Water Power

Run your car or bakkie on water with a HHO or HHOO generator.

You may e-mail me at john@hydroxypower.co.za with questions, suggestions or if you can help in any way.

Latest update   5 January 2012

New! How to build a basic hydrogen-on-demand wet cell.

Keep it green with organic Septic Tank and Pit Latrine Treatments

.

PETROL CONSUMPTION                       

As at beginning August 2009:                                                      

I have revised my guestimation regarding my consumption following on my previous figures. I am now getting about 50% better consumption with a combination of town and in between town driving! 

But ... I found today that my reservoir has sprung a leak from which electrolyte and HHOO gas has been escaping.  The problem is that the fuel pump which circulates the electrolyte creates pressure in the system and as I have explained before the 110mm pvc pipe with the end caps is a pain in the gaai to seal properly. I am presuming that the leak only occurred sometime this morning (4 August 2009). The leaks occurred at the top of the reservoir between the collar and the pipe, as well as where the end cap screws onto the collar. 

(The silver foil on the pipe was to shield it from the radiator hose where the two touched)

So ... I have now taken another look at my setup and I have found that the Merc radiator that I have does fit in the space where I had the copper core radiator, I just had to rotate it into a horizontal position and reposition it slightly.

I decided to stay with the old pc power supply fan as it is better that the 4 fan combination which I previously had thought of using with this radiator.

This radiator holds a slightly greater volume of electrolyte than the previous radiator.  I turned the fan around on the housing onto which it is screwed so that it now sucks air through the radiator. The assembly is positioned so that there is a flow of air through a vent in the front bumper and through the radiator. The fan is the on the opposite side, or rear of the radiator.

Today (4 August 2009) I ripped out the leaking reservoir and copper core radiator and I installed the bigger radiator above. But it still does not make up for the extra electrolyte which the reservoir held. I firmly believe that about 4 litres of electrolyte or more must be in the system to help control the temperature.

I have set the amps with the new setup (when slightly warm, 35ºc) at 9,1A. This includes about 1A from the fuel pump which is on the same circuit. This setting gives me just over 1 litre per minute. I think that I can expect problems with heat issues but I will run it like this and see how it goes.

If I can find a suitable container for a reservoir then I will install one again. It has to be robust and able to take pressure. So the hunt for something suitable continues...

And I still have the problem of the fuse heating up! But I will overcome it...!

5 August 2009                                                                                           

I went for a short drive today (about 20km), mostly in 80kph zones. I was very surprised that the fuse held out. When I returned home I felt the fuse holder and it was cooking! The electrolyser  was drawing 9,5A (minus about 1A for the fuel pump). When I opened the fuse holder the fuse picked that moment to fall apart. If the fuse holder hadn't been bakelite then that would definitely have melted! I felt the negative terminal on the electrolyser and that was also very, very hot and when I checked whether the wing nut was tight, the whole bolt assembly turned. I now knew that I had no other choice but to take the electrolyser out and strip and re-assemble it.

I have been puzzling over why the negative terminal gets so hot, and the only thing that comes to mind is that the cable to earth might not be making good contact. This cable was bolted onto a handy bolt on the underside of the engine. I then remembered problems that I had when I still had my Toyota Venture. These problems were caused by a faulty earth. In this case I eventually decided to put my own earthing strap on from the chassis to the engine - problem solved. So I have now earthed the cable from the negative terminal directly onto the chassis, I hope that this solves the problem. I'm also hoping that this might have something to do with my problem with the fuse on the positive line - if this is the case then I've solved it as well. Here's to hoping...

The electrolyte was a very rusty brown colour. The plates also had a layer of gunk over them. The inside of the housing and the tubes were stained the same colour. I am hoping that the colour and gunk is more from the copper core radiator that I had been using rather than from the plates or my choice of electrolyte.

The housing after being washed out with pine gel. The effects of the heat on the negative terminal can clearly be seen. The plastic has melted and warped, and the fibre washer is now embedded into the housing! This photo was taken of the inside of the housing.

Instead of stripping the cell down, I decided to use the other cell which has the 2 extra neutrals (sorry for the bad pic). I did make it slightly smaller by removing one nut (used as a spacer) on the right of the left (extra) neutral, i.e. I removed one nut on the right of the plate which is fourth from the left.

When I re-assembled the cell in the housing, I decided to use an extra 2 fibre washers with the embedded one. When I assembled this cell, it was the opposite way around to the one which I had taken out, so the positive terminal was now swapped around with the negative (you can see where I remarked the + and - on the outside of the housing with a marker pen).

The cell assembly with the inlet and outlet pipes in place.

The inside of the lid. I have run a bead of silicon along the rubber insert to ensure a good, airtight seal. The lid is now in place on the housing and I will leave it to cure overnight and tomorrow I will fit it into my car.

I think I need to connect the fuel pump and the fan for the electrolyte radiator directly to the on/off switch on the dashboard to also take some of the load off the relay and the fuse. The fan draws minimal current (milliamps) but the 1A or so that the fuel pump draws might just make a difference. I am hoping that all the little mods will eventually make the whole system run better.

6 August 2009                                                                                                 

I have now connected the fuel pump and the fan for the electrolyte radiator directly to another on/off switch on the dashboard.

I ran the electrolyser and to my dismay there was no gas! Ag no, it could only be a leak from somewhere. I took it back to the bench and connected it up to my pc power supply...plenty of gas! What now? I felt a real poephol when I realised that I had not connected the leads to the terminals when I tried to run it in my car the first time!

I connected it all up again, and the power leads, and fired it up. The fuse popped within about 1 minute. I thought that this fuse might just have taken too much strain from previous trials and so I put in another. Same result, ag nee!

I first checked all the connections and for any obvious short circuits - nothing. With my multi meter I then checked across the positive and negative terminals on the electrolyser. The beep told me that it was a dead short!

I pulled the electrolyser out and opened it up, fully expecting to see a faulty assembly with contact somewhere between the positive and the negative...nothing! But then I saw at the bottom of the cell between 2 of the plates was a bridge of bicarb paste. In the bottom of the housing was quite a thick layer of bicarb in which the plates had been  immersed. I know that bicarb does not readily dissolve and it was this that was then causing the "short", not just this bridge but the whole bed of bicarb which the bottom of all the plates rested in. A good argument for using KOH as an electrolyte as it dissolves completely. However, with my cooling system (aluminium radiator) it won't be too clever!

Plan B: I stripped the whole cell and with my trusty angle grinder I lopped off about 10mm from the bottom of each of the 11 plates. I then decided that they could all do with a good clean with pine gel before being re-assembled.

Assembly of one of the sets of 3.

The 3 sets assembled, with the 2 additional neutrals also visible.

The re-assembled cell in the housing. The tubes for the circulation of the electrolyte in place.

Because the cell no longer rested on the bottom of the housing, I had to support it with a block cut out from a thick chopping board (I think the kind that butchers use, the household ones are certainly nowhere near as thick). The support is at the end where the connecting bar (+ terminal) reaches over the cell, i.e. the end which puts more strain on the side of the housing where the bolt goes through the wall.

End view. As you can see not all the plates are uniform in length, so I cut grooves in the block for only some of the plates to slot into. I only thought about recording this after it was all in the cell, but not 'n donder was I going to pull it all apart just to get a clear photo, sorry!

I started the car and switched on the pump and the electrolyser. I had the gas pipe in a 20 liter bucket of water ready to test production and it bubbled like crazy. I connected my multi meter and it was hauling 9,3A. Gas production was 1,3 liters per minute...for 14 minutes before the fuse popped! The fuse holder was too hot to hold.

I emptied one liter of electrolyte out and topped up with a litre of plain tap water. 8A and gas just over 750ml/minute. After running for about 20 minutes the fuse holder was quite warm, but not too hot, and just under 1 litre gas p/m @ 8,3A. The electrolyte was just warm, I guess about 35ºc.

Dankie tog!

I'll have a run tomorrow again and see how it goes.

But I still need to find out why a 30A fuse cooks at 9 - 10A! Nou ja...

Tonight's moonrise...time to call it quits for the day!

7 August 2009                                                                                                        

I opened the fuse holder up again and filed the soldered ends of the 2 incoming leads flat so that they would make the maximum contact with the ends of the fuse. I again used cable ties to pull the assemble together (now very necessary as the fuse holder partially broke last night!).

I only drove for a few k's this afternoon so there was no real test of the system. I decided to visit 2 auto electricians in the nearby industrial area to see if they had any fuse holders and fuses that would not overheat so readily as the others that I have tried.

At the first workshop he showed me a HEAVY DUTY FUSE  and equally HEAVY DUTY FUSE HOLDER. It was similar to the other spade-types that I previously tried, just very dik en stewig! I liked the terminal connections to the fuse holder - broad copper plates. R130,00 for one fuse holder and one fuse!

At the second workshop I was showed the same fuse as the previous guy had, except he did not have the fuse holder. His answer was 2 connectors which clip onto the "spades". I did not fancy this too much as the connection was not very tight or secure. BUT THEN he turned to me and said that I needed a trip switch. He showed me 40A unit and demonstrated how when it tripped out it would automatically reset itself. The connecting terminals looked to be robust so it looks as if this is my answer! I liked it because of the convenience (and cost saving) of not having to replace a fuse every time there's a problem! Cost? He looked at me for a few seconds before saying "R50,00". I am convinced that I can get it cheaper elsewhere. This I need to investigate!

When I got home after about a 10 minute return run, the fuse was only slightly warm so I think that something that I did yesterday worked, but definitely the tweak to the fuse holder would have contributed to this!

10 August 2009                                                                                                     

Today I had a golden opportunity to test my latest tweaks: it was my Son's birthday and he wanted us to take him to the skate park at Gateway, Umhlanga Rocks (just north of Durban).

First problem even before we left home: the electrolyser wouldn't run! The LED which is connected to the power switch did not come on. I checked the fuse and that was also okay. I checked every lead, connection, that there was electrolyte in the system, etc. I could find nothing amiss! Well, time was awenting and so eventually I decided to go and investigate on the other side. My wife and I dropped the children off at the skate park and we then went to spend the day with my parents. The investigation, with my father's help, continued. Eventually we determined that the relay had gone faulty, it was a Hella nogal! When I shopped for the relay I was told that this make was considered the best (@R35). Well, we found a spares place open (today is a public holiday) and I decided to go with the R20 "generic" relay.

After that was fitted I decided to sort out some of the wiring in that I changed the connections of the electrolyser switch: it now only powers the system when the car is running, and not just if the ignition switch is on (as it had been before). To do this I linked it to the fuel pump circuit.

Then I got clever. Remember I separated the electrolyte pump and cooling fan from the electrolyser cell circuit? Well, I thought that instead of having a separate on/off switch for these two items, I'll use the same switch as for the electrolyser. Ja boet... I switched the ignition on ... and nothing! No led light, no pump and no electrolyser! What now? Boy was my face red when I explained to my father (who had been standing there all the while with great expectations) that I had forgotten that it would now only run if the engine was running. Ahem. I turned the key further and it fired up. I switched the switch on again and ... smoke! I quickly switched off and remembered that the configuration of the pump and fan circuit were originally set up to be switched on the positive line, whereas the electrolyser circuit was different as a result of the relay being used. As a result the pump and fan wiring was down to a dead short with the earth and consequently it had a small meltdown. Luckily this did not affect any of the components!

Well, after rewiring the pump and fan to it's own, separate switch and circuit, it was action time and ... no led light! But when I checked, the pump, fan and electrolyser were operating! I then saw that the one leg had broken off the led. Phew!

I had driven exactly 176km without the benefit of the electrolyser. I then completed the round trip with the HHOO generator running. 352km later I calculated the results: The un-gassed portion of the trip I did 11,77km per litre. The gassed trip came to 14,6km per litre. Both legs were driven at more or less the same speed and with 6 of us in the car. This only amounted to a 24,5% increase in economy.

I was disappointed at the results as I had expected by now to be pushing past the 50% increase in economy mark. Immediately on our return home I checked the amp draw, the temperature of the electrolyte and finally (the big one) the amount of gas being produced.

First of all the fuse with which I had previously been having overheating problems with was now cool to the touch! I ascribe this to what I had previously thought to be a faulty earth, which I had rectified by connecting the electrolyser's negative wire directly to the chassis.

The amp draw was only 7,5A and then I remembered that I had on a previous occasion lowered the electrolyte concentration to help reduce stress on the fuse.

The temperature of the electrolyte was only 42ºc! This meant that the cooling system was working fantastically well. Considering that the last leg of the journey was a continuous run of one hour and the expected temperature for the day had been forecast for 27ºc (a nice winter's day), the new radiator and fan were obviously working very well.

Then the gas ... only 700ml per minute. This was to be expected considering the low amperage. Tomorrow I will increase the concentration of electrolyte so that the start-up amperage is in the region of 9A. If the fuse handles it and the cooling system can cope, then I'll consider adjusting it again for an even higher amperage draw.

At 9A I expect to produce gas at a rate of about 1,5 litres per minute or perhaps more.

11 August 2009                                                                                         

I increased the electrolyte concentration today and raised the start up amperage to 9A. Gas production now 1,2 litres per minute (cold start up). I went for a short 5km round trip and on my return the fuse had blown. I am now using 25A fuses as the stockists all seem to have run out of the 30A. In any event it is apparent that my fuse problem still persists as even a 25A fuse should not be popping at 9 - 10A.

I lowered the electrolyte concentration so that the system now draws 8,5A when warm. At this amperage the gas production is 1,1 litres per minute. I expect this output to climb to about 1,25lpm or more after it has run for about 20 minutes.

So, it looks like I will still be acquiring a trip switch to replace the fuse holder and fuse setup.

13 August 2009                                          

Today I acquired a 30A circuit breaker (until now I have been incorrectly calling it a trip switch!) which I have now fitted in the place of the fuse holder. I also replaced the broken LED, but I think I'll change this one for a red one which I am presently using on my bench top power supply, as the green LED is difficult to see in the day time. Tomorrow I will increase the concentration of the electrolyte and raise the start up amperage to 9A to see how the circuit breaker stands up to the test.

and...

a (not so) massive increase in gas production with a change in design and voltage!

I have been experimenting since yesterday with six 9cm x 19cm plates in this HHOO configuration.

I decided to apply low voltage to see if this would increase the efficiency of the cell. I used the red wires from the pc power supply to give me 5 volts. I had run out of bicarb so for this test I thought I'd use potassium hydroxide (KOH) as the electrolyte. I had to add a lot of it before there was some gas to speak of. I kept on adding KOH until the amperage reached 4A. By now the cell was chucking gas at a rate of knots! More than I've ever seen before with any other design! I tried the 9 plate design that I originally started with and even with an increase in electrolyte concentration I could not achieve anywhere near the output as with the 6 plates.

Others have also experimented with using lower voltages and have achieved similar or better results, but I regard this as a breakthrough for me personally.

Here are the results I recorded of the 3 tests that I ran with this (6 plate) cell so far:

Last night:

time                amperage    electrolyte temperature

Test 1

23h05             4A                cold (25 - 30ºc)        Start

23h30            4,2A              cold

23h40            4,2A              35ºc                          End

Test 2

00h15            4A                 cold (25 - 30ºc)       Start

00h55            4,1A              33ºc

01h30            4,2A              37ºc                          End 

Today:

Test 3

12h30            4A                   cold (25 - 30ºc)      Start

14h10            4,1A                35ºc

15h50            4,3A                43ºc

17h20            4,4A                45ºc                        End

Moerse nê?

These were recorded in my open test container. The cell configuration and the spacing between the sets of plates, as well as the lower voltage have obviously resulted in a very efficient cell. I also found in this test that almost no gunk was given off by the plates. I reason that the high concentration of electrolyte has resulted in better conductivity between the plates which places less stress on the metal, therefore the iron is less inclined to be drawn out into suspension than compared to when using a lower electrolyte concentration.

I will now fit this cell into my plastic "spaghetti" canister and run more tests, including gas production. In this container, which holds far less electrolyte than my test container, I expect the heat (and amperage) to build up more quickly. The downfall with using KOH is that I cannot use my radiator cooling system as the electrolyte will eat it away. I will also be impractical to use bicarb with this cell at 5v as so much more bicarb is required than KOH (in this setup I estimate at least 10x more!).

I am excited to see how many litres per minute the new setup will produce. I can control the heat (with reduced amperage) by lowering the concentration of electrolyte. Also, there is a container that I have in mind to acquire to use as a electrolyte reservoir. If I pump the electrolyte from the electrolyser through this container and back it will help to keep the temperature in check (a system which I have used with success before).

Here is the cell which I have now fixed in the container: it was a job to get the connecting straps right!

The completed article. The cable ties are there to hold a patch that I epoxied on the away facing side. On a previous occasion the container fell on the floor with a cell inside and it split the plastic.

I will leave the electrolyser until tomorrow so that the silicon can dry properly.

14 August 2009                                                                          

I upped the amperage in the electrolyser in my car to 9A... and the circuit breaker trips out after about 5 minutes! Ag nee... nou is ek eers moedeloos! The circuit breaker doesn't even get warm. I adjusted the electrolyte concentration so that the amp draw reduced to 8,3A and it still tripped after a while, with the circuit breaker still cool. I have now reduced the concentration further so that the amperage draw is only 7,6A and gas production is 800ml per minute. I need to seriously check everything to find out where the problem is. At this level of gas output I still expect the improvement in consumption to be significant (much greater than the 24,5% last recorded).

Update on my "new" design:

I ran the new cell in the sealed container and to my disappointment the gas production is not as wonderful as I expected. At 4A it was only producing just over 500ml per minute. While it was still in the open test container it gave off very strong smelling KOH fumes. I suspect that these vapours became trapped and condensed in the new container and it was then (mostly) only the HHOO exiting to the bubbler. The vigorous bubbling action that I observed in the open container was probably inclusive of the generation of the KOH vapours. Anghasi. I like the fact that this cell is producing no gunk, but I do not like working with KOH as it is not as "friendly" as bicarb. I will run further tests with this cell to see if it is worth the effort of taking it to the stage of running it in my car.

20 August 2009                                                                                      

The LED on my on/off switch on the dashboard popped a few days ago with me fiddling and resoldering the wires. As a result I decided today to test the system to see if the circuit breaker was holding out or whether it was still tripping. I diverted the gas line into my trusty 20 Litre water-filled bucket. Production was only 800ml per minute @ 8,2A and the temperature of the electrolyte was 37ºc. I left the system to run while I was checking everything and after about 10 minutes the circuit breaker tripped out. I lowered the concentration of the electrolyte to draw 7,7A. The gas output was now 400ml per minute which is certainly not good enough.

So... I decided to go back to basics. I took the electrolyser out and stripped it.

Where did all this gunk come from? The 2 fibre washers on the left wall of the container came from one of the connecting terminal bolts and slid to where you see them now. Then I looked at the plates!

The positive plates and the sides of the other plates facing the positive plates all had gunk on them. This was obviously the iron leaching out on the oxygen producing side. I have revised my opinion on using bicarb. It does in fact make gunk like KOH. But I have seen in the experiment with low voltage where I used a very high concentration of KOH that there was virtually no gunk in the electrolyte and nothing on the plates. Unfortunately this concentration is too high for my 12v (or rather 13,8v) car system without the right kind of electronic circuitry to control it.

I read on the net, and I have subsequently observed it for myself, that bicarb produces very little oxygen until all the carbon component is gone. Also... my greatest gains in mileage (50%) were obtained when I used KOH.  So ...

Back to basics!

Don't mess with the man's design! I decided to scrap my "clever" alteration to Bryan's concept and return to the 9 plate setup. I scrubbed everything first with pine gel before re-assembling the cell. I decided to go back to using the original 9mm x 19mm plates (the previous cell I had cut the plates shorter by about 1cm).

I want to try KOH again but then I can't use the copper or aluminium radiator to cool the system. I am going to try using a fridge water bottle as a reservoir as it looks like it might take the pressure. Why a fridge water bottle? Because a decent container that I have in mind is very expensive and this bottle happened to be in the pantry loitering with intent. I'll try attaching the cooling fan to this container to help keep the temperature down. It won't be anywhere near as effective as the radiator but it will be better than nothing.

I'll leave the silicone to dry overnight and then I'll assemble it all tomorrow.

28 August 2009                                                                                                 

Well, I ran the new setup and I found that on a short trip into town (round trip about 7km) the electrolyte would heat up too quickly. Also, the positive terminal was getting hot very quickly. After a lot of testing I realised that I would have to drop the concentration of electrolyte to a level where the system would be generating less gas than what I wanted if I wanted to keep temperatures under control.

So... I stripped everything down again, cleaned the cell up, re-installed the radiator and went back to bicarb electrolyte. I added cooling fins on to the terminals in an attempt to keep them cooler.

The fibreglass on the far side (negative) terminal is to insulate it from the ss heat shield which unfortunately touches the electrolyser at this point. 

The reconstruction of the radiator set-up.

Pieces of plastic cut out from a 2 litre milk bottle, stuck on with silicon, are handy for protecting the edges of the radiator.

My previous circuit breaker was taking a long time to reset itself when it tripped out. I was given this second hand one by a kind auto electrician, it can be manually reset via a little hole in the plate on top of the unit (which is now facing down).  I have found that this circuit breaker is less prone to tripping out than the previous one, perhaps because it is second hand (worn-out)?

After trying a number of concentrations of electrolyte I have settled on an amp draw of 8,5A to try and keep the positive terminal from heating up too much (and the circuit breaker from tipping). Production is only a shade over 750ml per minute. I'm not happy with so little as I think that between 1 and 1,5 litres per minute is required to move me up to my next target which is a 50% increase in fuel economy.

In a few weeks time we will be visiting my twin brother, my step children, grandchildren and friends in Jozi. This run will an informative one as far as fuel economy performance is concerned. We will probably trade our car in for a better one (condition, mileage on the clock, model, make? etc.) while we are up there. If so, I will either transfer the present system or if I can source 316L grade stainless steel up there at a reasonable price then I will build a new cell for the run home.

In the meantime I will carry on tweaking to improve the performance of this cell as much as possible.

I have decided to group my updates per month. So as soon as Spring kicks in (come on 1 September!) I will add another page with which to continue on. I will also put separate links on the home page for each month.

As my son would say to his friends...later dude.

29 August 2009

My town consumption is about 10km per litre - not so hot.

Bad news!!

Today I bought heavy duty leads for my multi meter as the leads which come as standard would quickly get hot every time I measured the amperage. I checked the amp draw with the new leads and I was surprised to see the meter indicating an amperage of 18,5A! I checked and re-checked and I still got the same reading! I checked the gas output and it was 700ml per minute. I then tried an analogue ammeter (the dash mount type) and it also indicated a reading in the 20A range. This ammeter is very old and not too accurate, but it was close enough to validate the reading of my multi meter. Or does it? As much as what the analogue ammeter also gave a high reading, I know that there is something up with my multi meter as it often emits strange beeps when I use it for testing voltage or amperage. 

What a blow! All this time I thought I was able to produce in excess of 1lpm @ under 10A! If the meter is giving a faulty reading then I will be very happy. But considering that the positive terminal was getting hot I think that this was in fact due to the amperage being close to 20A. This would also explain why the fuses and circuit breakers were popping and tripping. 

Another plan B: I decided to revert back to using KOH (again!) which meant that I would have to get rid of the radiator and re-install a reservoir so that there would at least be 3 + litres of electrolyte in the system to help keep things cool. I used the reservoir that experienced leaks the last time. But this time I removed the collar and screw-on end cap and used glue and silicon to fit a female end cap. While I was running it to check gas production and in between adjusting the electrolyte level, I could not find any sign of leaks in the reservoir. Hopefully it will hold out this time. I set the cell to produce 850ml per minute at which rate my meter briefly reads 19 point something amps before zeroing out. The meter is rated at 20A so this tells me that the amp draw must be pretty much near 20A. It also tells me that there must be something wrong with the meter if it trips out this close to its limit. I don't know, I'm not a fundi on these things, all I know is that there could be a problem which I need to check out.

The next step is to rip out the cell and check everything out. Then I will run it on the bench first to set the benchmark for amp draw and production. Ag ja... ek isielisie!

The family is getting together next weekend in Toti (the land of the Sharks!) so I need to make sure that everything is bak gat before we go.

Try, try, try again ... sometimes it is very trying!

BACK TO TOP

Free website hit counter

The information presented on this website is for you to look at, learn from, laugh at, or whatever. But if you try anything that you see here it is at your own risk. I will not take responsibility for your stupidity should something go wrong.