Water Pump Info circa 2019
Printed From: TheAMCForum.com
Category: The Garage
Forum Name: AMC V8 Engine Repair and Modifications
Forum Description: AMC-made V8 engine mechanical, ignition and fuel from basic repair to high-perf modifications
URL: https://theamcforum.com/forum/forum_posts.asp?TID=99988
Printed Date: Apr/19/2024 at 5:24pm
Software Version: Web Wiz Forums 12.03 - http://www.webwizforums.com
Topic: Water Pump Info circa 2019
Posted By: Greyhounds_AMX
Subject: Water Pump Info circa 2019
Date Posted: May/13/2019 at 12:31pm
|
The situation changes with time, but here's the current state of affairs as I am seeing it. Excluding the early V8 variants, there are two basic water pumps that folks need the most: 1) 68-72 version: 4-1/2" hub height, bypass
connection at 11:00 position 2) 73-up version: 4-7/8" hub height, bypass
connection at 12:00 position At this point in time it's become very hard to find an off
the shelf 68-72 water pump. The only one listed at RockAuto for example is the
GMB 110-2908 (cast iron), which has the wrong shaft length, the bypass
connection is in the late model location, and it appears to hit the harmonic
balancer as well. If you want a correct 68-72 water pump you'll probably need
to send one out to have it rebuilt, or buy a rebuilt unit from one of the AMC
vendors who usually keep them on hand. They do have a core charge however. Aluminum 73-up pumps are available from multiple supplies,
and all appear to use the same basic GMB casting, except the Edelbrock pumps.
The casting quality is a huge step up from the cast iron pump bodies, with
larger passages and more concentric connections. Here's a rundown of those
aluminum pumps: GMB 110-1040AL ($28) - Standard replacement. 12 month /
12,000 mile warranty. GMB 110-1040P ($45) - High performance, better bearings,
better seal, better impeller. 12 month / 12,000 mile warranty. Airtex AW3401H ($52) - High performance, better bearings,
better seal, better impeller. Lifetime warranty. Flowkooler 1781 ($93) - CNC billet 16 vane impeller with
anodized coating, reduced impeller clearances. 4-13/16" hub height. 2 year
warranty.
Milodon 16271 ($104) - High volume. 4-13/16" hub
height. 180 day warranty. Here's some comparison pics of a typical cast iron pump
casting with the GMB aluminum casting, courtesy of one of the fsj sites. Based on this and the variety of
available offerings I can't see any reason to use a cast iron water pump.
Edelbrock offers water pumps in both common hub heights, and
therefore has the only 68-72 aluminum pump available. The same casting is used
for both, so the only difference is in the hub height. Edelbrock 8832 ($281) - 73-up long pump ------------- 1968 AMX 390 w/T5 |
Replies:
Posted By: 304-dude
Date Posted: May/13/2019 at 12:35pm
|
There is a new update on water pumps... Galvins and others now carry proper pre 73 pumps. I will look up the link and add it... this is from Galvins site. http://www.ramblerparts.com/product_info.php?products_id=665" rel="nofollow - http://www.ramblerparts.com/product_info.php?products_id=665 ------------- 71 Javelin SST body 390 69 crank, 70 block & heads NASCAR SB2 rods & pistons 78 Jeep TH400 w/ 2.76 Low 50/50 Ford-AMC Suspension 79 F150 rear & 8.8 axles Ford Racing 3.25 gears & 9" /w Detroit locker |
Posted By: 304-dude
Date Posted: May/13/2019 at 12:53pm
|
Here is Galvin's new water pump http://theamcforum.com/forum/6872-v8-correct-new-water-pump-photo_topic98457.html" rel="nofollow - http://theamcforum.com/forum/6872-v8-correct-new-water-pump-photo_topic98457.html As a side note... through the years now, a few have updated their water pump and pump pulley to retro fit newer pumps on late 68 to 72 cars. Also the two styles have different diameter bypass nipples, outside of the clocking. A second note is some of the performance brand named pumps are relabeled GMB pumps. I noticed with GMB pumps, they give two different height spacers for proper fitment with accessories. ------------- 71 Javelin SST body 390 69 crank, 70 block & heads NASCAR SB2 rods & pistons 78 Jeep TH400 w/ 2.76 Low 50/50 Ford-AMC Suspension 79 F150 rear & 8.8 axles Ford Racing 3.25 gears & 9" /w Detroit locker |
Posted By: ChillyB
Date Posted: May/13/2019 at 12:55pm
|
I have two engines: a 77 and a 73. I noticed one of the bokt boss heights were different so I had to move over a water pump when I swapped engines (dropping in a running engine to get by for a while). Just got a new one from NAPA. They no longer discriminate, cutting the bolt boss low then including spacers to raise it back up if thats the accessory bracket type you need. They also dont take a core return.
|
Posted By: Sonic Silver
Date Posted: May/13/2019 at 5:41pm
|
Greyhounds_AMX wrote:Posted By: 69 ambassador 390
Date Posted: May/13/2019 at 10:18pm
------------- Steve Brown Algonac, Mi. 69 Ambassador sst 390 84 Grand Wagoneer 69 Cougar XR7 65 Fairlaine 500XL 79 F-350 Super Camper Special |
Posted By: Greyhounds_AMX
Date Posted: May/13/2019 at 10:33pm
|
What type of failure, and which GMB aluminum pump? ------------- 1968 AMX 390 w/T5 |
Posted By: 69 ambassador 390
Date Posted: May/13/2019 at 11:09pm
|
The casting fails on the nose where the bearing is. It only happens when running high rpms and hard acceleration. A customer lost two radiators. The GMB is not as heavy in that area as the Edelbrocks are. I would limit the GMB castings to stockish applications and definitely not on clutch fan installs as that puts a lot of weight on the end of the shaft. Same customer put an iron pump on and no more issues. The GM just doesn't have enough meat or reinforcing ribs in that area. ------------- Steve Brown Algonac, Mi. 69 Ambassador sst 390 84 Grand Wagoneer 69 Cougar XR7 65 Fairlaine 500XL 79 F-350 Super Camper Special |
Posted By: 304-dude
Date Posted: May/14/2019 at 8:38am
Good info... I expected a limitation with aluminum pumps, being that they are cast, and nothing done to beef up the snout. IMO the Edlebrock would be the best option for crank driven perposes. All the other aluminum variants seem to use the GMB casting. Though, an electric pump seems to be one of the choices many strip racers opt for. It is nice to see, as of recent, new iron pumps are available for the two variants. Will end up using my GMB aluminum one for emergency spare perposes. As it was a cheap option, that will work in finalizing my modified accessories installation, using the 73 on up pump. ------------- 71 Javelin SST body 390 69 crank, 70 block & heads NASCAR SB2 rods & pistons 78 Jeep TH400 w/ 2.76 Low 50/50 Ford-AMC Suspension 79 F150 rear & 8.8 axles Ford Racing 3.25 gears & 9" /w Detroit locker |
Posted By: mmaher94087
Date Posted: May/14/2019 at 11:09am
|
The boys a MOPAR will tell of shearing the front off the pump with high horse power motors. They recommended removing every other vane off the water pump impeller. If one thinks about what's happening; it makes sense. Engine revs quickly and the impeller in the pump tries to move water quickly.. water cannot be compressed. Something breaks. ------------- Mike |
Posted By: 304-dude
Date Posted: May/14/2019 at 11:31am
A bit of common sense... though, it will kill city driving on a hot day. Though, my feeling is, if your engine is to run above 6000 RPM 90% of the time... then you would be better off with underdriven crank and waterpump pulleys. Problem is they will need to be a home brew custom setup, since nobody makes them for AMC. Though if Edlebrock has both AMC and GM patern on its water pump, and your using an ATI or early harmonic damper with Chevy pulley bolt patern, you could find under drive pulleys to work. Also I would think having power steering shared on the same belt would add side load on the pump to the torque being generated at the RPM extremes. NASCAR and many other motor sports will offset sideloading on the pump bearing by using a separate belt to the power steering pump. Though another benefit is making belt jump minimal with using smaller belts. There is a lot of subtle things to ponder that when added all together can make a big difference. ------------- 71 Javelin SST body 390 69 crank, 70 block & heads NASCAR SB2 rods & pistons 78 Jeep TH400 w/ 2.76 Low 50/50 Ford-AMC Suspension 79 F150 rear & 8.8 axles Ford Racing 3.25 gears & 9" /w Detroit locker |
Posted By: Greyhounds_AMX
Date Posted: May/14/2019 at 12:35pm
|
I suppose you could consider the early V8 pulley setups underdrive compared to the later 4-bolt crank pulley stuff. ------------- 1968 AMX 390 w/T5 |
Posted By: Steve_P
Date Posted: May/14/2019 at 4:45pm
| Modern cars all use aluminum pumps and many turn 7k+ RPM. If the AMC aluminum pumps are failing the problem isn't RPMs and aluminum, it's the design of the pump |
Posted By: 304-dude
Date Posted: May/14/2019 at 4:55pm
I dont think anyone said it was the use of aluminum that is faulty. Looking at the overall dimensions, nothing indecates the castings were enlarged enough at the snout end to retain the bearing from being pulled and breaking through. The lack of proper reinforcement on GMB castings, is never going to be as strong as iron. You would need to double the thickness at around the parameter to help with the lack of material from 90° cut for bearing and seal retension. Haven't taken any of these aluminum pumps apart to verify how well chamfered, and thickness at the bearing seat, but assume they did not make it a critical part of manufacturing design. ------------- 71 Javelin SST body 390 69 crank, 70 block & heads NASCAR SB2 rods & pistons 78 Jeep TH400 w/ 2.76 Low 50/50 Ford-AMC Suspension 79 F150 rear & 8.8 axles Ford Racing 3.25 gears & 9" /w Detroit locker |
Posted By: ChillyB
Date Posted: May/14/2019 at 7:04pm
| Agree with all of that. Replicating iron parts with aluminum castings skips any design effort and goes ahead with a lot of assumptions. Titanium gun parts are my favorite example of this. A well designed steel part is made out of titanium instead and advertized as an upgrade. For dimensionally equal parts steel is stronger than Ti. Ti advantage is only realized when size of part is increased such that for a comparable weight the larger Ti part is stronger, often with addition of extensive gussetting. A Ti scope mount dimensionally equal to steel is weaker but much lighter. Sometimes that is strong enough. Sometimes not. |
Posted By: FSJunkie
Date Posted: May/15/2019 at 3:13am
|
GMB new iron pumps routinely only last me about 30,000 miles before the shaft bearing gets loose and begins to wobble. It gets looser until I replace it. Six and V8. ...which is why I buy them from NAPA with a lifetime warranty and keep the receipt. No, I am not over-tightening my belts. I wonder if the shaft seal slowly leaks and contaminates the lubricant in the bearing. I've started using an anti-rust and water pump seal lubricant additive in my coolant so we'll see if that helps. ------------- 1955 Packard 1966 Marlin 1972 Wagoneer 1973 Ambassador 1977 Hornet 1982 Concord D/L 1984 Eagle Limited |
Posted By: Hurst390
Date Posted: May/15/2019 at 1:10pm
 I've been running GMB aluminum pumps at high rpm for years no failures. in circle track and drag. Hundreds of laps------------- SC/Hurst Rambler 11.62 120 100% Street Legal |
Posted By: Hurst390
Date Posted: May/15/2019 at 1:16pm
 Another basic aluminum GMB pump at work that I've ran for years with zero issues..
------------- SC/Hurst Rambler 11.62 120 100% Street Legal |
Posted By: Greyhounds_AMX
Date Posted: May/15/2019 at 6:17pm
|
My NAPA cast iron lifetime warranty pump started leaking after 5000 miles. Fat lot of good the warranty does me when they don't offer an early style pump anymore. ------------- 1968 AMX 390 w/T5 |
Posted By: 304-dude
Date Posted: May/15/2019 at 6:35pm
|
I wonder what accessories where on the engine that constantly broke the pump nose? Seems like the two replies with users of the GMB pump, have tested race only motors that may not have power steering or may have a belt setup that does not overly effect torque at the nose of the pump. Since my theory with bearing break out is based on how torque is applied through the belt setup with a base copy of tye iron core using cast aluminum. ------------- 71 Javelin SST body 390 69 crank, 70 block & heads NASCAR SB2 rods & pistons 78 Jeep TH400 w/ 2.76 Low 50/50 Ford-AMC Suspension 79 F150 rear & 8.8 axles Ford Racing 3.25 gears & 9" /w Detroit locker |
Posted By: Hurst390
Date Posted: May/16/2019 at 5:29am
 ------------- SC/Hurst Rambler 11.62 120 100% Street Legal |
Posted By: 304-dude
Date Posted: May/16/2019 at 5:50am
Thanks! Seems that side loading with accessories is not the issue with bearing breakout. The only other thing would be cooling fan type. Looks like you have chosen proper mechanical fan for your application. As noted in the picture, a 4 blade, which uses less torque for high rpm use. So, in short there is no reason for nose breakage from overall design, if properly installed, and use with a proper fan for engine application. As stated by the reply that brought up breakage, a heavy clutch fan in high performance use shouldnt be used. I dont know of any strip cars using them, if any thing, electic fans are common. Well, it seems like a lot of hot air on theory with why there is pump breakage. Maybe a wee lot of bad castings, who knows? ------------- 71 Javelin SST body 390 69 crank, 70 block & heads NASCAR SB2 rods & pistons 78 Jeep TH400 w/ 2.76 Low 50/50 Ford-AMC Suspension 79 F150 rear & 8.8 axles Ford Racing 3.25 gears & 9" /w Detroit locker |
Posted By: ChillyB
Date Posted: May/16/2019 at 6:05am
|
Here is a theory based on mostly nothing. The fan will have considerable gyroscope effect that torques it sideways under acceleration and deceleration. Holding a spinning bearing isnt a tough job and the side load on the bearing due to belts and weight of fan seems insufficient to break anything. But that gyro effect can apply side loads that seem disproportionate to the static weight of the loads hanging on it. Now cantilever all of that gyro load out on the end of a shaft with an extension. Some of you guys are obviously running thise pumps hard and not breaking them. But that's the nature of design reliability. Rarely is the design so bad that it fails most of the time. But a premature failure rate as low as 5% is still pretty awful. And a lot of customers will feel OK about the product because theyre part of the 95%.
|
Posted By: farna
Date Posted: May/16/2019 at 6:30am
|
The race cars are using a light weight fan or electric fans. A street application will use a heavier fan, and in some cases a clutch fan. I suspect those using a light weight fan or electrics fans (no engine fan) don't have an issue, but running something heavy like a clutch fan causes breaking.. as already mentioned. They seem to hold up fine at more normal street rpm though (only occasional short burst to 5000 rpm or so, hardly ever higher). Not really necessary here, but I'll point out that the stock AMC pump (or the MOPAR!) wasn't designed to work at sustained high rpm. Stock pumps are designed to work well in normal driving ranges -- like idle to 3000-3500 rpm. Start getting over 5000 for any length of time and I can understand having issues!! ------------- Frank Swygert |
Posted By: 304-dude
Date Posted: May/16/2019 at 6:31am
Its mostly about the snap of sudden change on rotational forces. The spin of the fan and its mass, along with accessories with pull the nose to the driver side, and torque twist. Plus if one is to romp and stomp roller coaster rpms, they are begging to snap a nose off, when a mechanical fan is too great for the demands of the application. ------------- 71 Javelin SST body 390 69 crank, 70 block & heads NASCAR SB2 rods & pistons 78 Jeep TH400 w/ 2.76 Low 50/50 Ford-AMC Suspension 79 F150 rear & 8.8 axles Ford Racing 3.25 gears & 9" /w Detroit locker |
Posted By: SC397
Date Posted: May/16/2019 at 7:16am
|
There are at least 2 guys in the Detroit area that rebuilds original water pumps if anyone is interested. I have been taking a few of my cores with the part numbers in to this fellow. He is very reasonable, ships, and has a one year warranty. I will post up pictures etc. when I get a chance. Dan Bailey 4706 Detroit St Dearborn Hts, MI 48125 313-530-2893 |
Posted By: bigbad69
Date Posted: May/16/2019 at 8:24am
There are lots of variables. If the bearing and bearing retainer (in this case the pump casting) design is marginal, some will fail, and some won't. ------------- 69 Javelin SST BBO 390 T10 |
Posted By: Red Devil
Date Posted: May/16/2019 at 12:09pm
|
I have a Flowkooler aluminum pump (GMB with Flowkooler impeller) with a thermostatic clutch 18" fixed fan (heavy compared to a stock flex fan), PS with AGR pump, Delco alternator on my '74 Javelin 401. Sees 6500 rpm on occasion (rev limiter). Been fine for 10 years, but just a summer cruiser so not lots of miles. Different combinations will have different harmonics and different loading. If you're concerned, save up and buy an Edelbrock pump. Hope this helps, RD.
|
Posted By: Greyhounds_AMX
Date Posted: May/21/2019 at 12:32pm
|
So I ordered the GMB-1040P, which is their aluminum high-performance pump version. Again, here's their lineup: 1) 110-2908: Cast iron
stock replacement pump, long housing, late model design. 2) 110-1040AL: Aluminum housing long late model pump. 3) 110-1040P: High-performance version of 1040AL. They tout the 1040P high-performance version as having a better impeller and better bearings, and compared to the stock replacement version that appears to be true. The shaft on the 1040P is larger than stock at 0.750" diameter, and is turned down to 0.625" for the pump mounting flange, which is really nice as then the bearings are larger. Unfortunately it does prevent us from pressing the flange on further to make a long pump into a short pump. The impeller on the 1040P is a different shape than the stock replacement version, having the same number of vanes, but they are longer. The lower cost 1040AL appears to use the same impeller design, so I expect that the difference between the 1040AL and the high-performance 1040P is the larger shaft/bearings used in the 1040P. Here you can see the larger shaft diameter as it comes out of the pump housing, then back down to 0.625" for the mounting flange.
 If you consider that the GMB pump flange location is actually 3/8" further out than the stock pump shown (late model vs early pump), you realize that the GMB housing extends out further than the stock unit did, placing the bearing closer to the load. And if you compare the relative diameters of the bearing housing area to the diameter of the fan/pulley flange, it's looks like the GMB casting is a bit larger OD in the bearing area. And it has a bunch of reinforcing ribs as well.
------------- 1968 AMX 390 w/T5 |
Posted By: Greyhounds_AMX
Date Posted: May/24/2019 at 12:48pm
|
I measured and found that the GMB aluminum pump only has 3/16" of clearance between the back of the fan mounting flange and the shaft seal face. So even if the lower tier 110-1040AL version has a 0.625" shaft all the way, there's not enough room to push the flange on further and make it line up with an early crank pulley. It would need to be pressed on 5/16" further, and there's only 3/16" available. ------------- 1968 AMX 390 w/T5 |
Posted By: ChillyB
Date Posted: May/25/2019 at 6:22pm
How old is the photo of that NAPA water pump? Here's the one I just got a month ago. Looks very different drom that photo.  |
Posted By: Greyhounds_AMX
Date Posted: May/25/2019 at 7:36pm
|
Yours is a later model long pump, and my Napa pump was an early model short pump for 68-72. It had an AMC part number on it so had to be a reman, I bought it around 2001 or so. We can't see the top of yours from the pic but the later model pump will have the bypass clocked at the 12:00 position as well as the 4-3/4" or so length. ------------- 1968 AMX 390 w/T5 |
Posted By: ChillyB
Date Posted: May/26/2019 at 6:06am
| You're right. Mine is for a 73. |
Posted By: 304-dude
Date Posted: May/26/2019 at 1:05pm
|
To add to your great thread... I purchased a standard GMB pump for mockup for working on my mixed year parts engine setup. GMB 110-1040A It seems that it shares the same shaft specs as the performance GMB pump. QUESTION: IF the diameter is the same from flange to bearing for the oem late pump version... Other than the impeller difference, I assume the GMB shaft design with its corresponding bearing, will be a better option than the factory with smaller diameter shaft at the bearing? Only if we had a bearing OD to compare with between the GMB pump and late model oem pump, to see if there is a slightly larger bearing bore OD at the GMB snout, for the larger GMB bearing ID... "If" the above question is validated by a confirmation on bearing ID size differneces between OEM and GMB. ------------- 71 Javelin SST body 390 69 crank, 70 block & heads NASCAR SB2 rods & pistons 78 Jeep TH400 w/ 2.76 Low 50/50 Ford-AMC Suspension 79 F150 rear & 8.8 axles Ford Racing 3.25 gears & 9" /w Detroit locker |
Posted By: Repsolroger
Date Posted: Jan/01/2022 at 6:43pm
Edelbrock - Water Pump AMC Short Style 1968-72 (Part No. 8831) will this fit a 1969 390 .. some sites say yes...then others dont...and some ask if my engine is a W or X ... that stuff is expensive expensive ...so want to make sure before i add to cart and send it to someone------------- 1969 sst javelin restomod |