Puppeteer

Author Topic: Another 6/1 Restoration  (Read 121316 times)

BruceM

  • Hero Member
  • *****
  • Posts: 3054
    • View Profile
Re: Another 6/1 Restoration
« Reply #195 on: October 25, 2017, 06:01:25 PM »
Watch out for the commercial under/over voltage monitors-  most will not work with a ST type (spikey) waveform. 

After finding a couple commercial units worthless, on my neighbor's setup we had to go to a more expensive Siemans true RMS voltage monitoring relay to trip a small latching relay controlling our load shed contactor- it works great.

I use RMS calculation via AVR microcontroller (Arduino processor) for my inverter.  With the 10 bit A/D and rectified AC scaled to 5 V, and 32 bit calculations, it has adequate resolution to keep voltage within 1%.  It only samples 64 samples per full wave, as the processor is quite busy with real time wave shaping. For a dedicated processor to do RMS voltage monitoring, there is plenty of RMS sampling code out there.  Mine is dependent with my other real time code which is timer interrupt driven. 

Dieselsmoker

  • If your genset fails, remeber Ps 119:105 - Your word is a lamp for my feet, a light on my path.
  • Full Member
  • ***
  • Posts: 142
  • Self-Certified Professional Engine Lover
    • View Profile
    • Get your spares back!
Re: Another 6/1 Restoration
« Reply #196 on: October 25, 2017, 07:14:40 PM »
Thanks for the warning and saving me some school fees!  I never even considered the fact that a voltage monitor could be sensitive to waveform...

 
I use RMS calculation via AVR microcontroller (Arduino processor) for my inverter.  With the 10 bit A/D and rectified AC scaled to 5 V,....


I more or less follow what you are saying, but
Rectified AC? I suppose you halve wave rectify?
What do you mean with "scaled to 5V"? Did you just use a step down transformer?

Seems like I have a lot of reading ahead of me  ::)


1963 Lister 6/1 genset - Restored
1942 Fairbanks-Morse ZC-208 - Restored
1945 Ruston & Hornsby PB 3HP - To be restored
1954 John Deere 40-S - Current project

BruceM

  • Hero Member
  • *****
  • Posts: 3054
    • View Profile
Re: Another 6/1 Restoration
« Reply #197 on: October 26, 2017, 02:54:55 AM »
To get more resolution with the 10 bit A/D, and to get a faster answer every half wave,  I did full wave rectification.
Some schemes put the scaled AC wave centered  at 2.5V, with peaks at 5 and 0v,  but this costs you over 1 bit of  resolution.   There are many different ways to do it that all work.  The complication in my inverter software case is that I wanted the computed RMS value in real time (8.33ms or a half wave), which gives me most of a half wave to shuffle the new waveform timing data from flash memory to the ram for the next full wave start. While that is happening the processor is banging out via timer interrupt table  the waveform control in real time.  I'm working the little 8 bit AVR chip pretty hard.


BruceM

  • Hero Member
  • *****
  • Posts: 3054
    • View Profile
Re: Another 6/1 Restoration
« Reply #198 on: October 26, 2017, 03:15:41 AM »
I've extracted some of my C code for RMS calculation via Arduino processor.
Because of more critical real time things happening, this uses timer 2 and NO interrupts so that it doesn't interfere with the critical timing of the inverter waveform generation.  I hope this is helpful.  My own AVR uses slightly filtered peak voltage in analog to determine over and under voltage, using analog timing as well to trip the load shed breaker.  If you PM me I can share that hand draw schematic for you if you'd like to roll your own analog circuit solution instead of an arduino based one.  It is only a couple 4000 series CMOS and a quad op amp and a lot of "glue" and trim pots.

Code: [Select]
  // initialize Timer2

  TCCR2A = 0;        //normal operation
  TCCR2B = B00000111;//timer2 prescale of 1024. 64 usec/count at 16MHz clock.
  //                   130 counts per 8.88ms.
  TCNT2 = 0;         //zero timer2 count
  TIMSK2 = 0;        //no interrupts for timer2

  interrupts();            // enable all interrupts
}





void loop()
{
  unsigned long SumV;
  unsigned long Vlong;
  unsigned int RMS;
  byte i;



  //RMS calculations
  //Sync with H-bridge output

  //AD 10 bit value of 1023 should equal 400 volts peak. So RMS goal (230V)= 588.

#define MaxIndex 62

  while (!StartWave) {}// wait for start of sine wave flag set in ISR

  TCNT2 = 0; //reset RMS sample timer counter (timer2)

  if (SwitchTimerFlag) //start of waveform now change index for ISR for new voltage
  {

    Tindx = NewIndex;
    SwitchTimerFlag = false;
  }
  StartWave = false; // reset interrupt routine flag

  // Now collect one half wave worth of RMS voltage-65 samples

  SumV = 0;

  for ( i = 1;  i < 131;  i = i + 2 ) // 64 usec per count so 128 usec per sample
  {
    Vlong = analogRead(A0); //compiler requires long for a sq() or multiply to work properly
    SumV = SumV + sq(Vlong);

    while (TCNT2 < i) {}; // hang for next timer2 count
  }

  // done with real time samples, now calculate RMS and load new TimeTable entry into Timer array
  // digitalWrite(11, !digitalRead(11)); //debug

  RMS = sqrt ( SumV / 65); // approx 512 should be middle of desired voltage range


Dieselsmoker

  • If your genset fails, remeber Ps 119:105 - Your word is a lamp for my feet, a light on my path.
  • Full Member
  • ***
  • Posts: 142
  • Self-Certified Professional Engine Lover
    • View Profile
    • Get your spares back!
Re: Another 6/1 Restoration
« Reply #199 on: October 29, 2017, 07:18:34 PM »




Things don't always work out as planned it seems...

First setback was the overload breaker. I got a "normal" 10A breaker of the shelve and installed it along with a neutral bar to tidy the wiring up a bit. To test the breaker I fired up the engine and lined up some appliances to load the alternator with while giving the old girl some time to build up a bit of heat. Great anticipation sometimes leads to great dissapointment. The breaker reacts too slowley. As soon as the engine speed drops, the voltage drops and the current - well yes you get the picture. There is now way I can get the breaker to trip. Any ideas? I'm not sure if there are differently specced breakers available or if a smaller one will do the trick?

Second problem was the Arduino. I build some circuits on the breadboard and decided to o follow the route of full-wave rectification. If you look at the picture - on the left-hand side I took the 12V tap from the transformer and put it through a bridge rectifier and a voltage regulator with some caps to smooth things out to serve as a 5V supply to power the Arduino.

In the middle is a 5V relay that will eventually be used to operate the contactor.

On thr right-hand side the 6V tap on the transformer is rectified for measurement by the Arduino. The 50K pot is to calibrate the circuit so that the 5VDC that feeds into an analogue input can be calibrated to measure 250V @ 1024 bits. There is also a polarised cap and a 5V Zener in the mix before the signal gets fed into the Arduino. Simple. So it works, but the serial monitor sometimes reports some erratic results - not sure why yet, and I couldn't figure it out since the Arduino stopped accepting sketches getting uploaded  :'(  It looks serious... I think it's dead...

Bruce - your PM is on it's way. I'd like to have a look at what you have. It's gonna take a while to replace the Arduino.
1963 Lister 6/1 genset - Restored
1942 Fairbanks-Morse ZC-208 - Restored
1945 Ruston & Hornsby PB 3HP - To be restored
1954 John Deere 40-S - Current project

BruceM

  • Hero Member
  • *****
  • Posts: 3054
    • View Profile
Re: Another 6/1 Restoration
« Reply #200 on: October 29, 2017, 10:27:56 PM »
Hi Dieselsmoker.  Pity about the Arduino.  Some protection from ac line spikes might be in order, or it could have been a neutral/ground loop issue if you programming computer was plugged in and the generator was on.  If you want to send me a schematic I may be able to spot the likely problems.  It's clear you have some skills if you put this together so quickly- bravo!

 I've sent you my prototype AVR schematic which has adjustable time delayed, high/low analog voltage level sensing...filtered peak, not RMS.  It has served my 6/1-ST3 setup well for over 10 years.  Sometimes analog is just easier.  I set the trigger points while it was on the bench just as you did-  loading the harmonic regulated ST head until voltage sags, or conversely, lighten the load to get a bit of over voltage.  The I increased the time delays to allow for motor starts, or load shedding; allowable short over/under voltage situations.






Dieselsmoker

  • If your genset fails, remeber Ps 119:105 - Your word is a lamp for my feet, a light on my path.
  • Full Member
  • ***
  • Posts: 142
  • Self-Certified Professional Engine Lover
    • View Profile
    • Get your spares back!
Re: Another 6/1 Restoration
« Reply #201 on: November 15, 2017, 07:58:14 PM »
I've made some good progress with the voltage regulating circuitry. Thanks BruceM for all the help and advice thus far. I however must interrupt this topic to report back on the transfers that finally arrived. It was a looong 15 week wait... but well worth it.







1963 Lister 6/1 genset - Restored
1942 Fairbanks-Morse ZC-208 - Restored
1945 Ruston & Hornsby PB 3HP - To be restored
1954 John Deere 40-S - Current project

Dieselsmoker

  • If your genset fails, remeber Ps 119:105 - Your word is a lamp for my feet, a light on my path.
  • Full Member
  • ***
  • Posts: 142
  • Self-Certified Professional Engine Lover
    • View Profile
    • Get your spares back!
Re: Another 6/1 Restoration
« Reply #202 on: January 06, 2018, 08:30:44 PM »
This update is waaaaay overdue. Been on holiday and without any photographic material I dared not post anything!  :)
 
Milestone met! The genset is good enough to show off and it works 100%.
There are only a few more boxes to tick to get to exactly where I want to be - for now I am however content with where I am with the project.

It's been an interesting journey to get the electrics sorted... With the Arduino fried the next option was to go analogue. Thanks again to BruceM for the advice and encouragement.

Circuit operation in a nutshell: Voltage coming from the alternator is scaled down to 12V and monitored by op-amp comparitors. As soon as either adjustable over- or under voltage thresholds are exceeded, the circuit energises a power relay to shed the load going to the plug outlet. There is a time dealy build in to filter out dips and spikes. When the fault condition is resolved, the circuit can be reset to connect the load again. There is a switch in the panel where the monitoring circuit can be switched off - with the power relay wired N/C the circuit can thus operate without intervention. The green LED on the panel door indicates that voltage monitoring is active.

From left to right:
Plug outlet.
Non-latching pushbutton to reset trip.
Green LED - Voltage monitoring active.
Red LED - Output tripped.


Complete circuit assembly:


From left to right:
Neutral bar.
Main circuit breaker.
Voltage monitoring circuit breaker (On/Off).
2x transformers:
     One is to scale 230V to 12V for input into the comparitors.
     The other is 230V to 18V to power the delay-on relay.


The brains of the circuit:


From left to right:
Power relay to disconnect load. (230V Coil).
12V relay controlled by the delay timer. These contacts operates the power relay.
Delay-on timer triggered by the voltage monitoring circuit.


A brief demonstration:
https://youtu.be/vpnZOMdcmFM

For the alternator connection box I also made a cover. Found some plate in the shed and some heavy hammerblows later I had the cover.




On the way back from our coastal holiday, we made a road trip of the 1300Km journey back home. We stopped at a couple of interesting places and avoided the main roads so we could pass through some small towns. In one of these towns called "Willowmore" we browsed through some antique shops. In one of these shops I found a Lister Diesel tank, exactly the same as the one I have, in the back of a dark shelve  ;D The tank is beyond repair, but the Lister cap is intact! Of course it had to come home with me. In no time I had it cleaned up and the piece of crap chinese-made cap  was chucked sidways!



My to-do list to round the project off:
Guages in the panel door. (Volts, Amps, Hour meter, Frequincy)
Probably paint the panel green.
Gib head key covers.
Rocker bushes. (I'll turn these myself)
TAKE IT TO A SHOW!
1963 Lister 6/1 genset - Restored
1942 Fairbanks-Morse ZC-208 - Restored
1945 Ruston & Hornsby PB 3HP - To be restored
1954 John Deere 40-S - Current project

BruceM

  • Hero Member
  • *****
  • Posts: 3054
    • View Profile
Re: Another 6/1 Restoration
« Reply #203 on: January 07, 2018, 01:52:58 AM »
Very nice work on the voltage monitor with time delay and latching relay control of the main power relay!  Your video shows real talent for concise and clear technical explanation...not easy to do.

The whole setup is a restoration work of art and should make some heads turn at the engine shows but would also make a fine home power source as well.  Bravo Dieselsmoker!






 



Hugh Conway

  • Hero Member
  • *****
  • Posts: 529
    • View Profile
Re: Another 6/1 Restoration
« Reply #204 on: January 07, 2018, 06:34:20 PM »
Very nice work Dsmoker! And Bruce, I admire your knowledge of electronics.....All that is a completely foreign language for me.
Here, we are at the other end of the scale with a daily driver listeroid. after all these hours, it is a bit blackened  from the carbon in the oil that inevitably oozes out. A good weekly wipe down helps, but cannot really restore it to the original (none too great )finish.
The only electrics are analogue volt  and ammeters and a vibrating reed freq meter.
I''ve been very fortunate to have sorted out my governor to hold the freq.  to within 1 Hz from no load to full load.
I am at about 60 meters above MSL and can pull a continuous 3600W@ 240V without anything complaining, though normally ask it for about 2900W when charging batteries. We once pushed it to output a measured 4000+ watts, but it was asking a lot.  That great Utterpower PMG was worth every penny!
We've logged just over 1500 hours now and the machine sits in a corner of a 3 sided carport exposed to the cold and damp of the coastal islands. As they say, "it ain't pretty, but it does work" Clean fuel and regular oil changes seem to be all it wants. A Solberg air filter/sound suppressor screws right into the standard intake adapter and  really quiets the intake noise.
I do have a Dursley 6/1 that was intended to be more of a show machine, but in the end it will also have a regular job.
It is quite a diverse group and large skill set here on this forum. I visit nearly every day to learn from you all.
Cheers,
Hugh
JKson 6/1  (Utterpower PMG ) Off-grid
Lister 6/1 Start-O-Matic engine......running with PMG
1978 Royal Enfield (glutton for punishment by Indian iron)
1963 BMW R-27 project

BruceM

  • Hero Member
  • *****
  • Posts: 3054
    • View Profile
Re: Another 6/1 Restoration
« Reply #205 on: January 07, 2018, 07:04:08 PM »
Thanks, Hugh, I have learned much here as well.  George B's management of the development and production of the PMG heads was very impressive and a huge undertaking.  I was an R&D manager and can really appreciate the task. 

Regarding intake noise, I think 38AC's work has convinced me that this is likely an intake cam timing issue,  since some have it an others (like my older Metro and neighbor's DES 8/1 propane conversion) do not.  I wonder if you could also make it go away by increasing intake lash to retard intake open timing?  It's certainly an easy experiment.



Dieselsmoker

  • If your genset fails, remeber Ps 119:105 - Your word is a lamp for my feet, a light on my path.
  • Full Member
  • ***
  • Posts: 142
  • Self-Certified Professional Engine Lover
    • View Profile
    • Get your spares back!
Re: Another 6/1 Restoration
« Reply #206 on: May 21, 2018, 08:18:09 PM »
Hi Listerines!
I took the Lister to its first show a couple of weeks ago. Pity I only learned about it the day before so I was not prepared at all. It was just load and go! It was close to where I live from so I pulled the trailer with the Lister on to the show with the John Deere. There is a bigger show coming up in September I'll prepare better for. I'll think of something nice for it to do and also make a nice sign for it with some specs and pictures maybe. 

It was a car-show organized for a church charity that was held at the local Primary School. I stole a lot of limelight with the only stationary engine and tractor there  :D The Lister drew a lot of attention - especially with the older generation that grew up with these around on the farms.



The John Deere 40 was a hit with kids. They climbed on one after the other to pose for pictures on it.  8)


My son demonstrating the PTO pulley:


On Sunday we FINALLY had a power cut for two hours and I could drag out the Lister to do some REAL work.
https://youtu.be/2o6lEUi1cbY


I have a question about the exhaust of the Lister. If it's nicely warmed up and under load - how clean should the exhaust be? It always smokes a bit - never totally clean. It definitely gets better when nicely warmed up so this is not entirely as bad as it is when worked for 2 or more hours. Check this short video: https://youtu.be/zWEAvepyJRM?list=UUhqCC4-8MJq7uqgqdUSmlzg

I upped the injector opening pressure somewhat:


The spray is not perfectly atomized... hoe does this look? (It took a couple of tries with the cellphone camera to catch the image)
1963 Lister 6/1 genset - Restored
1942 Fairbanks-Morse ZC-208 - Restored
1945 Ruston & Hornsby PB 3HP - To be restored
1954 John Deere 40-S - Current project

BruceM

  • Hero Member
  • *****
  • Posts: 3054
    • View Profile
Re: Another 6/1 Restoration
« Reply #207 on: May 21, 2018, 08:45:29 PM »
Bravo on your show debut. Your restoration is most impressive.

I don't recall your engine's break in status, but it often takes 100+ hours under load for a CS to fully seat the rings.  My exhaust looked even better after 200 hrs. The exhaust should be nearly clear when warmed up under moderate load.

It's hard to judge your spray pattern from a photo so close to the nozzle but it might be too spray/jet like and not like a "burst of fog".  (May not be atomizing well.) I'd swap in a new nozzle or injector and see if that helps.  Check it visually first so you can see what a good one looks like.  Your injector may be fine; I can tell more from the sound and mist from it hitting in a coffee can since that's what I use with my DIY grease gun pop tester.  A jet is not ideal. A bit of carbon in the nozzle can cause a jet, as can poor pintle fit, in my limited experience.




Dieselsmoker

  • If your genset fails, remeber Ps 119:105 - Your word is a lamp for my feet, a light on my path.
  • Full Member
  • ***
  • Posts: 142
  • Self-Certified Professional Engine Lover
    • View Profile
    • Get your spares back!
Re: Another 6/1 Restoration
« Reply #208 on: May 22, 2018, 10:03:50 AM »
Bravo on your show debut. Your restoration is most impressive.

I don't recall your engine's break in status, but it often takes 100+ hours under load for a CS to fully seat the rings.  My exhaust looked even better after 200 hrs. The exhaust should be nearly clear when warmed up under moderate load.

It's hard to judge your spray pattern from a photo so close to the nozzle but it might be too spray/jet like and not like a "burst of fog".  (May not be atomizing well.) I'd swap in a new nozzle or injector and see if that helps.  Check it visually first so you can see what a good one looks like.  Your injector may be fine; I can tell more from the sound and mist from it hitting in a coffee can since that's what I use with my DIY grease gun pop tester.  A jet is not ideal. A bit of carbon in the nozzle can cause a jet, as can poor pintle fit, in my limited experience.

Thankyou!
I have no idea how many hours are on the engine... I have that hour meter on my wishlist  ::)

It is possible that the engine is not fully run in, but I am not totally happy with the spray pattern... I have a Mercedes 300D nozzle I'll try as reference (if it fits in the injector holder) - never thought of doing that while I had everything set up! 

What is the best way to determine the optimum opening pressure? Surely things like atmospheric pressure will have an influence? Or is it not too critical as long as it is near the 90 atm mark the manual recommends? I just checked something: 90 atm = 93 Kgf/sq cm so my setting is pretty accurate. I did play with the adjustment screw while the engine was running and a full turn either way had very little noticible effect on smoke or power output. (I'm not sure how much one turn affects opening pressure - I'll check when I take the injector out again)

The injector is as clean as I could get it - but I won't be surprised if the hole is not perfect as the injector tip is pitted by corrosion. I'll take it to a Diesel service centre who has the tools to clean this out properly.
1963 Lister 6/1 genset - Restored
1942 Fairbanks-Morse ZC-208 - Restored
1945 Ruston & Hornsby PB 3HP - To be restored
1954 John Deere 40-S - Current project

Dieselsmoker

  • If your genset fails, remeber Ps 119:105 - Your word is a lamp for my feet, a light on my path.
  • Full Member
  • ***
  • Posts: 142
  • Self-Certified Professional Engine Lover
    • View Profile
    • Get your spares back!
Re: Another 6/1 Restoration
« Reply #209 on: May 22, 2018, 10:21:28 AM »

You have done a brilliant job on your machine!
It looks awesome and runs beautifully smooth as well.

The injector on my machine fires in a circle pattern. The fuel comes out from every side, not firing like yours appears to be to one side. I agree with Bruce, I can tell more about how they are firing from the sound, at least on my china diesel which I have had more hands on time with.  Had to pull that apart a few times due to not running it for over a year and the veg fuel gumming it up. No big deal to clean it but as soon as I wind it up I can tell by the sound if it's right or not.

Given the time and incredible effort you have put into the thing, a few more bucks on a new injector may be worth while if you can't clean this one.
The ones on my china diesels come apart easily and there isn't much to them but they are easy to clean out stuff you can't see but just won't clean themselves with any sort of solvent when running.

Unscrewing the muffler may give you an indication as to how the engine is running. If it's wet in the exhaust port at all something's not quite right.
May not be bedded in with the rings or it may be the injector but I would eliminate the easiest possibility first and then give it some time and see how you go.

Would be best at this stage to always run with a load on the thing. If you have electric hot water heating you could always hook it up to that to give it a load to work from and get some benefit out of it.

Thanks Glort. The encouragement and advice gained from this forum helped a lot to get the job done!

The injector should spray to the side for the stream to align with the hole in the head. Other types of injectors do have a conical spray pattern where the tip is exposed directly to the combustion chamber. On page 12 of this thread there are some books that explains this - but you probably know this already. Are you using the conical spray tip in the Lister? If yes it will make finding a replacement tip so much easier!

I'll take the muffler off and check the port. The muffler definately catches some unburnt fuel or oil when the engine ran for a while with no load. When connecting a load there is clearly a lot more smoke for a few minutes before it improves. Usually when I let it run I connect it to the house so, like you said, it does something productive and not just burn the fuel for nothing.
1963 Lister 6/1 genset - Restored
1942 Fairbanks-Morse ZC-208 - Restored
1945 Ruston & Hornsby PB 3HP - To be restored
1954 John Deere 40-S - Current project