K8004 DC Pulswidthmodulator - nasty delay in setting in

Kits Automotive projects
1

Hello All,

I have bought and soldered neatly three kits K8004, the DC Pulswidthmodulator. (Specs: 6 -30V DC, 6 Ampere.)
I used one on a fan, I used one on 5x20W 24V/3A halogen lightbulbs, and the other on 4x20W 24V halogen light bulbs.
The ones on the lights seem ill or acting strange, giving problems.

The situation on board is: Boat batterys 24v DC, continous charged at 27v DC.
All DC aparatus as navigational equipment, fans and other motors work as they should.

Problem:
They control the light with the external trim-pot as they should, also the tuning trim-pots on the print.
But why the heck do they take so long to kick in (feeding the users) when switching the light switch on?!
The delay is about 13-17 seconds, everytime over and over again.
I dont want that to happen at all, it is inconvenient and annoying, waiting for the light to come on.

In a test situation they work fine. Once in the same situation as they are mounted now, I lengthened one wire with 30metres solid core 1,5mm2, and it worked. But I don’t think thats a proper solution.

Any one any idea how this effect is created, the delay I mean? (supposing it is not the normal working order)
Any one suggestions for a solution to minimize the occuring delay?

All information welcomed,

kind regards Walter

2

When the bulbs are cold, they represent a very low resistance, which causes the protection circuit of the kit to kick in, as the initial current swings beyond the limit of the kit. Once they have heated a little bit, the resistance rises significantly and the current stays within the limits of the kit. You can try to solve this problem by lowering the value of the current sense resistors. Sorry, no values, please experiment.

3

Thanks for the explanation. A little surprised that the halogen would bring a problematic peak for its rating.

Fine, I can experiment at own risque.
But please what resistors are considered the sense resistors, from the diagram r #?

4

R8, R9 and R10 are sense resistors.

5

Havent made modfications or anything yet.

I tried once more if settings of the trim-pots make a difference. Couldnt see any effect, only the frequency (250k trimpot) gave me a few seconds gain.
Then it collapsed; none of the 4 trim-ppost have any effect.
The only thing it is doing is passing on the voltage to the light, at almost maximum output.
Goodbye dimmer.

Dissapointing conclusion: If you ask me the K8004 kits just cant endure 100W Halogen.