By making a “ladder diagram” I was able to use ladder logic to design the function of the ECU. I charted the diagram out on legal pads and this turned out to be the “road map” I would use to complete this phase of the project.

       The ECU design that I had come up with required some time delay relays, an RPM sensor and a new solid state ignition coil driver. Most of these had to be built so my work was cut out for me. Before I started the ECU I wanted to install the gauges so I could work “down the ladder” so to speak. I took some cardstock cutouts of my gauges and laid out the holes to be cut in the control panel by taping them to the panel. I then used hole saws and a mechanical chassis punch to cut the gauge holes out in the aluminum control panel.

       I had earlier modified an automotive tachometer so the faceplate would read out the power turbine’s RPM. It’s amazing what you can do with a scanner and a printer!!! The tach will be coupled to a photo optic RPM sensor I will mount on the output power shaft.

       I mounted all of the gauges in the control panel. The indicator lights, switches, fuse holders and key switch soon followed. Even though the gauges don’t match, it still looks good. My new design had eliminated the need for multiple switches so only a couple were needed. Most of the control would be performed by the new ECU system and then relayed back to the indicator lights. The new panel includes a tachometer, volt meter, oil temperature gauge, pyrometer, combustor pressure gauge, oil pressure gauge, fuel pressure gauge, high/low oil pressure indicator, high/low RPM indicator, on/off fuel flow indicator, high/low combustor pressure indicator, generator switch, panel lighting/headlight switch, keyed ignition switch, run/start/cool down indicator lights and afterburner switches (future use).

       I connected up the gauge power and switches to a terminal strip that will later connect to the ECU via a set of control conductors. For the most part, the control panel was complete and ready for operation. I now needed to put a new material list together for my ECU circuitry. A laundry list of parts was needed that just about cleaned out the local Radio Shacks. After driving all over town I got the basic parts list filled and started working. The first task on my list was to build a solid state ignition coil driver. I figured that the GR-5A should graduate to a modern ignition system so I used a diagram that I found on on the internet to build one.

       An output relay was added to each timer allowing for normally open and normally closed contacts on all three timers. I may post my circuit schematic for the timer board in a future post so check back later. I now was going to tackle the main relay board for the ECU. I started out by mounting all of the needed relays to the main ECU board. I then added the terminal strips to the boarders of the circuit board. The terminal strips will help connect the ECU to the different sensors and devices.

       Following my ladder diagram, I wired up the relay board. I also installed indicator LED’s on each relay so I could know what function is operating on the ECU board when it is powered up. Even though there are no IC semiconductors on the main board, it ended up being a fairly complex circuit. Well, complex to me anyway :0) It will be a lot of work to publish the finished schematics which I might get to one day. I mounted the timer board (TIM1) and the ignition circuit (IGN1) to the main ECU board to prepare for testing. By using some jumper wires, I was able to test the primary function of the ECU and it seemed to be a success. Can you tell that I enjoy using my label maker?

       There was one more support circuit I needed to complete the ECU and that was the RPM sensor switch. After looking on the internet for a schematic to build an RPM sensing switch, I remembered that I saw a similar device made for automotive use. Having some store credit at Summit Racing prompted me to buy a RPM switch made under their name. The switch was designed to turn on a device after a set engine RPM is sensed. I planned to use the switch to signal the ECU when the turbine is spooling fast enough to ignite the combustor during startup. Coupled with the photo optic sensor already installed on the inducer ring, the sensor would complete the ECU.

       The basic function of the ECU is to control startup, running, and cool down phases of operation. In starting mode, the ECU first looks for oil pressure. Once oil pressure is achieved, the ECU will allow the spooling blower to run at 100% speed. The ECU now looks for the RPM sensor to report the minimum RPM needed to inject the propane gas into the combustor. After a 3 second delay (T1), the gas flows into the combustor. The ECU now checks to see if the throttle switch (TS1)  indicates a proper fuel flow and if so, starts the ignition coil after a 1 second delay (T2). Now the ECU waits for the combustor pressure to reach operating pressure. Once achieved, the ECU shuts down the blower and goes into running mode.
       In running mode, the ECU monitors oil pressure and combustor pressure continually. EGT (exhaust gas temp.) is not monitored directly by the ECU on the GR-5A. If the turbine (GR-5A) starts to overheat, the oil becomes thinner causing the oil pressure to drop. The low oil pressure will signal the ECU to shut down the fuel so a pyrometer switch is not necessary on this model. The combustor pressure switch also tells a lot about the engine’s operating state. If a turbine failure occurs, combustor pressure will cease. If the compressor hose bursts, combustor pressure will cease. If the engine’s RPM is too low, the combustor pressure will fall. All of these states will signal the ECU to shut down the fuel on the GR-5A.
       In shutdown mode or “cool down mode” the ECU starts the spooling blower at 30% operational speed. Cool air is forced through the engine which cools the turbo down enough so that the oil cannot be scorched inside the turbo bearing. After 4 minutes, the turbine is cooled to 300 deg. Fahr. which is a safe temperature for the oil pump to stop. This is where the timed to open relay (T3) is used to meter the cool down period which is activated when the engine ignition switch is turned off.

       Now with all of the electronics hammered out I was able to mount the ECU inside of the box I had prepared earlier. I had installed a cooling fan under the box to help offset any power turbine heat that may get past the heat shield. The RPM sensor (RS1) was first to be installed as it lays under the main ECU board. I oriented it so I could adjust the RPM limit without removing the main ECU board. I also installed the resistor bank (RB1) to the back of the enclosure which hopefully will act as a heat sink to cool the resistors.

       By using long stainless 6-32 machine screws and nylock nuts I mounted the main ECU to the back of the ECU box. I used 1/4” nylon tubing as standoff spacers which also double as vibration dampeners.

       I imagine this whole unit would be reduced to the size of a matchbox if it were a commercial unit :0} Without any hesitation I set out to connect the ECU to the engine’s sensors and devices. Using different color coded wires, I loomed up a wire harness that connects all of the engines electronic devices to the ECU.

       I installed an Omron E3X-A11 photo optic sensor to the main output jackshaft so the tachometer would indicate the turbine speed. The sensor disc was made earlier for this purpose.

       I installed the control conductors on the Turbotug in preparation for the control panel hookup. I used 1/4” nylon tubing to relay the oil, fuel and combustor pressure to the analog gauges in the control panel. I ordered special “K” type thermocouple connectors so I could disconnect the pyrometer wires from the thermocouple on the engine. This will allow me to use other “K” type thermocouples on other engines with the same pyrometer. I also installed Molex style connectors on all of the control wires so the GR-5A could be removed from the GRV-1 in minutes.
       This was part of my “power dock” design that will hopefully be used to connect up to future engines. All that was left to hook up was the control panel.

       I had all of the necessary wires landed and had just finished plumbing the gauges. There was nothing holding me back from testing the GR-5A. I charged the battery a bit and rolled the kart out of the garage. I had some anxiety in the pit of my stomach as this project represented a lot of hard work. As with my model airplanes, the maiden flight is always the most nerve racking. I turned on the ignition key and the control panel indicated proper oil pressure. I then turned the key to start and with a vacuum cleaner like whine, the GR-5A spooled up.
       The RPM sensor needed to be set so I set up a remote starter switch at the ECU and dialed it in. I spooled up the engine again and got the fuel solenoid to open but no ignition. After fooling around with the throttle switch I figured that I had not opened the emergency fuel shut down valve. Argggggg!!! I spooled it up once more to hear that familiar “poof” followed by the distinctive whine of a turbine. All looked OK as I watched the pyrometer but the engine could not sustain itself.
       The blower valve would not open at the proper time to allow the engine to breathe. This became a big problem as my design was heavily based on the pneumatic cylinder opening the blower valve. I had no other choice but to roll the kart back into the bat cave and rethink the blower valve. Nick, my friend from work dropped by and helped me try to start the engine again. With his help I was able to manually open the blower valve and start the GR-5A. This was good news as the blower was a partial success. Nick had suggested that I remove the air cylinder and replace it with an automotive door lock actuator. This was a great idea so off I was to my local Circuit City to buy a Code Alarm brand door lock actuator for $19.95.                                                                

       I took the actuator apart (mostly to see how it ticked) and greased up the cog gear inside. The brackets on the blower valve had to be revamped to accommodate the new actuator. I installed the actuator on the new brackets and then tested it to see how much current it took to close the valve. The actuator drew too much current for prolonged use so I installed a 5 ohm resistor which reduced the current supplied to the unit. A return spring was installed to open the valve when the blower is not turning. I also installed a new set of terminals on the ECU and reworked the wiring a little to incorporate the new system.

       I tested the blower valve and was very happy with how it performed. The valve opened instantly after the blower relay opened. I rolled out the Turbotug once again and crossed my fingers. I turned the key and “poof” she spooled up slowly. I warmed up the combustor with the bower on to allow for the oil to heat up a bit. I cracked the throttle and watched the pyrometer as the engine whined higher and higher. I was waiting for the combustor pressure switch to turn off the blower and it did after a minute. The GR-5A was running!!! My home-built turbocharger turboshaft engine started itself :0) Needless to say I had a smile from ear to ear.
       I turned on the generator and heard the power turbine RPM dip a little. The engine was running at 850 deg. Fahr. and the power turbine was spinning at 5,000 RPM at 4 PSI combustor pressure. I fine tuned the idle-up set screw and let her run for 5  minutes while I did back flips in the front yard. I had to see if the thing was going to roll so I hopped into the seat and throttled it up. The power turbine spooled up to 6,500 RPM and engaged the clutch. The Turbotug started to roll forward  so I drove it up the driveway almost into the garage. I can’t forget that this thing puts out some serious exhaust as the trees above my driveway were loosing leaves.
       I tested the cool down timer by shutting off the key switch. The blower ran for 3 minutes 49 seconds, cooling the turbo just enough as to prevent the oil from burning on the turbine shaft. I wanted to see how the engine would start after being run already so I turned the key again. Immediately the engine fired up and was idling within 15 seconds. I was incredibly happy with the performance of the ECU.

       I really wanted to road test the Turbotug but it became apparent that the noise level would be too high. My neighbors were very tolerant with my noisy experimentation and I did not want to press my luck. I will plan a day to test at the dry lake bed and take video of the first test runs. In the mean time I opted to tear down the kart and paint it before it makes it’s big debut. As it is, the GRV-1 was accumulating more and more rust with our recent wet weather and I wanted to stop it before I had to sand blast the entire frame.
       For now, I have the frame to strip. I will also start making arrangements to have my friend help video the Turbotug in action.  I can’t wait to see if the Turbotug will be as useful as I planned. Either way, the GR-5A shows a lot of promise and that is good enough for me. I hope to be able to get the kart back together in a week so I can road test it. So far so good!!!

Tune in again for another exciting episode......

Don Giandomenico

[Home] [What's New At RCDON] [Articles] [Aircraft Projects] [Aircraft Videos] [Aircraft Videos II] [Experimental Projects] [GR-1 Turbojet Engine Project] [GR-5 Turboshaft Engine Project] [GR-5A Turboshaft Engine] [GRV-1 Turbotug Project] [GR-7 Turbojet Engine Project] [GRV-2 Jet Bike Project] [GR-6 Turboshaft Engine Project] [GR Engine Series Q&A] [Readers Projects] [Experimental Projects II] [My Collection] [RC Links] [About Myself] [Contact RCDON]