- Joined
- Jul 8, 2009
- Messages
- 773
- Reaction score
- 248
+1 on the surreal comment.
1/3 Scale Ford 289 Hi-Po
- Thread starter mayhugh1
- Start date
Help Support Home Model Engine Machinist Forum:
With all engine parts now finished except for the piston rings, it would make sense to start on them next. Piston rings are really boring, and with so many ahead of me it's hard to get moving in that direction. After lots of procrastination, work is continuing instead with the more interesting running accessories like the radiator, ignition module, and fuel tank. They'll eventually be needed to run the engine, and finishing them before the rings will remove any temptation to jury rig a first-start that I might regret.
The radiator was tackled first. The holy grail for a model V-8 is getting one to idle indefinitely without overheating. Subjective results from my last two engines suggest that more cooling capacity may be available from a simple hollow core compared with one having the same outside dimensions but made up of finned cooling tubes. This may be nonintuitive, but there are some reasons that support it.
Jerry Howell's V-4 radiator is a well thought-out example of a finned cooling tube radiator, but if it instead had a plain hollow core it would hold nearly 3X more coolant. Cooling fins add realism, but in most models without an electric fan forcing air through them, the fins are more cosmetic than one might think. Cooling tube construction is typically done with brass because of its solder-ability, but aluminum has 4X better thermal conductivity.
Construction began with the radiator core and was greatly simplified by starting with an extruded rectangular aluminum workpiece. The piece of 6063 architectural material that I had on hand has slightly better thermal properties than 6061, but it's soft and more difficult to machine. An array of grooves milled on both sides of the core adds the look of fins and even increases the core's surface area for better heat transfer.
A full-size 289 with a three core radiator has a coolant capacity of 14.7 quarts with of 25% of the coolant inside the radiator at any time. If scaled by volume and assuming all else equal (it probably isn't) the goal for the 1/3 scale model's total coolant capacity would be .54 quarts. The engine's actual coolant capacity works out to be only .15 quart, but a 1/3 scale radiator with a hollow core will add another .85 quart giving a one quart system total.
The upper and lower tanks were machined from 6061 billet with features similar to those on the drawn tanks of a typical 60's era radiator. They'll eventually be JB Welded to the core, but in the meantime they were bead blasted to provide bite to the epoxy and paint. - Terry
The radiator was tackled first. The holy grail for a model V-8 is getting one to idle indefinitely without overheating. Subjective results from my last two engines suggest that more cooling capacity may be available from a simple hollow core compared with one having the same outside dimensions but made up of finned cooling tubes. This may be nonintuitive, but there are some reasons that support it.
Jerry Howell's V-4 radiator is a well thought-out example of a finned cooling tube radiator, but if it instead had a plain hollow core it would hold nearly 3X more coolant. Cooling fins add realism, but in most models without an electric fan forcing air through them, the fins are more cosmetic than one might think. Cooling tube construction is typically done with brass because of its solder-ability, but aluminum has 4X better thermal conductivity.
Construction began with the radiator core and was greatly simplified by starting with an extruded rectangular aluminum workpiece. The piece of 6063 architectural material that I had on hand has slightly better thermal properties than 6061, but it's soft and more difficult to machine. An array of grooves milled on both sides of the core adds the look of fins and even increases the core's surface area for better heat transfer.
A full-size 289 with a three core radiator has a coolant capacity of 14.7 quarts with of 25% of the coolant inside the radiator at any time. If scaled by volume and assuming all else equal (it probably isn't) the goal for the 1/3 scale model's total coolant capacity would be .54 quarts. The engine's actual coolant capacity works out to be only .15 quart, but a 1/3 scale radiator with a hollow core will add another .85 quart giving a one quart system total.
The upper and lower tanks were machined from 6061 billet with features similar to those on the drawn tanks of a typical 60's era radiator. They'll eventually be JB Welded to the core, but in the meantime they were bead blasted to provide bite to the epoxy and paint. - Terry
- Joined
- Jun 24, 2010
- Messages
- 2,445
- Reaction score
- 968
Very nice! Just out of curiosity, have you ever tried some of the low temp aluminum braze alloys? I was eyeing Harris AluxCore 98/2 which melts at abut 65% that of aluminum. Its 98% zinc so maybe its like the Miracle Magic rods shown when the circus comes to town LOL. My failed attempts at the higher strength braze materials have been a disaster, the part inevitably resembles a puddle because the melting temp is ~90% of parent material.
Attachments
peterl95124
Well-Known Member
- Joined
- Feb 22, 2020
- Messages
- 524
- Reaction score
- 343
funny you should say that, I just ended my Duesenberg ring procrastination...
am experimenting with yet another method of cold formed rings (my heat treat furnace element burned out and haven't found a replacement yet). my previous attempt at cold formed rings resulted in them needing to be lapped, and that ended up taking a full .005 away at the gap before they passed the light test in a ring-gauge, so this time I'm starting with rings that are intentionally eccentric by .005 so that they will be uniform by the time they are lapped in. since the previous rings are fully functional in spite of not being uniform width this is mostly cosmetic but my OCD is taking over and I have to try it. Time will tell...
FYI, my way of cold formed rings is to machine them to the size of the gap-enlarged Trimble Method rings, take the diameter of his gapping dowel and add that to the circumference of the ring and machine to that size, then cut out a gap the width equal to the gapping dowel diameter. These rings when inserted into a cylinder or ring-gauge touch the cylinder at the gap and opposite the gap and pass light everywhere else, so lots of force at the edges of the gap so be careful to not scratch your ring-gauge. I try to apply lapping compound at the gap and not opposite the gap. Time will tell if this is the right approach...
Peter,
I've never tried that particular brand, but I have played with others I was tempted to buy at shows. It was never as easy to use at home as it appeared in the demos with the right amount of heat coming from practice, and the window is pretty narrow. With the mixed results I've personally had, I've never used them on anything but practice parts. - Terry
I've never tried that particular brand, but I have played with others I was tempted to buy at shows. It was never as easy to use at home as it appeared in the demos with the right amount of heat coming from practice, and the window is pretty narrow. With the mixed results I've personally had, I've never used them on anything but practice parts. - Terry
A primarily cosmetic shroud was designed for the radiator and machined from black Delrin. It's mounted to a pair of side rails attached to the sides of the radiator core. Machining Delrin is a hoot and an opportunity to run with the big boys with speeds and feeds at near full tilt. Both faces of the shroud's workpiece received a lot of machining.
The shroud will be mounted to the radiator through four spacers that will allow the fan to draw air across the rear face of the radiator. The spacers will also provide for the removal of the shroud with the radiator in place in front of the engine.
The core's thin wall was a problem for the fasteners that attached the aluminum side rails. The fasteners penetrate the walls of the core but provide only three or so threads for a Loctite seal. Loctite 620 was added to the primer'd surfaces under the rails for additional sealing and the elimination of air spaces that would create bubbling problems when the assembly's paint is heat cured. - Terry
The shroud will be mounted to the radiator through four spacers that will allow the fan to draw air across the rear face of the radiator. The spacers will also provide for the removal of the shroud with the radiator in place in front of the engine.
The core's thin wall was a problem for the fasteners that attached the aluminum side rails. The fasteners penetrate the walls of the core but provide only three or so threads for a Loctite seal. Loctite 620 was added to the primer'd surfaces under the rails for additional sealing and the elimination of air spaces that would create bubbling problems when the assembly's paint is heat cured. - Terry
That is awesome!
Terry
What brand of heat cured paint are you using?
Awesome build, thank you for sharing.
The radiator parts were bead blasted and Gun-Kote'd satin black. The finished radiator was too big to fit inside my shop's heat treat oven, and so the core and the tanks were painted and heat cured separately prior to final assembly. In the past I used our kitchen oven to bake large painted parts, but fumes from curing the 289's engine block created some domestic problems. After that, shop work was limited to the shop.
Although 99% of the radiator's surfaces were painted and cured in the shop oven before final assembly, the seams between the core and the tanks were masked off for the JB Weld used to bond them together. The gaps were about 1/8" wide, and it was important for air not to become trapped in the epoxy and bubble up later during the final heat cycle. JB Weld, thinned with acetone, was injected into the gaps with a hypodermic and allowed to cure with a smooth filleted surface. After curing, the assembled radiator needed two final (low fume) 300F cycles - one to insure total outgassing of the epoxy and a second to cure the subsequent painted-over seams. These were done in a secret late night kitchen operation.
The filler neck had to be machined as a separate item and was Loctite'd in place during final assembly. The cap itself was machined from 303 stainless. Threaded hose barbs were also turned from 303 stainless, but since I'd evidently used the wrong tap drill in the tanks their fits wound up sloppy. Rather than fill the gaps with sealer or Teflon tape, the barbs were permanently epoxied into the tanks.
The radiator will eventually be supported on a display base with a pair of mounting brackets attached to the sides of the lower tank. Eighth inch fabricated steel brackets will be trimmed to final size once the radiator's mounting height is finalized. A coolant drain added to the bottom of the lower tank with a 1/4" NPT pipe plug wrapped up work on the radiator. - Terry
Although 99% of the radiator's surfaces were painted and cured in the shop oven before final assembly, the seams between the core and the tanks were masked off for the JB Weld used to bond them together. The gaps were about 1/8" wide, and it was important for air not to become trapped in the epoxy and bubble up later during the final heat cycle. JB Weld, thinned with acetone, was injected into the gaps with a hypodermic and allowed to cure with a smooth filleted surface. After curing, the assembled radiator needed two final (low fume) 300F cycles - one to insure total outgassing of the epoxy and a second to cure the subsequent painted-over seams. These were done in a secret late night kitchen operation.
The filler neck had to be machined as a separate item and was Loctite'd in place during final assembly. The cap itself was machined from 303 stainless. Threaded hose barbs were also turned from 303 stainless, but since I'd evidently used the wrong tap drill in the tanks their fits wound up sloppy. Rather than fill the gaps with sealer or Teflon tape, the barbs were permanently epoxied into the tanks.
The radiator will eventually be supported on a display base with a pair of mounting brackets attached to the sides of the lower tank. Eighth inch fabricated steel brackets will be trimmed to final size once the radiator's mounting height is finalized. A coolant drain added to the bottom of the lower tank with a 1/4" NPT pipe plug wrapped up work on the radiator. - Terry
- Joined
- Jul 16, 2007
- Messages
- 3,036
- Reaction score
- 1,124
Terry, a unique idea for the radiator I just wonder what the efficiency will be. I'm sure 10 minute runs wouldn't be a problem. At the time I made mine heater core stock was still available at the local radiator shop but that was in the eighties.
The ignition module is designed around one of Roy Sholl's CDI's. Roy sold both regular and 'magnum' versions of his CDI's with the higher sparks/sec magnums intended for multi-cylinder engines. I believe the differences between the two were that the magnums which were intended to be driven at a full six volts and had smaller storage capacitors and beefier drivers for faster charging rates.
A schematic is shown in the first photo. A relatively simple front-end board of my own design includes an led triggered by the Hall device without any need to power the CDI. This feature simplifies troubleshooting and provides a visual indicator to aid timing adjustments without the need to deal with the high voltage. Inadvertently triggering a live CDI without a suitable return path for the spark can create latent damage inside the coil that over time can cause the unit to fail.
The ignition module is powered up upon connection to 6 volts dc. A blue led on the unit lights up whenever the distributor's Hall device is activated by the distributor's magnet putting the ignition into 'dwell.' When the distributor is rotated and the Hall device is deactivated, the light turns off. The CDI, if powered up, then immediately fires. The CDI is powered up by a toggle switch on the enclosure along with a red warning light.
The ignition electronics will be packaged inside a machined aluminum enclosure reminiscent of the MSD 6A aftermarket CDI's. MSD ignitions actually didn't appear on the market until several years after the 289 HiPo. Many owners including myself were dazzled by their slick marketing, and lots of perfectly working stock ignitions were replaced with these higher voltage units. On daily driven cars they sometimes created more problems than they solved, but their smart-looking red enclosures and the impressively long sparks produced by their displays sitting on auto parts counters were hard to ignore. I was surprised to discover that the 6A unit which is the one I added to my '65 Mustang decades ago is still available from the manufacturer.
The next steps are to create the front-end circuit board and cram all the electronics into the insufficient space I've left for myself. I tried to keep the dimensions of the enclosure to a reasonable scale, but the enclosure's height which was driven by Roy's CDI made it thicker than I wanted. After bead blasting it was Gun Kote'd 'blood red'. - Terry
A schematic is shown in the first photo. A relatively simple front-end board of my own design includes an led triggered by the Hall device without any need to power the CDI. This feature simplifies troubleshooting and provides a visual indicator to aid timing adjustments without the need to deal with the high voltage. Inadvertently triggering a live CDI without a suitable return path for the spark can create latent damage inside the coil that over time can cause the unit to fail.
The ignition module is powered up upon connection to 6 volts dc. A blue led on the unit lights up whenever the distributor's Hall device is activated by the distributor's magnet putting the ignition into 'dwell.' When the distributor is rotated and the Hall device is deactivated, the light turns off. The CDI, if powered up, then immediately fires. The CDI is powered up by a toggle switch on the enclosure along with a red warning light.
The ignition electronics will be packaged inside a machined aluminum enclosure reminiscent of the MSD 6A aftermarket CDI's. MSD ignitions actually didn't appear on the market until several years after the 289 HiPo. Many owners including myself were dazzled by their slick marketing, and lots of perfectly working stock ignitions were replaced with these higher voltage units. On daily driven cars they sometimes created more problems than they solved, but their smart-looking red enclosures and the impressively long sparks produced by their displays sitting on auto parts counters were hard to ignore. I was surprised to discover that the 6A unit which is the one I added to my '65 Mustang decades ago is still available from the manufacturer.
The next steps are to create the front-end circuit board and cram all the electronics into the insufficient space I've left for myself. I tried to keep the dimensions of the enclosure to a reasonable scale, but the enclosure's height which was driven by Roy's CDI made it thicker than I wanted. After bead blasting it was Gun Kote'd 'blood red'. - Terry
Last edited:
Your work is amazing and inspiring
What would the cycle time be for machining the ignition cover
What would the cycle time be for machining the ignition cover
Time to come up with the design was about 20 hours. Total machining time was about 6 hours. - Terry
- Joined
- Jun 24, 2010
- Messages
- 2,445
- Reaction score
- 968
So presumably you selected the magnum model? Is it from prior inventory you sourced from the new supplier/builder? Any thoughts on how tolerant the nominal 6v input voltage would be to say... a 2S lithium battery ~7.4v. I'm sure it would handle the current requirements, but that's a moot point if the whole system revolves around 6v +/- some narrow deviation.
https://www.cncengines.com/
It's one of Roy's magnums. I'll be using a voltage regulator module to drop the 12 volt battery voltage down to exactly 6V. Not sure how the lifetime of Roy's CDI's are affected by voltages higher than 6V, but he was always adamant about not going any higher. - Terry
rsholl
Active Member
Terry, you are correct on the slightly smaller value capacitor so it can charge faster. The Magnum was not designed to run strictly on 6 volts. Depending on the RPM's you need, it will operate very well on any voltage from 4.5 to 6. The reason I stressed 6 volts as an upper limit is we also drive the front end of the CDI harder. The more voltage in the more current draw. Voltage X RPM's = total current draw. The Magnum will operate up to about 20K on 5 volts, for more speed it needs the 6 volts to get to 30K RPM's
All my systems are that way, the least amount of voltage and speed you run the engine at the more efficient the CDI is. Dan is building the CDI's to the same specs as I did so nothing will change. I would not try 7.4 volts because the switching transistor in the DC-DC circuit would not handle the current long term. I would like to see you set your regulator at maybe 5.6 volts to see if your engine shows any signs of missed sparks at the top RPM you will shoot for. If so then go ahead and up the voltage to 6 but I'm betting you aren't planning on running that very fine looking engine at more than 7,500 RPM. Excellent build and you have set the bar very high for us builders, me included.
Roy
All my systems are that way, the least amount of voltage and speed you run the engine at the more efficient the CDI is. Dan is building the CDI's to the same specs as I did so nothing will change. I would not try 7.4 volts because the switching transistor in the DC-DC circuit would not handle the current long term. I would like to see you set your regulator at maybe 5.6 volts to see if your engine shows any signs of missed sparks at the top RPM you will shoot for. If so then go ahead and up the voltage to 6 but I'm betting you aren't planning on running that very fine looking engine at more than 7,500 RPM. Excellent build and you have set the bar very high for us builders, me included.
Roy
- Joined
- Jun 24, 2010
- Messages
- 2,445
- Reaction score
- 968
Can you elaborate - what are the dimensions of current draw in this formula?
And are these reference RPMs on a single cylinder basis, so a 5 cylinder would be 30K/5=6K max crankshaft RPM for example?
Peter,
Roy can step in and correct me, but I think there's a little confusion created by one of the units in his reply. An 8 cylinder 4-cycle engine running at 5000 rpm needs 5000 rev/min x 4 sparks/rev = 20,000 sparks/min which is what Roy says his unit will deliver when driven at 5 volts. What may have confused you is when he said "it needs 6 volts to get to 30,000 RPM". He really meant to say 30,000 sparks/min which would be equivalent to 7500 RPM from an 8 cylinder engine.
As far as "voltage x RPM = current" goes, I think he was just saying that the current draw will increase as either the applied voltage or the RPM is increased and not necessarily giving an equation with formal units which probably isn't strictly linear with respect to both variables. - Terry
Last edited:
