, oil may be inside the line and may "seep" but not show bubbles in soap. Sometimes you need to move the machine in different positions to allow any pooled oil to move to a different location to get bubbles. The bubbles from microscopic leaks will look like a tiny grain of rice growing, or a pinhead size foam droplet. Worst case, you might have to take the whole system out of the cabinet and immerse the parts in water with it charged with N2, to locate the leak. The pressures sound good. They are determined by the temperature of the condenser for the high side, and the evaporator for the low side. For frosted evaporator designs, that 4 or 5 PSI low side is in the range one would expect. I haven't specifically done a video on leak checking, but many of my general repair videos cover some of that. Hope they have been helpful. Sincerely, David
|
|
|
Post by majestek12 on Apr 21, 2023 12:46:30 GMT
Many thanks turbokinetic. It is relieving to hear the increasing average evap temp isn't unexpected, and the logic makes sense that it will have to run less, and cool down less, to maintain temp once stabilized. I have noticed that the run times are about the same (3-4 minutes), but perhaps this makes sense. That being said, I've not analyzed the run times or cycles per hour as closely as the evap temp, and may have missed a decreasing average in them as well.
Current thermostat settings maintain an average top-shelf (adjacent to evaporator) temp of 30-31 and bottom shelf 35 but the evaporator never gets above freezing (lowest is 12°F, average is ~22°, high ~27°). As the unit maintains stability, will the frost line move at all due to increased evaporator average temp, or will it not because the evap is still below freezing? Intuition tells me it should not move but I'd like to educate myself further.
Thanks for the tips on pressure testing if need be. Maximum test pressures on the serial plate attached to the refrigeration chassis state factory test pressures of 195/high and 140/low. Without a current way to isolate the sides I'll probably test at 140, though hopefully won't have to! Thanks for the tips on the potential oil traps, I would not have thought about that.
I hadn't specifically sought out your videos but after doing so realize I came across several during my prior searches. You have a unique voice.
|
|
|
Post by majestek12 on Apr 21, 2023 14:03:40 GMT
Two more questions while I have your ear.
The capillary tube on this unit is bonded to the return line and for the majority of the run is outside the refrigerator cabinet, only entering at the top just before passing through the wall and attaching to the evaporator plate. How much variance can I expect to see in high and low side pressures with changes in ambient temp - is referencing an R12 PT chart valid?
I've seen references that the frost line method is not preferred compared to weighing in a charge as if it is frost line charged in the winter it'll be undercharged in the summer, and vice versa. Obviously this is in comparison to weighing in a charge for an original system with a known capacity, but I'd like to know for any future investigations and avoid excess wear on the compressor from an overcharged system. For what it's worth, this was on that HVAC site and may pertain to external units or systems that see greater ambient temp fluctuations.
I really do appreciate the assistance, time, and knowledge and am humbly thankful. Backstory to this fridge, as it seems most on here will have an appreciation for it, is that my grandfather spent 40 years working as a refrigeration technician for the Army at the Fort Belvoir proving grounds. Back in his day home air conditioning systems were relatively new and he had a side business installing them in the Alexandria Virginia area. As part of this and due the tough times and frugal wallets operated with neighbors on a commodity economy. Family legacy is that he installed an air conditioning system for a new neighbor who recently moved from Detroit, and had previously worked for Nash Kelvinator. This fridge was supposedly a prototype or pre-production unit that had been decommissioned, but taken by the neighbor as his wife was a baker and wanted it for use as a bread and pie box. My grandfather wanted a beer fridge for the basement and took it as partial payment for his labor. Being in the industry he had access to refrigeration components cheap or free and use some of these to restore its operating condition. Stories in my family can be akin to fishing-tales but after sending pictures to John at Antique Appliances in Georgia he stated he's never seen a unit quite like this one. Apparently it has design language similar to units produced for Canadian markets, but John noted the serial plate is clearly of Detroit origin. Whether truth or embellishment it's an interesting facet of the story.
The main reason for my appreciation of your support is that beyond honoring my grandfather's legacy it's been the frosty soda and beer fridge for three generations of my family; my grandfather and grandmother, dad and aunt, my wife and myself, and will one day be passed down to my son. On my will I've bequeathed my grandfather's old Wurlitzer jukebox to my daughter, and your assistance helps to ensure I won't saddle my son with a non-operating fridge when parts and refrigerant are even more difficult to obtain. In the meantime, which is hopefully a long time, my son and daughter will be able to share the experience of grabbing hot-day sodas, and beers one day, while being in some fashion able to touch the hands that came before them.
|
|
|
Post by turbokinetic on Apr 21, 2023 15:04:55 GMT
Thanks for the back-story! That is a very interesting bit of info on the history of a Nash-Kelvinator prototype. Could you share pictures of it here? I would like to get a better understanding of how the system is constructed knowing it's likely a custom setup.
The best way to charge is by weighing in a charge, 100% accurate. This works if you have a completely unmodified factory system and use the original refrigerant it was tested with, to get the charge weight. In the case of a modified system, you won't have a charge weight to work from. Furthermore many of these systems have a theoretical maximum charge listed on the nameplate, but not the exact amount in this system. You can overcharge them massively and easily by blindly going by the nameplate on these old systems.
The frost line absolutely does change with ambient temp and cabinet temp. It will get higher in hot weather, and lower in cool weather, in general. Normally you see this during times where the run cycles are long and once the system is stabilized, it is pretty constant.
Glad you've been enjoying my videos. As for my accent, I've lived on the for decades, spending time in many states and countries. You just pick up little bits of all the accents you're around, I guess.
|
|
|
Post by majestek12 on Apr 21, 2023 16:54:53 GMT
Gladly; some pics can be found here drive.google.com/drive/folders/1qA67cBlmNbbwqw3tMoA_JF8FA56F-guYAs you mention, it's definitely a custom setup. Best I can make of the serial plate on the non-original compressor makes it out to be a 1959 which corresponds with family history. On the compressor itself, the low side service valve is welded to the casing, so I expect it is original to the compressor, though of course not original to the refrigerator. High side is definitely custom plumbed/retrofitted, with the service valve being legged off the main high side line with a T, after the condenser and before the connection to the capillary tube. Valve itself is connected to the compressor housing with a custom bracket. I wasn't able to find the specific high side valve through Google, but researching the Kerotest name and finding similar parts indicates that it's a packless nut design intended for refrigeration purposes. There is an inline filter drier after the condenser and before the cap tube. During my copious leak checks I made sure to sniff around all compression fittings, solder joints, valve stems, seals, etc. After research, seems that if we have to open this system up it would be best to remove the high side valve and replace any compression fittings with solder joints to reduce potential for future leaks. Duct tape on the front of the refrigerator casing covers a recessed hole of about 2" diameter that was apparently intended for the original thermostat control. I've been looking around for a Kelvinator emblem to cover this and make it more aesthetically pleasing. Through Google I've been able to find one other model that's similar to this, but not the same model number. Closest I could come was a AC-L, while mine is CA-R. L and R designations are intuitively indicative of door swing, but I've not been able to find another CA model. Similarly, compared to the AC series, the Knight emblems are different and mine has Kelvinator script on the bottom of the door. The kick plates and handle appear the same. www.reddit.com/r/Antiques/comments/pfzcws/help_with_year_of_vintage_refrigerator/Shiny workbox is my own recent addition; added a three-prong with ground to the original two-wire. Junction box on the factory schematic is actually a porcelain unit affixed to the bottom of the cabinet inside the compressor cavity. Thread below has a similar design with more info and pictures of what the original compressor probably looked like. www.applianceblog.com/mainforums/threads/1940s-or-1950s-kelvinator-wiring.58403/I originally thought that the enamel paint was a goner, but some careful buffing and polishing brought it back unexpectedly well - I guess it was protected by the 60 years of cigarette smoke film ;-) EDIT - some additional added, such as link to wiring information.
|
|
|
Post by turbokinetic on Apr 21, 2023 17:18:58 GMT
Gladly; some pics can be found here drive.google.com/drive/folders/1qA67cBlmNbbwqw3tMoA_JF8FA56F-guYAs you mention, it's definitely a custom setup. ..... I originally thought that the enamel paint was a goner, but some careful buffing and polishing brought it back unexpectedly well - I guess it was protected by the 60 years of cigarette smoke film ;-) Very nice! That compressor is a Tecumseh low-speed design. It's as durable as the original Kelvinator was. The only issue with it is that the system is assembled with flare fittings and therefore there is a chance for leaks which would otherwise not be there. That HS valve is very likely a packless style valve although the handle looks a little different from what I have seen before. The body does appear to have the oversize gland area where the sylphon for the packless valve will be located. It does look nice all polished up, too!
|
|
|
Post by majestek12 on Apr 27, 2023 18:49:12 GMT
Bringing this back to the top, progress has been made though it has produced some more questions.
After being recharged the fridge has been running reliably though I could swear the frost line on the evaporator was beginning to drop ever so slightly after a few weeks, but I wasn't sure if it was or just my mind or ambient conditions playing tricks on me.
I started poking around and noticed what appeared to be oil residue under the HS service port cap. I thought I had cleaned the entire valve body earlier in my hunt but wasn't sure, and it was so slight that I figured it could be the reflection of my flashlight. Making sure the valve was closed I removed the cap and heard a brief "psssh". Pretty telltale, but I grabbed my sniffer and it corraborated the valve/port as the source of the leak.
This is a huge step forward, and while the valve obviously has to be removed from the system I'm not sure of the best way to do so. The valve is T'ed into the HS line with a roughly 12" pigtail. Can the pigtail be crimped relatively close to the T (not close enough to disturb the brazing at the T), cut near the valve, and brazed to eliminate the valve? Opinions and experiences with crimps seem to be mixed.
My main concerns are not opening the system if at all possible, not creating a dangerous situation by brazing on a charged system (even though it would be a dead leg and the valve would be opened or removed to relieve any pressure build up), and not creating microfractures/another leak on the system side of the crimp. The copper is about 60 years old and has been in the relatively "hot" compressor cabinet which I would assume would make it brittle over time and I'm concerned about the crimp making "cracks".
Is recovering and recharging the system the only way, or can the valve be eliminated reliably without doing so? It's obviously a kludge approach, but can the service port cap be sealed somehow? I guess if the valve is leaking it can be expected for the stem or diaphragm to eventually do so as well...
|
|
marko
7 Cubic Foot
Posts: 144
|
Post by marko on Apr 27, 2023 22:02:20 GMT
Yes, the service port tube can be pinched and soldered, but may I suggest making sure that the Schrader valve core is snug and installing a 1/4" flare cap with a touch of "Leak Loc"?
Edit: So I may not have completely understood the type of valve you have. If it is a "king valve", with a square topped stem, most have a packing gland that can be tightened or more packing installed, if needed.
My flare cap suggestion still stands.
|
|
|
Post by majestek12 on Apr 27, 2023 22:33:50 GMT
Yes, the service port tube can be pinched and soldered, but may I suggest making sure that the Schrader valve core is snug and installing a 1/4" flare cap with a touch of "Leak Loc"? Edit: So I may not have completely understood the type of valve you have. If it is a "king valve", with a square topped stem, most have a packing gland that can be tightened or more packing installed, if needed. My flare cap suggestion still stands. Thanks Marko, I wasn't aware of Leak Loc and will definitely give it a shot before taking on of more invasive approaches. The valve is a "Kerotest" name brand, old school, packless, valve from the 1960s. The service port itself is just an open port, no Schrader or provisioning for one. The port has a cap on it, but no sealant like you educated me to. Pictures can be seen here redirect.viglink.com/?key=71fe2139a887ad501313cd8cce3053c5&subId=5399178&u=https%3A//drive.google.com/drive/folders/1qA67cBlmNbbwqw3tMoA_JF8FA56F-guY
|
|
marko
7 Cubic Foot
Posts: 144
|
Post by marko on Apr 27, 2023 23:24:32 GMT
Yes, the service port tube can be pinched and soldered, but may I suggest making sure that the Schrader valve core is snug and installing a 1/4" flare cap with a touch of "Leak Loc"? Edit: So I may not have completely understood the type of valve you have. If it is a "king valve", with a square topped stem, most have a packing gland that can be tightened or more packing installed, if needed. My flare cap suggestion still stands. Thanks Marko, I wasn't aware of Leak Loc and will definitely give it a shot before taking on of more invasive approaches. The valve is a "Kerotest" name brand, old school, packless, valve from the 1960s. The service port itself is just an open port, no Schrader or provisioning for one. The port has a cap on it, but no sealant like you educated me to. Pictures can be seen here redirect.viglink.com/?key=71fe2139a887ad501313cd8cce3053c5&subId=5399178&u=https%3A//drive.google.com/drive/folders/1qA67cBlmNbbwqw3tMoA_JF8FA56F-guYAhhh, I see your packless valve. That should never leak (from the "packless" part), maybe it just needed a slight bit more tightening against the seat. Yes, a dab of "Leak-Loc" on the flare angle will insure a good seal on the cap.
|
|
|
Post by majestek12 on Apr 28, 2023 0:35:07 GMT
Ahhh, I see your packless valve. That should never leak (from the "packless" part), maybe it just needed a slight bit more tightening against the seat. Yes, a dab of "Leak-Loc" on the flare angle will insure a good seal on the cap. I've got a bottle of Nylog Red on order, seems similar to Leak Loc but specific to older CFCs and mineral oil systems instead of POE. I don't know that it makes a difference on a leak and it's probably most important when the sealant could get INTO a system such as when under vacuum, but I figured it may better stand up to similar chemicals. I've got the valve closed pretty tight, I think it's just the seat giving up after 60+ years. I'm intrigued by your statement on putting the Leak Loc/Nylog on the flare angle. I had planned on giving the cap threads a decent coating, are you implying it would be better put on the flare surface on the cap and valve? Perhaps both the flare surfaces and the threads?
|
|
|
Post by turbokinetic on Apr 28, 2023 14:08:04 GMT
Ahhh, I see your packless valve. That should never leak (from the "packless" part), maybe it just needed a slight bit more tightening against the seat. Yes, a dab of "Leak-Loc" on the flare angle will insure a good seal on the cap. I've got a bottle of Nylog Red on order, seems similar to Leak Loc but specific to older CFCs and mineral oil systems instead of POE. I don't know that it makes a difference on a leak and it's probably most important when the sealant could get INTO a system such as when under vacuum, but I figured it may better stand up to similar chemicals. I've got the valve closed pretty tight, I think it's just the seat giving up after 60+ years. I'm intrigued by your statement on putting the Leak Loc/Nylog on the flare angle. I had planned on giving the cap threads a decent coating, are you implying it would be better put on the flare surface on the cap and valve? Perhaps both the flare surfaces and the threads? Needs to just be on the flare seat area, not the threads. Another option you can use are FlareSeal gaskets, or even just regular service port caps with rubber o-rings in them. FlareSeal can be bought at almost any HVAC shop. https://flareseal.com/ These caps can also be found almost anywhere: https://www.jbind.com/product-category/access-valves/quick-seal-o-ring-caps/ Those are just two options which have zero risk of system contamination. I use the FlareSeal gaskets extensively on the old belt drive systems which are almost always assembled with flare fittings. They are nearly 100% reliable even on 90 year old fittings. Sincerely, David
|
|