Maker Pro
Maker Pro

repair Tribest Sedona SD-P9000 dehydrator -- low DC voltage from power board: maybe rectifiers or caps?

.brush

Oct 5, 2022
12
Joined
Oct 5, 2022
Messages
12
hi folks, really appreciate finding this forum and look forward to learning more from you all!

i have an oldish (probably 2011) dehydrator that is digitally controlled, so without the control panel nothing works. worked fine for years, then suddenly turning it on does nothing: no lights on the control panel. not even a hum, which i associate with power transformers, so i assumed it was a switch issue -- and indeed there may have been. but having worked with the switch, it is now correctly providing 120V to the power transformer. in turn, the transformer is providing about 11.5 VAC off the secondary wiring... which is higher than the nominal 9V printed on the power PCB (kind of blocked in images below).

however, only ~0.5VDC is getting to the control panel (and this seems to fluctuate from 0.3 to 0.6V). i have a question in to the service department for Tribest whether that is nominal, but i'm fairly sure it should be a normal 5V based on the fact it uses two HC574 ICs (datasheet here i think: https://www.alldatasheet.com/view.jsp?Searchword=HC574&sField=3).

this would explain why the control panel shows nothing. there are no fuses or thermal cutoffs that i have found yet, which seems strange... unless the reed switch is supposed to function as one. (why is there a reed switch on the control panel? anyway, shorting the reed switch does not result in powering the control panel.)

so my presumption is that something in bridge rectifier or caps from the 9VAC is fried?

there are a sequence of 9 diodes coming out from the 9VAC (i would expect 4 or 5) -- almost as if they ran them in parallel, which seems like a no-no? or perhaps it's two separate rectifiers at different voltages?

there's nothing obviously wrong with the components. but the coloring of the PCB on the back of this area is faintly yellow/brown, which adds to my suspicion.

but i'm not really sure how to test for what's blown? not even entirely sure where to probe to test for voltages. any help at this point very helpful.

can't find schematics online, and the service department "doesn't have them available" -- not sure how helpful that department will be, they told me they'd have to call me back (who knows?) to tell me whether 0.5V is too low for the control panel.

thanks!

(i obviously wish this wasn't a digital control panel, but just had basic components you could easily test and fix, such as thermostats and thermal fuses, etc. if this troubleshooting doesn't work, i may want to figure out a way to sidestep the control panel and put in analog controls of such a sort into the machine... anyone have experience with something like this? it seems the heaters take line voltage, but i think the fans take something else, but i can't trace the PCB clearly... suggestions how to figure that out?)
 

Attachments

  • control.sm.jpg
    control.sm.jpg
    216.3 KB · Views: 11
  • PCB.back.jpg
    PCB.back.jpg
    220.2 KB · Views: 11
  • power.sm.jpg
    power.sm.jpg
    220.2 KB · Views: 11
  • power.trans.sm.jpg
    power.trans.sm.jpg
    227.5 KB · Views: 10

73's de Edd

Aug 21, 2015
3,462
Joined
Aug 21, 2015
Messages
3,462
Thir .brush . . . . . . . . .
( Thats stee-range ? . . . . . I seem to have just typed in a lisp . . .th. )


Looking at the boards foil side, one will initially see three hot spots from heat level subjecture and accumulated time of use.
They relate to the two rectifier diode sets and the two heat sinked triacs that switch heating functions in and out.
Seems like one triac area is being more cooked than the other, so maybe one gets principal use, while the other is intermittently switched in for supplementing heating level..
Those areas are RED boxed with the YELLOW boxing being their optical drivers. . . . . currently . . . .of minimal interest to us . . . . 'til we initially find the presence of DC supply voltages.

MARKED UP REFERENCING . . . . . ( two lef' clickee-clickees . . . makes much mo' bigga ) '
DEHYDRATOR DATA.png

Co-hosted @
TROUBLESHOOTING . . .
One would expect the raw 9VAC coming in and initially getting routed to the diode sets for rectification, with the resultant DC flowing over to 3 sets of RED starred filter capacitors .
One should expect having a decent DC voltage level across them, when powered up.. Sound KLAXXON HORN . . . .that's not happening now !
Plus there seems to be the utilization of a heat sinked 3 terminal regulator U5 ( LM 7805 ? ) which will be outputting a reduced voltage,
Viz 5 VDC.
You should have one voltage level supply feeding the white central connector to your control and display panel.
And U1 seems to be a comparator associated with temperature control regulation . . .fill us in on its part number !
The . . .ULN 2003 is using 2 of its sections for switching the two cooling fans. Two more sections go to the two unmounted connectors beside the cooling fan ones. The balance feed into your white cable, going to control panel . . . . as pins 4-5-6 and 7 th down from end pin at center of board.
That connectors next two pins connect to U1 temp controller I.c. 's two pins and the last pin is ground.
The temp sensors feed into 2 of the U1's pins.
The 2 and 3rd pins down from top center are feeding the WHITE "Zero crossing" optical isolator drivers for your power control TRIACS.
The very top pin of the cable is power for the control panel and should have the same voltage as
the 2 star E- cap.

Initially you found some . . . or all . . . . of the of the 3 RED starred E- capacitors have / had only ~0.5 VDC present.
With you reading such a somewhat higher 9 AC voltage, suggests its not being subjected to loading.

" Reading" the foils suggests that the rectified DC is routed thru a series steering diode to the
3 star E-cap and the input of the 3 term regulator. That reg outputs into the 2 star cap closest to ULN2003.
To the side of that cap is an unmounted LED indicator with its series dropping res even further over.
Should you "LED" that hole with proper polarity LED , it would give you a power on indicator.
Our 1 star cap just seems to be supplying a heavier filtered supply to our control board electronics, thru its cabling.
Supply the 3 caps values and voltage ratings. Also U1's I.D. . . . .U4 says me too . . .ME TOO . . .ME TOO !

Standing by . . . . . . . .
.
EatingPopcorn.jpg


73's de Edd . . . . .


Ya' know what ? I don't quite have your solution yet, but I shore do admire your problem .!




.
 
Last edited:

kellys_eye

Jun 25, 2010
5,282
Joined
Jun 25, 2010
Messages
5,282
Commonly, those dehumidifiers tend to stop working when one of the off-board sensors detects a fault - in my experience I've seen condensation collecting around a temperature sensor that caused the wires to that sensor to corrode away and stopping the operation completely. I was 'fooled' into believing it was a PSU issue too!

The sensor in question was located at the lower part of the condenser module (dry) and was ultimately 'killed' by a blocked overflow.
 

.brush

Oct 5, 2022
12
Joined
Oct 5, 2022
Messages
12
these are great replies! thanks so much. i'll go get photos of the sensors, and part number/ratings for the ICs and caps. quick questions: 1) how would i test the sensors? do you remember, @kellys_eye, whether they should have continuity or...? and 2) should i test voltages on the power board? ie. by checking from one side of cap to ground? or from final diode (which is final?) to ground? or...?
 

.brush

Oct 5, 2022
12
Joined
Oct 5, 2022
Messages
12
update: the 3 capacitors are normal 470 microF, 25V.

U4 is, as you noted, labeled ULN 2003 APG
U1 is (CHN) / LM 358N(?) / (KFM102)
U5 is, also as you guessed, (KIA) / 7805A / (PI 116)

i'm not sure how the sensors work, but they have high resistance between the two wires on each one: 13.3 k-ohms on one, 12.7 k-ohms on the other. pics below. no obvious damage.
 

Attachments

  • sensors1.jpg
    sensors1.jpg
    199.5 KB · Views: 5
  • sensors2.jpg
    sensors2.jpg
    187.3 KB · Views: 5

kellys_eye

Jun 25, 2010
5,282
Joined
Jun 25, 2010
Messages
5,282
That's a picture of the thermal switch with inline thermal fuse - not the kind of sensor I'm referring to though. Look for a pair of wires disappearing into some heatshrink. Inside the heatsink you'll find a string of diodes used as a temperature sensor (assuming your dehumidifier is based on a generalise design) - this chain is often broken from condensation corrosion.
 

.brush

Oct 5, 2022
12
Joined
Oct 5, 2022
Messages
12
That's a picture of the thermal switch with inline thermal fuse - not the kind of sensor I'm referring to though. Look for a pair of wires disappearing into some heatshrink. Inside the heatsink you'll find a string of diodes used as a temperature sensor (assuming your dehumidifier is based on a generalise design) - this chain is often broken from condensation corrosion.
hmmm, i can't find anything that sounds exactly like that. is it possible the green thing in the left of the images, which is also on one of the sensor legs i believe, is the temp sensor you're referring to?

one more comment -- to be clear, this is a dehydrator for drying fruit and vegetables, not a dehumidifer for removing water from the air in a room. does that make a difference? ie. this is not storing water, it's just blowing hot air over the food, so there's not much chance of condensation...?
 

.brush

Oct 5, 2022
12
Joined
Oct 5, 2022
Messages
12
TROUBLESHOOTING . . .
One would expect the raw 9VAC coming in and initially getting routed to the diode sets for rectification, with the resultant DC flowing over to 3 sets of RED starred filter capacitors .
One should expect having a decent DC voltage level across them, when powered up.. Sound KLAXXON HORN . . . .that's not happening now !
[...]
The very top pin of the cable is power for the control panel and should have the same voltage as
the 2 star E- cap.

Initially you found some . . . or all . . . . of the of the 3 RED starred E- capacitors have / had only ~0.5 VDC present.
With you reading such a somewhat higher 9 AC voltage, suggests its not being subjected to loading.
i realize i'm not entirely sure how to test the voltages here.

the ~0.5VDC i measure is from the top pin to bottom pin of the white cable connector. however, if i measure from the top pin to *either* of the legs of the "9VAC" from the transofrmer, i get a bit over 5VDC. i don't really understand this, i guess. where is the "ground" of the DC circuits, and how is it related to the legs of the AC?

i also get interesting measurements in the rectifier sequence of diodes. if i measure from the last diode, that is at right angles (the series steering diode?) to the beginning of the first diodes, i get ~3.1 VDC (to any of the 4). however, if i measure to the "outer" diode leg next to the ID "D5", i get over 12VDC. i haven't really puzzled through the logical shape of the circuit, but it makes me think the rectifiers are probably working fine?

then should i start suspecting the caps? how do i test them? or the ICs?


again, thanks so much for your help!!
 

.brush

Oct 5, 2022
12
Joined
Oct 5, 2022
Messages
12
ok, ok, i'm starting to figure more things out. the 2-star capacitor (labeled C1), measured across its two terminals from the foil side, measures the same as the "first pin to last pin" on the control panel connector: 0.3 - 0.6VDC (it wanders). however, the 3-star cap (labeled C2) measures 2-3VDC, and the 1-star cap (C1) measures ~15.5VDC. i'm imagining that perhaps these differ in order to provide different voltages for different purposes -- such as for the fans, that are connected to the higher voltage at C1.

at any rate, i've also now followed the foil more or less like you already told me above, but i understand it a bit better.

the 7805A U5 regulator is outputting 0.5 VDC to the 2-star (C3) capacitor, but that's because it's input is just 3VDC, the same (i think) as the 3-star cap (C2). this seems to me to be the problem. it's coming directly from one end of the "steering" diode, i believe. is it likely that some of the diodes are blown, leading to low voltage, even though voltage tapped at a different part of the rectifier sequence is fine? (or maybe it's just two separate bridge rectifiers, one of which is failing?)

suggestions for best ways check diodes (on-board and/or off-board) using a multimeter, anyone? onboard, just look at the voltage drop, or....?

thanks so much!

(also, U1 that you asked about, which is connected to the sensors, appears (according to TI's website) to be an opamp. i'm assuming that maybe it's for amplifying the signal from the sensors before sending them to the main control IC? but this maybe is less important at this stage since it doesn't seem in line with the power issue.)
 

kellys_eye

Jun 25, 2010
5,282
Joined
Jun 25, 2010
Messages
5,282
the 3-star cap (labeled C2) measures 2-3VDC,
If component designation follows the usual standards then C2 will be the smoothing capacitor for the rectified 9VAC input and should have around 12-14VDC across it.

The 9VAC input should be measured with your meter set to its AC position - measure straight across the AC input (the two blue wires).

The most basic of issues is that the 9VAC is not present........

If the 9VAC is present then one (or more) of the diodes has blown. These can be tested individually using your meters built-in diode-test function where, if you connect the red/black leads across any ONE diode, the display will show anything between 0.5 and 0.8V in ONE direction and 'open-circuit' in the reverse direction (swap the lead positions).
 

.brush

Oct 5, 2022
12
Joined
Oct 5, 2022
Messages
12
If component designation follows the usual standards then C2 will be the smoothing capacitor for the rectified 9VAC input and should have around 12-14VDC across it.

The 9VAC input should be measured with your meter set to its AC position - measure straight across the AC input (the two blue wires).

The most basic of issues is that the 9VAC is not present........

If the 9VAC is present then one (or more) of the diodes has blown. These can be tested individually using your meters built-in diode-test function where, if you connect the red/black leads across any ONE diode, the display will show anything between 0.5 and 0.8V in ONE direction and 'open-circuit' in the reverse direction (swap the lead positions).
thanks for the response! and you nailed it!

as noted the 9VAC *is* present (or actually, it's 11.5 VAC, which was previously suggested might be the result of lack of load).

before getting your message, i checked the diodes with the power off, using the ohmmeter. all except one (D7) run 850-900 ohms in one direction; D7 is open circuit both ways. so that's it!

now the question is which diode to get to replace it. just a simple 1N4007? i don't have any on hand, and it seems strange just to order one with shipping and all (now that radio shack is gone) -- suggestions for good diode (or general component) assortment packs to get?

thanks again!
 

kellys_eye

Jun 25, 2010
5,282
Joined
Jun 25, 2010
Messages
5,282
Any of the 1N400x range will suffice for this job but I tend to get 'quantities' of 1N4005/6 to cover 99% of all my needs. The 4007 is rated at 1000V and whilst it costs the same as the other ranges I only keep half a dozen in my stock.
 

.brush

Oct 5, 2022
12
Joined
Oct 5, 2022
Messages
12
one more question -- as far as i can tell, the 1N4005/6/7 differ only in that voltage may flow backwards over 600/800/1000V -- is there any reason you'd want to have lower rated ones? why not just get the 1N4007s for all purposes? thanks!
 

kellys_eye

Jun 25, 2010
5,282
Joined
Jun 25, 2010
Messages
5,282
It's a minor point - the range of 1N400x diodes actually has a price range that differs some 120% between cheapest (1N4003 for some reason) which is only 3p and most expensive (1N4007) which is 7p. OK, this difference is 'pennies' and for most people the quantities ordered wouldn't amount to anything noticeable so go for the '07 if you want 'stock'.
 

73's de Edd

Aug 21, 2015
3,462
Joined
Aug 21, 2015
Messages
3,462
Sir .brush . . . . .

In reviewing your last photos, seems like they are using two hermetically coated HEFTY 10K disc thermistors, they should be troublefree.
The colorization of the heater coils reveal and suggest a Mean Time Between Failure lifespan of an additional 200 years . . . . . relative to the conservative manner in which they are being used.

Now the quirky thing is the extremely low 0.5 VDC voltage out of your boards power derivation portion.
But that is further confusing by the, yet, relatively high AC voltage out of your power transformer. A working condition of supplying the required DC voltage at its proper power level, one would expect a slight drop of that raw AC input voltage due to " pull down " effect of the supplied circuitry's loading effect.

And now . . . . . GITTENDOWNNTOTHEREFINEDHOWTODOITTOITASPECT . . . . .

REMARK UP OF BOARDS POWER FLOW . . . . . .


#2 Dehydrator PCB.png


INITIAL COLOR CODING . . .BLACK lettering within YELLOW SQUARE INSETS

AC Supply comes in A and B and feeds into the YELLOW RECTANGLE Full Wave Bridge configured with diodes D1- D4 and outputs its DC voltage as the C pair of diode anodes as the - connection into that C foil ground plane.
The D pair diode cathodes are the + supply.
D routes upward to E , where it branches up to F and gets its main filtering with 1 RED Star E-cap.
Also the other branch route is from E over and up G and a jumper to H to then pass up the YELLOW dot path and initially into I to feed power to the ULN2003 and then on up to the J-K-L-and M path to feed power into the two fans and two non terminated lamps terminals.
If there is activation of a ULN2003 section(s) , that will complete a power loop to ground and an operation of that function.
Now for tracing down the second power aspect . . .

Change to . . .
COLOR CODING . . .YELLOW lettering within BLACK SQUARE INSETS

Start at the previous bridge + output of E and now note that there is a series arrangement of diodes
D5-8 and that an additional D9 is being to their side and end up at J .
That means that these series stringed diodes create 5 X ~ 0.700V voltage drops combined and that their
combined ~ 3.5 V drop from the principal supply is then presented to the input pin of the UL 7805 3 term regulator at K .
AFTER first absorbing up some additional E- cap filtering down at 3 RED Star cap L.

Change to . . .
COLOR CODING . . .WHITE lettering within BLACK SQUARE INSETS
5 VDC comes out the UL 7805 3 term regulator and travels the WHITE dot path up to B terminal of the 2 RED Star filter cap and also over to C terminal of the connector panel which feeds power to your control and display board.

PROCEDURAL . . . . .

Now if I found a bad diode . . .or more . . . in a discrete built up of a FWB, I would replace all 4 to be somewhat assured of a balanced condition.
HOWEVER in your one found bad open diode I wouldn't hesitate to just replace that one with a 1n4000 series diode family. Expecting it to have a like voltage drop as the others, if you have multiple choice options, elect the closest value to the others tested.

So o o o o o o it looks like your one open diode is opening the second and third power paths and you are presently just reading a sneak /phantom/ back leakage voltage and is static and now, not being a DYNAMIC voltage on those associated filter caps.
Hopefully . . . . . . . and VEWY-VEWY likely . . . . just that one diode will get you dwying apple slices, etc again.


Thaaaaaaaaaaaaaaaaaasit . . . . . .

73's de Edd . . . . .


All the world is a stage . . . . . . . . .. and most of us are desperately unrehearsed.


.
 
Last edited:

.brush

Oct 5, 2022
12
Joined
Oct 5, 2022
Messages
12
super useful analysis of the board, helps me understand more things. i had been thinking that the second suite of diodes must be a separate FWB, but it wasn't quite symmetrical -- but you're saying it's just used to reduce the ~15.5VDC to more like 12VDC, something easier for the regulator to handle? is it normal to use diodes for that?

anyway, the replacement diodes should arrive tomorrow. i'll let you all know how it goes! deep appreciation for the wisdom and skill on this forum!
 

.brush

Oct 5, 2022
12
Joined
Oct 5, 2022
Messages
12
just a somewhat tardy update that the new diode works perfectly, solves the problem. dehydrator back on track!

thanks to everyone!

a small note: i confirmed what most of you doubtless know, that testing diodes via resistance can be hit or miss. one of my multimeters gave the results above, which correctly diagnosed the diode -- but another one had much more inconsistent results, probably due to using very low current maybe to test with? but this other one had a diode tester function, which the OG one did not, and that function worked perfectly to find the failed component.
 

kellys_eye

Jun 25, 2010
5,282
Joined
Jun 25, 2010
Messages
5,282
...and thanks for the update - nice to hear that things worked out for you.

Lesson learned = use the right tool for the job!
 

73's de Edd

Aug 21, 2015
3,462
Joined
Aug 21, 2015
Messages
3,462
Sir.brush . . . . .

You say . . .

" one of my multimeters gave the results above, which correctly diagnosed the diode "


Indeed, that reading is displaying in millivolts and that Vf of the diode junction might suggest a hint of the type of diode it is, also.

Expecting 600-800 mv for most common silicon diodes . . . . . .500ish for some higher voltage fast-power switching diodes and finally a lowish 140-300 mv for Schottkey diodes and some old family germanium diode junctions.


" is it normal to use diodes for that? "

Quite common to see being used to trim in small voltage drops, I sort of prefer it . . . . . to the, otherwise, using of a low value resistor ( + pure heat creation / power loss ).

73's
 

badger2

Oct 30, 2022
2
Joined
Oct 30, 2022
Messages
2
thanks all contributors to this discussion - indentified and repaired blown diode on my dehydrator thanks to all of you. control panel now powers up at 5volts and looking good, just need to figure out why the control panel buttons dont respond now! a reed switch is mentioned in the original post, note that there is a corresponding magnet in the door . about 4 inches from right hand side. kind regards
 
Top