[Reg in Bristol]

Just as I finally seem to have fixed the head geaket problem the

bloody temp gauge packed in

I had potted it in hot melt glue and destroyed it whilst trying to dig it out

and had no spares to build another

Bollaux!

So, undismayed I decided to try an idea I've have had for ages and

make an led array which moves in relation to the senders lowering resistance

I just fitted it and powered it up

It didnt burst into flames or belch smoke which I always find encouraging.



I did breadboard it first and was reasonably confident of it but

now its ready for some real world testing.

IF it proves to be useful it occurs to me I could OR the higher temp led signals

to control an electric fan, but first things first.

I used an old indicator case as it was to hand.

 

[Linny]

Clever. I hope it does what you want it to do. I enjoy the bouquet flavor magic tree! I hope you don't get a ticket for obstructed vision. It would be a shame to be forced to lose your ability to smell the roses as you ride!

 

[Toehead]

looks great. Be sure to keep us updated.....



What is that other LED gadget that you have on your bike?

 

[unbreakable7132]

When I removed the gauges on my cx I lost the temp gauge.I made a clear plastic shell for a small bulb and hid it under my carbs.I hooked it up cold and checked how bright it was, then hooked it up hot to check brightness again. I actually used it as my gauge for a while, now I'm trying to hook up a 2 inch digital gauge.It was a neat idea to me for the time I didn't have the gauge.

 

[Reg in Bristol]

I like that Brandon

you couldnt get much simpler could you?

The bulb gets brighter as the engine warms up.



I was going to make a graph of the temp to resistance ratio

but my digital temp probe has gone walkies so I've had to

use dead reckoning.



My sender has a resistance of 600 ohms at 20C

which drops in a non linear fashion to about 50 ohms at about 100C

Using a hot air gun, I had it go as low as 37 Ohms

Anything above 100 ohms can be regarded as cold in terms of engine temp

and disregarded so I used these figures as a rough guide.



Well after a brief test it works.

the leds lit in turn as the temp rose but using the old indicator body didnt help.

Putting yellow, amber and red leds behind an orange diffuser made it hard to discern which led was lit.

It did occur to me beforehand but I was impatient to test it and grabbed

the first thing to hand.



It needs tweaking and some refinements doing to it but afterwards I think

I may have a useful 'at a glance' indicator of coolant temperature

and not be reliant on 7V regulators or the old gauge.

It uses less then 100 milliamps when running by the way.



I know I could go much further and there are many ways to get very accurate

temperature readings but I wanted a simple robust and reliable indicator

thats only as complex as it needs to be.

Component cost about £3



Toehead: thats my battery charging indicator

2 green leds tells me its 'in the zone'

any more its over charging, any less, under charging

Those Suzuki clocks had a gear indicator which I ripped out and put that in instead

 

[Toehead]

Very cool. I was thinking about trying something like that out. Do you have any schematics?

 

[Reg in Bristol]

Heres the battery thingy in action

[media]http://www.youtube.com/watch?v=5EuL3m9pSJ4[/media]

Just watched it again, the bike sounds rough as hell ! but its only

some clutch rattle and in reality its not as bad as the rather sensitive

video cam mike makes it sound.

The battery voltage monitor is useful but I got caught out a while back thinking

all was well when the battery wasnt holding a charge and that reminded me

that its always wise to consider the limitations of any devices we use.



the circuit is dead easy and based on an old LM3914 one thats been about for years

I probably have a diagram somewhere

If you mail me on

[email protected]

I'll send it to you when I find it and correct the mistake I made when I drew it

 

[jasonh]

Hmm, interesting project. Any details on how the device was put together?

 

[Reg in Bristol]

Ok prototype 2

I used a cheapie headband torch case

and will run with this for while and see how it performs.



From brief testing on and off the bike,

the green comes on with ignition, this tells me its powered

As the bike warms up, the 2 yellows glow in turn until one of the 2 top red leds light

at normal/fully warmed up running temps.

If the larger/brighter leds light, this indicates overheating

I doubt the last large led at 5 o clock will ever light unless I torch it for the

insurance money.



3mm leds seemed a better idea

I wanted a clear but at a glance indicator, not something lit up like a christmas tree

that would be distracting at night.

 

[Toehead]

I like your creativity

 

[Shep]

Could you make one of these work Reg?



http://cgi.ebay.co.uk/52-mm-OIL-TEM...arts_Vehicles_CarParts_SM&hash=item23096aabfb

 

[Reg in Bristol]

Ooh pretty!

Cheap too.

Do you mean as a coolant temp gauge instead Shep?

I dont see why not.

It would work as is, but not sure off hand how oil temp relates to coolant temp

Does it run higher or lower?

Either way, you could jigger with the sender to compensate

 

[Shep]

Ooh pretty!

Cheap too.

Do you mean as a coolant temp gauge instead Shep?

I dont see why not.

It would work as is, but not sure off hand how oil temp relates to coolant temp

Does it run higher or lower?

Either way, you could jigger with the sender to compensate

I've got a feeling they can be used as either and the temp ranges of the oil are close to the coolant anyway Reg.As you know they are simpatico e.g if for instance my Efan was on all the time it lowers the oil temp a little below the nominal mark but not enough to effect the efficiency of the engine.



If you look at these more expensive oil temp gauges that go in the Oil Filler hole,



http://www.getgeared.co.uk/MOTO-DETAIL_Motorcycle_Oil_Temperature_Gauge?step=2&brandid=6&brand=HONDA



they pretty much match the coolant.My cylinders at idle are around 82 Deg C



you would soon know the range anyway by experience.



Just noticed the video.The gauge has a volt meter as well

http://www.youtube.com/watch?v=-jr-q48EQ3o

 

[marshallf3]

What I'd like is a graph of the actual voltage across the sensor at various temperrature levels, in particular at ambient and at several points between 180*F & 210*F



Resistances would be OK but I'd need to know the resistance of the gauge itself.



Wish I had the bike here, it would be pretty simple to figure out. Even though I've got the variable fan speed drive circuit designed for a 100K "tiny bead" thermistor it would be a heck of a lot easier to just tap off the exiting temp sender wire (at it, at the gauge or somewhere inbetween) and use that to control the PWM circuit.



Seems to me that somewhere in the old forum was a chart and description from Honda of how to test the sensor using an ohmmeter and a pan of water but I can't recall if it gave "in circuit" voltages as well.



Couple of people have already expressed interest in a true PWM variable speed controller that included a fault light. I've got a circuit that would not only accurately vary the fan speed but also light the light (LED) if the fan quit turning or if the bike got too hot.



If I had these voltage readings it would be easy to finish up the design and start testing a "plug and play" circuit this weekend. As it stands no you'd have to use thermally conductive epoxy to glue the thermistor to the radiator or ??

 

[Reg in Bristol]

My sensor was 600 ohms at room temp. 20C

and dropped in a non linear way to about 50 ohms in pan of just boiled water

I'd measured the senders a few times on a normal running bikes and

knew that 50ish was the norm when the guage showed at the junction of the

thin/thick lines.

I started a graph but couldnt find my digital thermometer so

I used these rough guides to set up my gauge and 50 ohms lights the first small

red led.

Before the gauge packed in the bike was running AOK temp wise, so I'm reasonably confident in it



I took it for a run earlier and the first red led lit pretty much all the time

after it warmed up and the second red only came on at times when I was on the motorway at 70 ish.

Not sure if the remaining 4 will be ever lit unless a serious cooling fault occurs

but will leave it be for now and see how it behaves for a week or so before doing any more tweaking.



The main thing is I have an indication of engine temp again and that was

the motivation for this project.

 

[marshallf3]

My sensor was 600 ohms at room temp. 20C

and dropped in a non linear way to about 50 ohms in pan of just boiled water

I'd measured the senders a few times on a normal running bikes and

knew that 50ish was the norm when the guage showed at the junction of the

thin/thick lines.

The thing is I'm wanting to leave the guage in the circuit, after all if you've got one you'd want to still be able to look at it.



I found some additional data in the service manual as to the sensor which may help me to estimate the ohmmage of the actual gauge but it isn't going to be the same as having a few true voltage readings off of someone's bike. Even knowing that voltage at a few temperature points (outside temp before starting, voltage when the gauge reads halfways etc) could be quite helpful.



Thermostat:

Starts to open at 80*C-84*C (176*F-183*F)

Fully open at 93*C-97*C (199*F-205*F)



Sensor ohmmage readings:

60*C = 104 ohms

85*C = 43.9 ohms

110*C = 20.3 ohms

120*C = 16.1 ohms



It's just the voltage at TP1 I'm interested in: http://www.innoengr.com/CX500/stock_sensor.jpg

 

[nolavox]

Hi Y'all



Great DIY idea Reg!



Ive been using a multimeter with a temp sensor as the tool for this type of diagnosic. This will allow you to corrrelate Temps, resistance and voltages exactly.



A link http://www.markertek.com/Tools-Test-Equipment/Multimeters/Rolls-Corporation/MU118.xhtml

 

[Shep]

Just noticed this one from the same seller.Costs a bit more bit very flash,



http://cgi.ebay.co.uk/52-mm-OIL-TEM...der-/eBayISAPI.dll?ViewItem&item=150484063923



You can quite easily drill and tap out the temp sensor like Steve F has done to use cheaper car temp sensors and the both of the above come with the temp fittings etc

http://www.dotheton.com/forum/index.php?topic=10072.120

 

[marshallf3]

Hi Y'all



Great DIY idea Reg!



Ive been using a multimeter with a temp sensor as the tool for this type of diagnosic. This will allow you to corrrelate Temps, resistance and voltages exactly.



A link http://www.markertek.com/Tools-Test-Equipment/Multimeters/Rolls-Corporation/MU118.xhtml

Thanks, but I've got at least 10 different multimeters. The bike is out of town getting the last of the restoration done and I'm trying to get this circuit together before the electric fan is mounted so obviously I have nothing to get the readings from. As it stands it would work fine if I want to use a separate pickup but others that already have electric fans have expressed interest so being able to contreol it with the actual sensor voltage would eliminate having to take the radiator off to mount the tiny sensor, instead getting its input from the circuit that's already in the bike. Voltages from that circuit are going to require a change in design from what I already have working that's controlled by a separate thermistor.

 

[nolavox]

Thanks, but I've got at least 10 different multimeters. The bike is out of town getting the last of the restoration done and I'm trying to get this circuit together before the electric fan is mounted so obviously I have nothing to get the readings from. As it stands it would work fine if I want to use a separate pickup but others that already have electric fans have expressed interest so being able to contreol it with the actual sensor voltage would eliminate having to take the radiator off to mount the tiny sensor, instead getting its input from the circuit that's already in the bike. Voltages from that circuit are going to require a change in design from what I already have working that's controlled by a separate thermistor.

LOL only ten... Are they the Sanwa and Kowa testers Honda shop mannys spec? Are they true RMS meters. I believe one meter in the hand is worth....



I think you may have misunderstood what I intended. The Multimeter I referred to can be used as a thermometer / as well as a DVVM, Mine is an Ebtech and uses a thermistor probe to read temp, not the regular test probes. I have used mine to simply see at what temp the exhaust gasses are at the pipe exit, or taped the probe onto the water pipe to get a rough idea of where the themostat is opening and if it's consistent. I also have used it to see what temp the inside the Fridge or coming out of the Auto AC vents is. It's just an alternative to a conventional thermometer or sensor gun and an easy way to confirm what actual Temps are in a variety of places.



Many temp sensing probes availiable from Fluke fluke temp sensors & probe accs

Here's a solution to convert one of your dvms to a thermometer.

Fluke Thermocouple







Anyway, It sounds like you are way beyond that and I apologize if I offended you in any way.