Dew Heater controller repair?

[erick]

I recently came by a Kendrick Dew Controller Model VI. Adverts say it is a pulse width modulation design from 40% to 100%. One control knob, four outlets. Control knob has an on/off switch built in.

When I put the output on the oscilloscope, loaded or unloaded, all I see is 0 volts out in the off position and 12volts out in any control position. "Houston, I think we have a problem here." Just before I drill out the rivets so I can get the lid off and have a look inside, is there anything I am missing here?

I have a home built PWM supply and I swapped that into place instead of the Kendrick and I saw the pulses, under control, as I expected.

Anyone know if I can get a circuit from anywhere? From Kendrick given this is a superseded model?

Thanks
Eric :sadeyes:

[kinetic]

Eric,

I have one suggestion.
If the control is a combination switch/potentiometer then maybe check
the central wiper voltage of the pot with a voltmeter.

In some wiring configurations of pots when the centre wiper goes
'open circuit' after continual use, the voltage across the pot
is as if the controller knob was at the 'top' end.
I hope this makes sense.

By putting one multimeter lead on the centre wiper (2 on the pic), the other on
one of the others(1), turning the knob should see a small
voltage change in a linear or logarithmic scale depending on if
it's an 'A' pot or 'B' pot respectively.

FWIW.

Steve

[Barrykgerdes]

Hi Eric
They are easy to repair. I just repaired one with the same symptoms. It turned out to be just the 555 timer but I also had a new driver transistor available. I did trace out the circuit but did not keep it . It was pretty straight forward. I cut the timer out to avoid pcb damage and installed a socket.


Barry

[erick]

Thanks Steve and Barry, I'll look into these suggestions tomorrow and report back. :thumbsup:

[erick]

Well, that was interesting. Took a bit of work to get the lid off - tamperproof screws!

Pot was working fine. I went round the circuit and found weird voltages. Slow rises in voltage - huh? Components looked OK. So, pulled the circuit board right out so I could get to the solder side. Started up by touching up all the solder joints. What's this? One leg of the output transistor was not soldered! Once done it started working, but not as I expected!

First link is what I expected to see - my homemade PWM controller - short pulse width. (All videos show circuits under load with an 8" DewNot heater strap.)

http://www.youtube.com/watch?v=1ViMA6thhI8

Here is what I get with the Kendrick:-

control at 0%
http://www.youtube.com/watch?v=bja2avB83Kc

control at 50%
http://www.youtube.com/watch?v=zuGuBq5AAiQ

control at 90%, then turned to 100%
http://www.youtube.com/watch?v=5Qg3AgaSf2s

Goodness - 4-6 sec pulse widths! I didn't expect that!

Well now it does work. Woo hoo!

[erick]

Update - it worked nicely last night on a 12V supply, then I transferred over to a 12V battery. I'm sure everything was right (polarity etc) but it didn't work, then all the smoke escaped! :eyepop:

Yet to pull the lid off, but I don't have a good feeling since all the smoke got clean away! It won't work without the smoke contained, surely. :D

Might rebuild it using a circult I understand better. I think the kit I have will fit the box.

[erick]

Opened it up. 555 is cooked - no other obvious damage? OK, I'll replace the 555.

[Starkler]

12v supply to 12v battery? Seems the 12v battery might really be something like 13.8v and that was too much. Doesnt sound right though.
Interesting with the switch rate. I would have thought a rate of say 1-10khz might be the go because you could then easily filter out transients.

[mithrandir]

Yep. Once the magic smoke escapes, things stop working.

[erick]

No, doesn't make sense unless some other component is damaged. It comes with a ciggie lighter plug, so is designed to plug into power from a car battery - 13.8V shouldn't be a problem and I'm fairly sure I have run it from a battery before (for those oscilloscope scope tests I expect).

I'll put in a socket and buy a few 555s, in case the next one goes up in the same cloud of smoke and I have to search deeper into the circuit.

[kinetic]

You probably will have to dig deeper Eric.
The 555 is only supplying the pulse timing and or pulse
width modulation for something else to handle the
larger switching current.
On it's own a 555 can only supply milliamps from it's
output, most configurations that's Pin 3.
I'd say the problem is associated with the device switching
the large currents.
It's a pity you can't show a circuit or maybe even take a close
up pic of the PCB both sides.
One of us might spot something obvious, albeit, hands off :)

Steve

[erick]

Haven't been able to get a circuit :sadeyes:

Photos in next post :thumbsup:

[erick]

I cannot see anything but a very unhappy 555 (U1) :confused:

[kinetic]

Great pics Eric,

I'll print them out and try and nut out that circuit.
PS: I looked through the thread again and saw Barry's post
about the 555. If Barry's saying it's a simple circuit, and
he's also saying it was the 555 on his, then I wouldn't
question Barry's call on it.
Did you put the 555 in the right way around?:shrug:

Edit: I can see by your pic that you have....:)

Steve

[mswhin63]

D3 looks like a Zener regulator very simple regulator and easily opens or shorts to cause problems. If opens then would feed the full supply voltage to the chip and BANG. I can't really see the joins underneath properly though.

Q1 is the high current switching which is fed from the main supply voltage and bias by the 555.

So need to make sure the right voltage zener is in place (If it is actually a zener).

[kinetic]

Eric,

just about got a circuit worked out from the pics, nearly there...
Good pics.:thumbsup:
Can you tell me what's written on the Q1 (transistor/FET)
and also on the side of VR1 , and finally the value of the
combination switch/potentiometer?

Steve

[kinetic]

Malcolm, not sure a zener would be across the rails (which it is) in reverse
bias without a limit resistor.
D3 looks like it's simply polarity protection

[kinetic]

Ok here is a rough circuit.
Hopefully no errors.
I'm not sure what the D1/D2 small signal diode setup does
on the 555.
I'd be checking the output Pin 3 of the 555 and turning the
potentiometer to see if the PWM or average voltage out
changes. A digitaml multimeter might not show this, I'm
not sure what the 555 is oscillating at. If it is up in the KHz
the reading probably won't make sense on a DMM.
An analogue multimeter might show variation, an oscilloscope,
definitely.

Edit: oops I forgot VR1...it's across the heater outputs exactly as
the C2 is shown. (varistor?)
Clearer circuit added with Varistor and LED

Steve

[mswhin63]

Looking at it again I am not sure it is good practice to use the same colour wires as far as I am concern.

I am having difficulty seeing the connection under the board, maybe a photo of the underside would be good but is look like D3 track goes under the rail 0V to R4/R3 V divider arrangement but it is really hard to see. There is a bit of black gunge blocking near D3 view of that area.

I can't imagine a 555 running without some sort of regulator as it may produce varying pulses. I also can't imagine putting a diode across the rail without a inline fuse to protect against reverse polarity.

Without a photo of the underside of the board I feel it is difficult to assume.

[mswhin63]

cct looks good, but I don't want to assume, I have concerns on best practices with the cct especially a diode across the rail without a inline fuse.

Have to go now back later for updates.

[kinetic]

I agree Malcolm but we've gotta start somewhere :)
The fuse could be in the cigarette lighter plug.
Over to Eric....:thumbsup:
Give us some numbers on components please Eric :)

Steve

[mithrandir]

If it helps make sense, I've straightened erick's photos, flipped the track side to match the components and overlaid then.

[mswhin63]

Good photo D3 is across the rail "BAD". Symptom blown 555. possible cause reversed at one stage blew D3 then blew 555. Still could be a zener for overvoltage protection fairly heafty diode (or Zener).

Check D3 (get the number) and replace anyway. replace 555 again if needed. Only one more component left to replace Q1. Fairly cheap items to replace, so would do them all.

Modifiaction - place inline fuse in lighter wire with god know current rating, need more testing for that.

Advice - I wouldn't buy one in the future, 2 design flaws already mentioned. (apologise for the bad)

[mswhin63]

Has been know to have a fuse, 2 things, would have blown the fuse before anything else in the past and the plug looks regular no fuse access I can see.

[erick]

Sorry guys, have been out all day. Great work however (ps. don't you guys have lives to live?)

You were confusing me re "photo of underside of the board" - photo # 3 is it?

I took them as close as my standard lens would allow me to focus.

I bought it secondhand. I can (and have) built my own based on the Jaycar kit (KC5225) and the one from Oatley Electronics ( http://secure.oatleyelectronics.com//index.php ), the K252 which is at a great price of $18 (cf Jaycar's at $25)! I also own the best - the Thousand Oaks unit - and it is being built into my SDM.

I bought it because cheap price (:D) and nice compact package - I end up with mine in big plastic boxes :sadeyes:

As you see below, I already found an unsoldered leg on the output transistor and the way it operated was nothing like I expected - but it worked after soldering.

I did some more checking today when I took the photos. Fuse is accessible in the ciggie plug - unscrew the centre contact end. It is not blown, but I didn't take a note of the amperage. I'll do that tomorrow - and get the details of the components that you are requesting. I checked (metred) the power leads into the board, via the switch on the pot - all fine. I looked for any lead damage or burning - none. I looked for any possibility of loose wire touching the circuit board - none.

Have a close look at the 555 - I won't use it again. There is a heat blister in the middle of its back and it has a crack running from the #1 pin end, down the middle to the blister. It won't have survived that.

I am absolutely sure that I could not have supplied the wrong polarity. I was working with two diferent 12V batteries with hard-wired ciggie sockets and two ciggie extension leads, both of which I have used before without problems. But I will have to check the leads out. Now I recall, I used one of the leads for the rest of the night with a different controller and one of the batteries - worked fine. No, I'm convinced that what went wrong was in this Kendrick circuit.

If you see any simple sensible circuit improvements, I'll make them - eg. put a better reverse polarity protection diode in place.

After I give you more details, happy to get down to lifting and testing some components. I'd like to pull the 555, insert an IC socket, then power up without the 555 fitted - and check power supply voltages to the socket. Don't want to blow another 555 - hell, that's $1.25 into the bin!

BTW, the 555 in the circuit is labelled 5BC9TYM C555CP. I'm hoping this is just Texas Instruments' way of labelling a garden-variety 555. Jaycar can sell me just one version of a 555 - an "NE555". I have tried googling with little success, but I expect this is a drop in replacement for this circuit?

Thanks again guys - I'll study the first version of the circuit diagram.

[erick]

Fuse in plug is 5A.
Q1 is marked RFZ34N R 533G amongst other random letters and marks.
D3 is IN581 (? maybe another number, but it's against the board)

I reckon that I lift and test D3. If OK, reuse, if not, replace.
Lift two legs of Q1 and check. If OK, reuse, if not replace (What would be the Jaycar replacement for Q1?)
Then remove the dead 555, replace with new 555 in socket. Power up and it will all work :D

[wasyoungonce]

Here ya go...a cct diagram..from what my beady eyes can see.

This is a standard PWM heater design.

Obviously the 555 is dead. This is usually caused by overvoltage. D3 looks like a zener lets say 16Volts Zener. D1/D2 are IN4148's or equivalent.

The FET may have shorted gate to source (or drain). Take it out & replace the 555 & power up. If Ok it may be the FET. A replacement FET would be IRF540N from Jaycar.

Hope this helps.

edit:
opps the LED is connected to R1...I just drew it as a diode

[erick]

Thank you kind sir!

I'll make a Jaycar visit later today and try to find time to fire up the soldering iron this evening.

[Louwai]

Do you remember a couple of yrs ago when smoke came pouring out of my kendrick ctrlr at SV Eric????

Darren fixed it.......... Did a top job...

[erick]

Didn't yours have + volts to the shield and - volts to the central plug on the RCA connectors?

I'm sure I checked this one and it's +ve to centre. I better check that circuit diagram again.

Edit - yes +Volts to centre of RCA.

[wasyoungonce]

err I should have said that if the FET was shorted to the drain then R4 would have smoked up and more than likely L1 as well.

Suspect it's just a bad 555 then.

I didn't see the 1st page drawings...slaps forehead..Do'H!

[kinetic]

Woohoo, so some progress. Ahhh it WAS a FET...and Malcolm was
right too, it was a protection Zener....apologies Mal:thumbsup:

[mswhin63]

No need to apologise it is entirely possible for D3 to be a rectifier diode for reverse protection as they can handle higher currents. The important thing is that Eric is one step closer to getting it fixed. Zener would need to be 15V for 12V system Although 16V might be better as long as the rest of the cct can handle that sort of ver voltage. Car alternator can go as high as 14.5V so .5V is not much of a cover. I can only assume it is primarily design for general battery operation so a Zener under 23V should be fine. 1N581 is not all the numbers so it is going to be difficult to know the exact replacment zener value is.

I won't be so quick to judge the no fuse in the design. It would be pretty hard to forget it in this design. I retract my advice on not gettting one if needed.

[Louwai]

Not being overly electronic orientated Eric, I don't really know what the issue was.
But I do recall the smoke was caused by the power to the unit being turned on before all ofthe heater straps were connected & a lose strap power feed cable came in contact with the tripod......... which has the controler attached to one leg......

So a direct short I assume....

[erick]

OK, I'm back in business! Thanks all.

Lifted a leg of D3 - it checked out OK. Forgot to read the full number before I resoldered it :rolleyes:

Tried to lift Q1 but my crappy solder sucker wasn't doing a good job, so I gave up.

So I removed the 555. It came out in several pieces - it was severely cooked.

Added socket and a new 555. Very cautiously turned on - all fine. Added some load - all fine. No components heating up. Up to maximum setting - no problem. Put the oscilloscope onto the output to the load - it was behaving just as I described before. Loooooooooooooooong pulse width - seconds! At minimum setting output 12V is on for some 40% of the pulse. At maximum setting output 12V is on some 95+% of the pulse.

Ran for a while and all seemed fine. No components got even warm to touch. (the 3" dew strap was only drawing about 0.5A @ 12V).

I think it is on too long at the minimum setting. Can anyone suggest, from the circuit, which resister value I need to tweak to have it on for less of the pulse width at minimum setting - perhaps down to output on for around 15-20% of the pulse width?

Thanks again.

[DavidU]

Erik, running Astable operation I would replace R2 with a trim pot of twice the value of R2. The trim pot will use only one end lead and center.
The pot value and R2 change the width state when run as a multivibrator.

[erick]

OK, R2 it is. I'll get onto that.

Now what is R2 - cannot work it out from the photo - grey/green orange/brown orange? I'll have to pull the lid off again. But it's definitely orange - so maybe a 50 or 100k trimpot should do the trick.

If the pot at minimum gives me about 40% on-time, then reducing R2 value should reduce that below 40%. OK, I can play with that easily and check the results on the scope.

[wasyoungonce]

Eric I have played extensively with 555s PWM. The mark space ratio (range of control) is set-up by the Pot and (capacitor) Cx value. F = 1.44/(Rx * Cx).

I found if you have these wrong the PWM starts at around a duty cycle (mark space ratio) of around 40%...this is pretty much what you are seeing. Thus you have less range of control.

By changing the Rx & Cx (Pot & Capacitor) you can get this down but you also effect the PWM freq.

Since your Pot is a switch/pot you may have troubles finding a higher value pot that suits. In any case I found a pot of around 500K (& up) and a Cx value of around 1 to .3 microfarads gives a freq of oscillation of 2 to 7hz and the mark space ratio is 4.25% to 99.9% (range of control).

I can send you a spreadsheet I did on pot/Cx PWM freq & mark space ratios if you want.

[DavidU]

Nice work Wasyoungonce !

[erick]

PM sent for spreadsheet, many thanks.

Don't need a switch so changing the pot is no problem.

I wouldn't mind if the PWM frequency was increased - many seconds seems odd to me compared to speed controller circuits which I have used otherwise. But I guess it is not an issue for this heating application. But 2-7Hz - easier to monitor/adjust on the oscilloscope. :thumbsup:

Eric :)

[asimov]

I have one of these brand new that I'm about to put up forsale Eric..Sounds like yours is fixed though?

[erick]

Yep working now. I will tweak it a little to give a greater range of control. I have uses for it. I won't be buying another of these. Would prefer more control of individual outlets.

[asimov]

:thumbsup: