[R100ELL]

My 7 volt regulator was a bit melted so thought I would have a bash at making one. Got the regulator from Farnell & the gel filled enclosure (came in pack of 4) from CEF, picture below shows the old & the new.

Soldering done

Fitted into a small gel filled box

and finally the video of it working

 

[Sidecar Bob]

This is the recommended circuit that Reg in Bristol developed many years ago. In those days 7805 (5V) regulators were just becoming common and 7807 (7V) ones were expensive and next to impossible to find so he added the green LED (junction voltage approx. 2V) to raise the voltage relative to ground and produce the required 7V). 7805s are still cheaper but 7807s are more reasonable now but some still prefer to use the 5V reg so they have the LED to indicate that it is working (or at least connected to power). The capacitor is recommended to control spikes caused by load changes and to prevent oscillation within the regulator (it could live longer with this) but the most important part is the heatsink. I don't know if the gel in your enclosure will dissipate the heat but if it won't would recommend either bolting the tab of the regulator to a couple of square inches of sheet metal or potting it in black epoxy.

 

[R100ELL]

This is the recommended circuit that Reg in Bristol developed many years ago. In those days 7805 (5V) regulators were just becoming common and 7807 (7V) ones were expensive and next to impossible to find so he added the green LED (junction voltage approx. 2V) to raise the voltage relative to ground and produce the required 7V). 7805s are still cheaper but 7807s are more reasonable now but some still prefer to use the 5V reg so they have the LED to indicate that it is working (or at least connected to power). The capacitor is recommended to control spikes caused by load changes and to prevent oscillation within the regulator (it could live longer with this) but the most important part is the heatsink. I don't know if the gel in your enclosure will dissipate the heat but if it won't would recommend either bolting the tab of the regulator to a couple of square inches of sheet metal or potting it in black epoxy.
View attachment 205675

Thanks for the information Sidecar Bob, I did read the operating temperature of the regulator Im using is -40c to +80c. Given the size of it I think I would struggle to get any sort of heat sink onto it. The gel in the enclosure, as well as insulating & IP protection, should hopefully offer some sort of heat dissipation. I will be keeping an eye on it once I get the engine running long enough to check the temperature gauge works.

I found out the right hand HT lead is faulty so have ordered a couple of ignition coils, I feel it’s best to replace both whilst I’m there and as I had the fuel tank off I cleaned the terminal of the temperature sender unit & replaced the crimp terminal attached to it as I’m hoping that was an/the only other issue. Time will tell.........

 

[Sidecar Bob]

You would be surprised how much power that regulator would have to dissipate. 140 mA of current through the meter circuit would would develop over a watt at the regulator.....

 

[murrayf]

now you just have to fix that broken guide that is hammering around

 

[R100ELL]

You would be surprised how much power that regulator would have to dissipate. 140 mA of current through the meter circuit would would develop over a watt at the regulator.....

Out of curiosity how much power is expected to be drawn by the temperature gauge. As mines hasn't been working & the fact I don't have a clamp meter means I don't know what to expect on that front

 

[murrayf]

it will draw up to one amp

 

[Sidecar Bob]

So the regulator would need to dissipate up to 7W. I'd say you need something like 2-3 sq. in of plain sheet metal for that...

 

[Reg in Bristol]

7807s were rare as hens teeth back then so I hacked a 7805 with a green led in series with the ground
pin to bump it up close to 7V
It was what I had to hand
Nowadays I'm using an LM317T and two resistors (2200 and 470) for the temp gauge in my GL clocks
which gives around 7.1V
A heat sink is needed, a to220 type works for me, similar to this

I left one off one gauge out of curiosity and it appeared to work fine for 10 minutes then the gauge shut down
and only a power cycle would get it to work again.
The ICs have self protection which shuts them down if the internal temp passes a certain limit
Even if they shut down and cool, they stay off until the power is cycled, well mine did anyway.

 

[R100ELL]

Fired it up tonight once I replaced one of the HT leads which kept arcing and for the first time I saw the temperature gauge working, took about 5 minutes or so to register. For now I will have to wait for better weather before I can take it out for a run And a shake down to make sure all is ok. I will check the gauge then and look at getting a heatsink fitted if the gel doesn’t dissipate enough heat. It’s good to know it will just shut down if it overheats but I did buy a spare just in case I stuffed it up.

 

[Reg in Bristol]

I have dim memeory of a stock gauge on its own having a resistance of around 50 ohms
and with the sender going from circa 500 ohms cold to 20 ohms hot
with a running supply of 14v and 7V out you'll have current ranging from
25mA cold to 200mA hot through the temp gauge circuit.
that's a max wattage around 2.8 maybe 3 Watts.
Doesnt sound a lot and not all of that has to be dissipated by the linear regulator
probably only around half of it I'd guess, but its still enough to overheat an unprotected IC

 

[brinkcx]

Just a minor thing but long, long ago some engineering teachers insisted I looked at failure modes for any of my silly designs. I had ideas about a 7807 regulator as well and as Reg already said you'd have to expect 200+ mAmp max, that boils down to dissipating about 2 Watts. These nice 78xx regulators have a thermal self-protect, excellent, no problem?

Yes, there is a problem. It may work very well on any normal boring day. If your engine runs hot on a hot day and your 7807 or any other chip has no heat sink, the 7V regulator is protected, very nice. But not your engine ...... what the meter will show is an uncertain lower temp because the regulator specs don't say what current will flow in thermal protection mode, just that the regulator will survive..

Take home message : heatsink really needed !

( Guess that's why many if not all originals fail in the long run .....)