Wood gas electronic governor controll

[Patricktj]

Hi tcmtech
Unfortunately I can't remove the gearbox off the front of the alternator, it was built as a tractor driven unit with the gearbox cast in to the front housing! The alternator runes at 1500 rpm, I even looked at extending the shaft out of the rear of the alternator to power if from the back. But it would have cost less to buy a new one!
Yes your right wood gas has a slower flame, so lower revs are better, they say you can't exceed 3000 rpm but the guys In holland and Sweden are running Volvoes up to 4500 rpm quite easily. So I think it depends on the quality of the gas. I have taken the Chevy to 3500 rpm on wood gas, but not under load! Only time will tell !

Hi Alec
Thanks for the diagram.
Is the system able to be adjusted, ie variable speed ?
The standard servo you suggest, where can I get one and what is the cost ?
Thanks patrick

 

[tcmtech]

Going in from the gearbox end would be make more sense being there has to be some way that gear and bearing set on the drive end of rotor shaft gets put on.

Some pictures would be helpful here.

 

[alec_t]

Is the system able to be adjusted, ie variable speed ?

Yes. It includes the trimmer for minor adjustment. Major adjustment would require simply changing at least one other component value. But the servo could be mechanically arranged to provide only part of the throttle movement, a hand throttle providing the bulk of the movement.

The standard servo you suggest, where can I get one and what is the cost ?

Any good modellers' shop/supplier would have one. Cost could be ~ $5 - $100 and depends on quality, e.g. heavy-duty, high torque, ball-raced shaft, metal gears all add to the cost. I guess you'd want a fairly robust servo. What force is needed to move the throttle?

 

[Patricktj]

Hi tcmtech
Will try get some photos today.

Hi Alec
The throttle valve can be moved with a piece of cotton thread. So probably about 0.1 Nm - 0.16 ftlb.
Alec I'm going to give your diagram to a firm in town that say they can build it for me, if that is ok with you?
Thanks


Thanks guys for the help.
Patrick

 

[ericgibbs]

hi alec,
Ref your circuit, I see its producing the accepted 20mSec frame period for a servo, but the amount of change from 50Hz to 45Hz in 'high period' is too small from .4mS to 1.33mS, also what happens when its 55Hz.?

To cope with +/- variations in throttle control, the 50Hz high period for the servo drive should be mid range, approx 2mS high.?

Am I missing something.?

Eric

EDIT:
A typical min would be 1mS, mid range 1.5mS, max 2mS, within a approx 20mS frame period.

 

[alec_t]

Check the robotics/modeller's sites for a servo with enough torque. There should be lots, at a reasonable price, that can exceed 0.2 ftlb. I envisage the servo being mounted on a support movable by a hand throttle control for major adjustment, the servo arm being coupled to the actual throttle lever to provide auto minor adjustment to hold the speed constant. The circuit will respond noticeably to rpm changes of < 1%. Its response time is 2 input pulse periods, i.e. 20ms for a 100Hz pulse input rate. The servo response time is dependent on the chosen servo but I'd guess is < 50ms. Whether the overall response time is short enough to avoid hunting is unknown, but I don't see how you could reduce it significantly. If hunting does prove to be a problem you would need a more elaborate PID control system.

Alec I'm going to give your diagram to a firm in town that say they can build it for me, if that is ok with you?

Fine by me, providing you accept the caveats in my posts' signature and providing they're aware that the schematic is conceptual, rather then a full circuit diagram, so is missing supply decoupling capacitors, pinouts etc which they'll have to allow for. In particular, depending where you get your input pulses from some signal conditioning circuitry (filtering, limiting, amplifying) may be needed before the circuit input. As mentioned above you may need to swap a component if your target rpm is not 3000 as originally specified.

 

[alec_t]

@Eric

Am I missing something.?

The divide-by-two action of the D-latch?

A typical min would be 1mS, mid range 1.5mS, max 2mS, within a approx 20mS frame period.

That's what I got with my sim, with input pulse rates 105Hz, 100Hz, 95Hz (pulse periods 9.5ms, 10ms, 10.5ms) and the trimmer set at 0.72 . The circuit design is for rpm errors up to ± 5% (not your 10%). Errors beyond that will give output pulses < 1ms or > 2ms, but the servo will (I think!) regard anything <= 1ms as "full one way" and anything >= 2ms as "full the other way".

 

[ericgibbs]

alec,
I see a possible point of misunderstanding, your posted asc file is set for an ignition rate of 1mS+10mS = 11mS == 90.9Hz.??

Lets hope Patrick can get a nominal 100Hz pulse.

Eric

EDIT:
I see whats happened, I have assumed 1mShigh 10mS low...=11mS

 

[alec_t]

Lets hope Patrick can get a nominal 100Hz pulse.

If not, changing R1 would be simple. The consequence of lower pulse rates would be coverage of a proportionally smaller error range.
Can you (Eric) /anyone confirm how standard servos behave for pulse widths < 1ms or > 2ms ?

 

[ericgibbs]

If not, changing R1 would be simple. The consequence of lower pulse rates would be coverage of a proportionally smaller error range.
Can you (Eric) /anyone confirm how standard servos behave for pulse widths < 1ms or > 2ms ?

Most have a rotational range of 180deg between their end stops, some have 90deg end stops, also there are 360deg continuous.

For the mS values you quote the mechanical end stop should limit the rotation at either end limit.

Typical petrol driven carb' throttles have a spring loaded return to the closed position, the servo will have to hold the throttle open.

From memory, the accelerator operating arm moves the throttle over a 90deg rotation for full open, which would suggest a 90deg servo with direct drive.??

My concern of the basic circuit and servo, as I would expect a slow engine speed change response, would be a 'hunting' control response.

I would suggest Patrick makes a short video of the carb/throttle movement range, say for +/-5% on 50Hz

E

 

[alec_t]

the servo will have to hold the throttle open.

Most servos have high enough gear ratios that, even unpowered, they can hold against a considerable force.

From memory, the accelerator operating arm moves the throttle over a 90deg rotation for full open, which would suggest a 90deg servo with direct drive.??

A simple mechanical lever/linkage/cam/whatever between servo and throttle could provide movement amplification/reduction if required.

My concern of the basic circuit and servo, as I would expect a slow engine speed change response, would be a 'hunting' control response.

That would be my concern too; hence the reference to a PID control.

 

[ericgibbs]

hi Patrick,

To help you choose a suitable Servo and gear train if needed, I have put together a list of links that may help.

E

https://www.sciencebuddies.org/scie...ct_ideas/Robotics_ServoMotors.shtml?from=Blog

https://www.galilmc.com/learning/tutorials.php?gclid=CMXn9srmibkCFfIPtAodi3AAbg

https://www.youtube.com/watch?v=vdYFJx4FrBQ

https://www.youtube.com/watch?v=4HlmNelwGUA

 

[tcmtech]

I am curious to see where this goes to be honest.

I hear a fair amount about wood gas powered engines but I have yet to find anyone who ever used it long term that is willing to talk about the long term maintenance issues and realistic operating costs.

They all start out bragging about how cheap it is to run but somewhere it mysteriously falls off to the side never to be touched again without explanation.

 

[shortbus=]

I am curious to see where this goes to be honest.

I hear a fair amount about wood gas powered engines but I have yet to find anyone who ever used it long term that is willing to talk about the long term maintenance issues and realistic operating costs.

They all start out bragging about how cheap it is to run but somewhere it mysteriously falls off to the side never to be touched again without explanation.

I don't think it would work too well on a modern high compression engine. It did work pretty good on the old T-Ford and Model A type engines, that were low compression. They were in the 5/6:1 range instead of the modern ~9:1 range.

 

[Mr RB]

... but I have yet to find anyone who ever used it long term that is willing to talk about the long term maintenance issues and realistic operating costs.

They all start out bragging about how cheap it is to run but somewhere it mysteriously falls off to the side never to be touched again without explanation.

Because they don't like pulling the entire engine apart every couple of months and cleaning all the mungey tar buildup out of the cylinders, piston rings, valves etc?

In the old days that was much easier, especially with simple old-style engines that an amatuer could pull the head off with a few bolts and do the cleaning themselves.

 

[tcmtech]

That's what I suspected.

I figured wood gas is rather dirty so there had to be a fair amount of tar and ash buildup going some place!

 

[Mr RB]

People usually use one or even two cyclonic filters to get the particulates (ash) out, but there will still be tars in the wood gas itself.

Still, a 6cyl Chevy stripped down to the minimum (ie carb and pollution gear removed) is not to hard to take the rocker cover and cylinder head off.