DC generator braking for engine dyno 2

SACEM1: There are two basic ways to load the generator. The first is a rheostat to add resistive load. Instrument for voltage and current and you do not need a load cell or pivot.

The second is similar. The navy uses a salt bos to load generators for test. I believe, however, that the salt box is not controlable for load.

You also need an electrical person for the details, such as the output vs. rpm for the generator.

Regards
Dave

Measuring the torque is as easy as you say.

A cheap source of bulletproof electrical loads is old immersion heaters. From memory they are 3 kW or so each. Dunk them in a drum of water.

I don't know how to adjust the load electrically, those are big currents.

I think you'll find the voltage is proportional to the speed of the generator, in which case you are going to have an interesting problem or two - at low speed you will exceed the current capacity of the generator. I don't know what that means in practice.




Cheers

Greg Locock

Please see FAQ731-376 for tips on how to make the best use of Eng-Tips.

Sounds like an interesting project.

Use the generator to feed electrical power into some sort of fan or water cooled resistive load bank. The load will need to suit the voltage and current that the generator is rated. The rating plate should tell you everything you need to know.

Power to the field winding can come from a variac, and a suitable transformer/rectifier, again sized to suit the ratings of the particular generator.

By turning the variac up and down, or adding/removing electrical load, braking torque can easily be adjusted over a wide range.

You could use domestic water heater elements.
They are 4 KW at 240 volt. You could exceed their rating for short times if they were clean. The screw mount version screw into a one inch pipe thread which you could weld into a tank. You might want to pressurize the tank to limit boiling but use safety valves and discharge away from personel.
At 100 Hp you would produce 75 KW or about a quarter million BTU/Hr. If the generator is 240 volt this would be 312 amp.
Exceeding element rating by 50% would give you 6 KW per element and require 12 elements. The 50% over rating would require 294 Volts which your generator probably would produce with slightly higher exciter voltage and/or over RPM. Current at 100 hp would drop to 255 amp.

Another way is to go out and buy a large quantity of standard 1.2Kw coiled ceramic electric radiator bars and simply bolt them between two rectangular steel plates. A very large forced air fan will be needed to get rid of the heat. An excellent winter project.

It is best mounted outside, the hot smells, if not the heat itself will be fairly unpleasant for a while, every time it is first fired up.

You could be the guy with the biggest electric toaster on the block.

Thanks to all of you guys for the answers and suggestions:

As I need to load gradually the engine at a set rpm until it bogs down and read at precisely before that point what was the torque reading I will have to go to 24 water heater elements (I've found 5 Kw rated elements @ 240 V at cheap prices and put them in the lower part of a water tank left open for evaporation cooling at the outside of my testing room (or I would end in a sauna bath) install several switches to allow me to vary the load applied ( controlling: one for 1 resistance, other for 2 resist, other for 4, other for 8 and one for the 9 left) that way I can switch around any load at 5 Kw intervals up to full 120 Kw load.

A variac would be nice to have if I can get my hands on a cheap used one.

I will go by the pivot and load reading because it is more reliable than reading the current from the old generator and being sure that all other variables and efficiencies are taken into account.

What is still not clear is what GregLocock said about what will happen when the engine is at 1500 rpm's and the generator at 386 will it have enough resistance to the power output of the motor or will it go its Amp output to the sky.

If voltage is speed dependend and the output is constant then the Amps will be very high at low generator speeds just to keep the output constant so thats why I will have the variable switching resistances and that in turn would make it very complicated to deduct final motor output by any other way that it is not by load reading.

Thanks

SACEM1

The electrical portion of this situation should be posted to

Electrical Engineering -

Electrical Motor group and
Electrical power group.

Thanks

I will do so.

SACEM1

First off, restrain the engine, generator and scale as a unit; if you put a scale on the ground and load it, you may tip over a light buggy. Don't know what you have in mind physically, but I got that off my chest.
Back in college- welding engineering- we checked large welders with a plastic garbage can of salt water, raising/lowering 2 sheets of steel spaced about 2" apart using a simple string/crank mechanism.
I did a small 100 amp welder using a small jar of water (only a few seconds at a time) and a door spring- I clipped it a various locations to change resistance.
You know, you may get enough load adjustment out of generator using a good sized potentiometer to vary field strength. I did this with a 7 hp diesel engine used in a hybrid electric car project.
GOOD LUCK

Well thanks for your worry about tiping the buggy, but its not really so little, I have not put in on a scale but it should be around 1,200 lbs. But the engine will be on a stand not on the buggy, the stand will be fixed to the floor and the engine tested standing alone.

I'm still looking for an electrical guy who will check the DC generator to be sure its working after being so many years on storage.

I'll keep you up to date on further advances,

Regards

SACEM1

In my engineering school ME lab we had a dyno just as you described, a very old, very large DC generator (can't remember how many kW) mounted on bearings, connected to a big resistor bank electrically - and to a Toledo scale via a torque arm.

My senior project involved testing a racing engine, so we drove the generator through a transmission, much as you describe.

There were, however, a couple of major drawbacks to this approach:

The biggest was that the rotational inertia of the system (mostly the big generator) was so high that the darned thing acted more like an inertia wheel than a brake - when we'd change the load to change the rpm, the torque would take forever to stabilize.

Second, and more obvious, was that an unknown but significant amount of power was being consumed in the transmission. And that consumption was (presumably) pretty temperature dependent, so it was hard to be sure that small changes were real or simply test noise.

All in all, I wound up hating the setup. Hopefully yours will work out better.

Al Seim

Hey thanks Al:

You know I had been wondering about the rotational inertia of the generator and the effect that it would have on the motor speed regulation, just what you have told me that it would happen, and that fact was keeping me from going ahead with the project.

I thing I'll go the water brake route as soon as I can get my hands on a suitable design.

Thanks for your information, it made my choice easier.

SACEM1

One way to overcome inertia is to use a gearbox as already suggested. That lowers generator shaft speed, but it also reduces the inertia by a very large amount.

The trick is to solidly fix the gearbox case to the generator frame, and pivot the whole lot together. You will then actually be measuring the torque INTO the gearbox. Gearbox losses then just become part of the total absorption load that you are measuring.

It may also be convenient to change gear ratio for different types of engines. That will not effect the torque measurement if gearbox input torque is what is being measured.

Fit amp meters to the field winding and armature. As long as either rated current is not exceeded, the generator will be perfectly happy over a wide Rpm range. Sustained operation at very low generator Rpm may require an external cooling fan for the generator, as the internal fan may not be moving much air at low speeds.

An ancient truck gearbox may be ideal. They are usually very short having no extension housing, and the ratios will be wide with a very low first gear. That will certainly reduce inertia for smaller capacity engines.

I was planning on fixing the engines directly to a gearbox as yhese will mainly be aircooled VW's used for sand buggy's but maybe I can get an old gearbox (pre '61) that is bolted to the bellhousing and use a standard speed reducer at the shaft input (I manufacture those so have plenty of sizes around) and retry the idea.

Well seems like my mind is not made up yet but with so many good ideas flowing in it makes you look at it in different ways

Thanks

SACEM1

Sacem:

I like the idea of mounting the reduction unit on the generator, that certainly eliminates the transmission loss variable.

Depending on the inertia of the generator armature, how much you can gear it down and still absorb the power, and how happy your engine is at sustained high revs, you might have a usable setup after all.

In my case, I had been having bearing problems, and it was torture watching the engine scream at 7000+ RPM for 30+ seconds per data point (I really don't remember how long, but it seemed like forever) waiting for the reading to settle down. And when I tore the engine down post-test, the bearings really were starting to go.

On the plus side, I was amazed how easy it was to find power with a real dyno!

In any event, good luck!

Al

Top speed for an aircooled VW engine is around 5,500 rpm if it is prepared for high torque more than max HP, that's what you need when you are going hill climbing or loose sand dune climbing.

I really want to try for high torque values in the 2k to 3k range where most driving is done with a broad torque band so that the car is very tractable on really rough roads.

Thanks again

SACEM1

I have made good power and a wide torque band with air cooled VWs used in speedway midgets at engine speeds to 7000 rpm and in excess of 200 real (Dynoed) HP from 2.2 litres with injection and methanol.

You can always increase the torque by varying gear ratios.


I have made daily drivers from VW type 1 motors that ran for years with regular excursions to over 6000 rpm. These used original VW 1600 cranks with welded on counter weights and Cima 94 mm bore kits, original VW head castings with welded up inlets to allow more porting and twin 40 mm Webber DCN twin choke, down draft carbies with 32 mm chokes. They were 1913 cc and made around 130 HP, with good torque and tractability from 1000 to 6000 rpm.

There was a book by HP publications called "How To Hotrod VW engines" If you can get a copy, read and understand it, you will find all you need to know. Pay particular attention to oil cooling and filtering, sump depth, oil pump and air filters if you want long engine life.

For high rpm and long crankcase life, a fully counterweighted crank is a must.

To run high compression ratios, you will need to keep the oil very cool, de bur and polish the combustion chamber and valve heads, seal all air leaks from the tin work (I used silicone rubber) and take great care than the ducting is such that all air that is pumped by the fan, passes evenly over the motor and is not recirculated into the fan inlet. Air being drawn back from the region of the crankshaft pulley to the fan inlet, and air escaping from missing heater ducts is a major source of overheating.

If you are running a Superbug motor, the stock cooler and ducting is barely adequate for mild hotroding. Anything earlier you definitely need a remote cooler.

If you run a remote cooler and filter off a line from the back of the pump, you will need an oil thermostat so as not to try and push cold oil through a long line, filter and cooler in series. I always used the stock Superbug pump, hard anodised back to 0.001" clearance with excellent results. I found the original biggest available aluminium pumps always fitted and sealed in the crankcase better than cast iron aftermarket pumps, which were prone to suction leaks between the crankcase and the pump suction port

I had a good oil cooler, and I ran the old 36 HP top shroud and fan, and an aftermarket "small pulley". This gave a power boost you could feel, and although it pumped a lot less air, it put it all where you need it. I also ran 36 HP J pipes from the 2 front exhaust ports as they are lighter and less restrictive, being tubular steel instead of cast heat exchangers.




Regards

eng-tips, by professional engineers for professional engineers
Please see FAQ731-376 for tips on how to make the best use of Eng-Tips Fora.

Hello Pat:

Thanks for your good advice and your time for doing it.

I use rright now a single relief case with late model Brasilian van heads with the biggest stock valves that come from VW, dual port with homemade header to a AFV 446 carb (used as stock on 318 cu in Mopar V-8 that is a brazilian made copy of the Weber 446 carb, dual choke) each one feeds a cylinder bank has proven fail proof on really long treks that have DNF cars with webers and solexes.

Right now internals are pretty stock but oil is roted out from the pump via 1/2" hidraulic line into a multitube cooler (12" x 16") that has an electric fan fitted with a shroud and electric switch to operate, fitted just below the roll cage level, oil goes back into engine in drilled and tapped port next to oil relief valve. Filtering is done by a second line that comes from the stock oil unused (now) port and goes to a fram filter and back to the return port of the stock filter. The pump is aluminum with 30 mm high gears, pressure is not a problem here.

Right now I'm building a new engine with a crankshaft that has been welded with counterweights and stroked to 82 mm with stock VW conecting rods and plan to use 92 mm pistons, your Cima recomendation will be followed, but I will have to import the pistons as no larger than stock pistons can be found here in Peru.

The heads are being ported and polished following the advice found in the Fishers book wich I allready have and recommend to anyone involved in this subject. I've manufactured here a set of high lift rockers in order to use them with a stock cam I aimed at 1.3:1 but after assembly and checking with a dial depth gage found out a 1.36:1 ratio wich is pretty good I think.

My concern right now is the pressure plate and its ability to hold the new power output as right now when in deep loose sand or triying to pull someone else who's stuck the clutch makes a high pitch cry but the engine does not bog down too much and if I press the clutch it will not stall so definitly I'm at the clutch limit. The disk upon dissasembly does not show too much deterioration so i think the problem is in the pressure plate, by the way it is 200 mm stock van pressure plate.

My gearcase is stock type 1 with swing axle and old type 3 gearreduction cases at the wheels so torque is really multiplied by 1.46:1 and I have another set of reduction gears that give 1.63:1 but have not tried them yet, the reduction cases are laid down and have double shocks and restraining strap in order to limit downward fall of the wheels upon a jump, just like it is described in "Baja Bugs and Buggies" by Jeff Hibbard also an excellent book on that matter.

Finally the reason that I want the dyno is to be able to tune and try out different settings for my motor and other friends also in order to evaluate the mods we are doing to the engines as most are done locally and we do not have easy and quick access to over the counter pieces as in the states.

Those figures that you gave me about the engines HP's are really interesting and maybe I'll try also with double 446's one over each cylinder bank.

I am also thinking on using, if I go the 2 carb route, a 911 Porsche style fan, located in the center and with shrouding that forces air to go around the cylinders, does anyone have experience in that type of fan, my theoretical data says that a fan will not increase its power absortion in such a steep way as a centrifugal (stock type) fan does, so that would allow good refrigeration without the high power loss that represents a centrifugal fan at high speed.

Cheers

SACEM1

Having run engines on dynos through a transmission, I'd strongly suggest coming up with a means for cooling the tranny itself.