VapCap Induction Heater for Desktop and in Car Use

tuuwie

New Member
You hit the same dilemma as me. Many bad batches these days. In fact, all have been bad since last fall it seems. I test a couple thoroughly from each new batch and sadly find they fail. At that point, I just change out 100% of the FETs for that batch. Found failure rate has gone way down to just a few percent since starting this practice.
BTW, have tried multiple Ali suppliers and Banggood to find they are all the same. At least they have the diodes in the right way around these days...
My advice would be to always order a few extra FETs when getting one of these ZVSs and must have some soldering experience. Or know someone that does.
:science:

Hi Pipes. Thanks for chiming in. Just out of curiousity, my problem definitely seems to be a blown MOSFET right? These are encased under the board covered with that casing right?

These NTD4806NG have being working out great. Not the highest specs in the world but good enough for our task. And we know what they are..... lol. Unlike the ones that come with it with no markings. At least not any of the ones I've changed out had markings on them.

Swapping them out for these should fix the problem? Do you have a spec for these? Maybe I can find some locally instead of waiting for the 20+ days of shipping.

Much appreciated guys.
 
tuuwie,

Pipes

Addicted DIY Enthusiast
Accessory Maker
Yes, it's the FETs are gone. The reason is most likely quality issues but look for cold solders as well.
I've gotten use to changing them. Twist those big coils as much as they will go so you can access the legs. Use hot iron with needle tip in good condition and put under each leg and prey up until lifts off the board. The good news is that the solder bath is so bad they lift easy. Once the two legs are lifted, heat behind the heat sink at the other end. Again with hot iron. Applying a tad of solder will speed up the process. Apply side force until it slips off it's location.
Solder in new ones. I wedge under the big coil and use the coil to hold in place while I dab solder on the legs. Then solder the heat sink portion back on. Clean up and check work before firing.
Also, remember to never fire up without the work coil installed.

Just google specs for NTG4806NG. I'm sure DigiKey or alike will have them.

Good luck! ;)
 

mr_cfromcali

Accessory Maker
Accessory Maker
May or may not be of interest to you diyselfers. :shrug:

https://imgur.com/AngRCFo

Hi, thanks for linking to this here. I am the OP of this post on Reddit and the linked Imgur album of the built unit, and that's my build. It works great and I am very happy with it.

As I said in my notes in that post, I am definitely aware the hard work and trial and error you guys and gals in this thread did to get things to this point. Thanks!

My build. Notable points: momentary switch under coil, MOSFET board and wiring under unit. Works great.

AngRCFo.jpg
 
Last edited:

zhong8051

Member
Hi Pipes. Thanks for chiming in. Just out of curiousity, my problem definitely seems to be a blown MOSFET right? These are encased under the board covered with that casing right?



Swapping them out for these should fix the problem? Do you have a spec for these? Maybe I can find some locally instead of waiting for the 20+ days of shipping.

Much appreciated guys.
My circuit using IRFZ44N and it working good
 
zhong8051,
  • Like
Reactions: Pipes

zhong8051

Member
Hi everybody,
I make a new circuit and have some testing the position of the vap in heater.



I want to know the position of the vap in your heater in any case, 1 or 2 and what the best case ? thank you for reading ^^
Case 1 the vap tip inside the coil.


Case 2 the vap tip outside the coil.

 
Last edited:

phattpiggie

Well-Known Member
Accessory Maker
@Squiby, @Summer, @Ramahs and @Blu. I'll answer your question here.
This is the way my first Skelly was put together, it's now in the Sumac with a DC jack instead of batteries.
I'm probably going to make a desk top one up for when I'm in the shed.
qX0w5cY.jpg

After I've finished this Mini version off, apologies for the focus.
NAgS4MP.jpg

All my IH's together for a bit of scale
AD15VHW.jpg

To get the 'log' style heater involves a reposition of the coil. This is done to keep the diameter down.
If I was to 'log' a Skelly I would be inclined to put the heater on top of the battery sled, this means I can make an access door in the base. It also means they would end up more cone shaped.

I'd be happier retro fitting them than sending out an enclosure and hoping that the end user could fit the heater properly.

No idea on a price or time frame. Not having the tools or space makes them an involved build.
 

Mr. Me2

Well-Known Member
@Squiby, @Summer, @Ramahs and @Blu. I'll answer your question here.
This is the way my first Skelly was put together, it's now in the Sumac with a DC jack instead of batteries.
I'm probably going to make a desk top one up for when I'm in the shed.
qX0w5cY.jpg

After I've finished this Mini version off, apologies for the focus.
NAgS4MP.jpg

All my IH's together for a bit of scale
AD15VHW.jpg

To get the 'log' style heater involves a reposition of the coil. This is done to keep the diameter down.
If I was to 'log' a Skelly I would be inclined to put the heater on top of the battery sled, this means I can make an access door in the base. It also means they would end up more cone shaped.

I'd be happier retro fitting them than sending out an enclosure and hoping that the end user could fit the heater properly.

No idea on a price or time frame. Not having the tools or space makes them an involved build.
You’re on the path of our earlier conversation. Love it!!!
 

Ramahs

Fucking Combustion (mostly) Since February 2017
@Squiby, @Summer, @Ramahs and @Blu.

If I was to 'log' a Skelly I would be inclined to put the heater on top of the battery sled, this means I can make an access door in the base. It also means they would end up more cone shaped.

I'd be happier retro fitting them than sending out an enclosure and hoping that the end user could fit the heater properly

I'd want something like that for home-use, so no battery sled needed. Just DC, as I'd never put batteries in it anyway.
 

tuuwie

New Member
@Squiby, @Summer, @Ramahs and @Blu
To get the 'log' style heater involves a reposition of the coil. This is done to keep the diameter down.
If I was to 'log' a Skelly I would be inclined to put the heater on top of the battery sled, this means I can make an access door in the base. It also means they would end up more cone shaped.

Hey @phattpiggie, do you change out the FET's on the ZVS Heating circuit too? If yes, which ones have you had good experiences with?

Thanks in advance
 
tuuwie,

rz

Well-Known Member
@tuuwie
Look for:
N-channel MOSFET
Low Rds ( < 20mOhm?)
Vds breakdown 40v at least (Lower values *may* work, 60v is even safer)
Vgs threshold should be low, below 4v.
Low gate capacitance.

Look for keywords in the feature description like:
High-speed switching
Small gate charge
Low drain-source on-resistance (Rds)
Logic level (= low Vgs threshold)
etc..
(these may not be featured, but need to be found in the specs)

What works best for you depends on your specific build ( IH frequency, maximum current draw, power supply voltage and how it responds to current draw)

Below are some examples of ones that I've found to work well (and are probably over-kill in some aspects). There must be thousands/hundreds available that would work, and if you have any sort of large electronics component store in your area they're likely yo have at least one.

https://www.mouser.com/ProductDetail/757-TK6R7P06PLRQ
https://www.mouser.com/ProductDetail/726-IPD079N06L3G
https://www.mouser.com/ProductDetail/863-NVD5C688NLT4G
https://www.mouser.com/ProductDetail/512-FDD9410_F085

So look around for what you can find that looks relevant and ask us what we think.
 

tuuwie

New Member
@tuuwie
Look for:
N-channel MOSFET
Low Rds ( < 20mOhm?)
Vds breakdown 40v at least (Lower values *may* work, 60v is even safer)
Vgs threshold should be low, below 4v.
Low gate capacitance.

Look for keywords in the feature description like:
High-speed switching
Small gate charge
Low drain-source on-resistance (Rds)
Logic level (= low Vgs threshold)
etc..
(these may not be featured, but need to be found in the specs)

What works best for you depends on your specific build ( IH frequency, maximum current draw, power supply voltage and how it responds to current draw)

Below are some examples of ones that I've found to work well (and are probably over-kill in some aspects). There must be thousands/hundreds available that would work, and if you have any sort of large electronics component store in your area they're likely yo have at least one.

https://www.mouser.com/ProductDetail/757-TK6R7P06PLRQ
https://www.mouser.com/ProductDetail/726-IPD079N06L3G
https://www.mouser.com/ProductDetail/863-NVD5C688NLT4G
https://www.mouser.com/ProductDetail/512-FDD9410_F085

So look around for what you can find that looks relevant and ask us what we think.

Thanks so much for linking me with this info.
Really appreciate it.
Will ask around and let you guys know the MOSFETs available here.

Any chance you guys happen to have a circuit diagram for the ZVS Induction Heater? I have been looking around and haven't found any part numbers / datasheets / circuit diagrams. The ones i've found, I am not sure match the spec of the heater I have thanks to the lack of knowledge on my part. I was looking for it in case I have to take this blown one to a friend to replace the MOSFETs.

Thanks in advance
 
tuuwie,

rz

Well-Known Member
@tuuwie
http://fuckcombustion.com/threads/v...top-and-in-car-use.23211/page-75#post-1220762

http://kaizerpowerelectronics.dk/high-voltage/mazilli-zvs-flyback-driver/

in general, google Royer oscillator / Mazilli flyback driver.

@zhong8051 looking good. Keep up the builds ! :) The fast click in your second video is due to the shallow insertion. Since the bi-metal disc at the end of the cap (that does the clicking) is actually being induction heated itself, it heats up quickly and clicks. This does NOT indicate the actual chamber temperature. It is more likely to result in slight charring of any material touching the end of the cap, and a tip that cools down way too quickly since the rest of the tip doesnt get a chance to heat up enough. Your first video's insertion depth looks much better.

What are you measuring there, Amps @ 120 / 220 V? Either way thats some powerful stuff. How much capacitance do you have on there? Got an oscilloscope? :)
 

zhong8051

Member
@tuuwie
http://fuckcombustion.com/threads/v...top-and-in-car-use.23211/page-75#post-1220762

http://kaizerpowerelectronics.dk/high-voltage/mazilli-zvs-flyback-driver/

in general, google Royer oscillator / Mazilli flyback driver.

@zhong8051 looking good. Keep up the builds ! :) The fast click in your second video is due to the shallow insertion. Since the bi-metal disc at the end of the cap (that does the clicking) is actually being induction heated itself, it heats up quickly and clicks. This does NOT indicate the actual chamber temperature. It is more likely to result in slight charring of any material touching the end of the cap, and a tip that cools down way too quickly since the rest of the tip doesnt get a chance to heat up enough. Your first video's insertion depth looks much better.

What are you measuring there, Amps @ 120 / 220 V? Either way thats some powerful stuff. How much capacitance do you have on there? Got an oscilloscope? :)
I using step down from 24voltDC to 12volt DC. The ampe meter measuring DC ampe, 4x330nF 400V. I dont have oscillos. What you want see on osc ?
 
zhong8051,

tuuwie

New Member

Wow, that was quick. Thanks so much for sharing this man. Really makes my life so much simpler.
Will try and swap out the MOSFETs of my busted induction coil and will share the MOSFET components i am able to source here. I am sure it'll be similar to one of the ones you've linked. Thanks again. Much appreciated.
 
tuuwie,

rz

Well-Known Member
@zhong so thats about 7.5A @ 24Vdc? ~180Watts. Assuming 90% efficiency in the step down that's still over 150W delivered to the IH. Those transistors must be getting toasty. You could try reduce capacitance. 0.33~0.66uF range should be good depending on your coil. I'd experiment a little there, maybe use 2x0.33uF in series for 0.66, then parallel with another 0.66 string to give you 0.33 (using just one of those 0.33's alone may cause too much heating in the single capacitor), or just use two in parallel for 0.66uF. Reducing capacitance will INCREASE the self oscillating frequency which will INCREASE switching losses however it will also increase both the capacitive and inductive reactance (Xl == Xc), which causes general current consumption to DECREASE. There's no need to heat up the vapcap TOOOOO quickly, which often results in burnt loads anyway. An oscilloscope will show you the frequency, which should be around 20~30kHz, and help you tune your coil/capacitance, but is not entirely necessary.

EDIT: Corrected a backward statement about frequency behavior above.
 
Last edited:

polykoma

Active Member
Thank for your advice,
I'm check again, measuring direct input IH, the curent is 14A
Temperature of transistors is under 60*C I think is not too high because we using IH with shot time.


yo - but when you have a session then you maybe heat up repeatetly over lets say 3-5 minutes.

can you simulate a "session" and monitor the transistor temp?

regards poly
 
polykoma,

tuuwie

New Member
Yes, it's the FETs are gone. The reason is most likely quality issues but look for cold solders as well.
I've gotten use to changing them. Twist those big coils as much as they will go so you can access the legs. Use hot iron with needle tip in good condition and put under each leg and prey up until lifts off the board. The good news is that the solder bath is so bad they lift easy. Once the two legs are lifted, heat behind the heat sink at the other end. Again with hot iron. Applying a tad of solder will speed up the process. Apply side force until it slips off it's location.
Solder in new ones. I wedge under the big coil and use the coil to hold in place while I dab solder on the legs. Then solder the heat sink portion back on. Clean up and check work before firing.
Also, remember to never fire up without the work coil installed.

Just google specs for NTG4806NG. I'm sure DigiKey or alike will have them.

Good luck! ;)
Just realised I completely missed this post and I am back to report that after careful analysis, the FETs on the heater seem to have cracked and completely busted.
I am checking for the ones you linked earlier and also for the ones @rz linked as well in the open market for a quick replacement. Hopefully the FETs are the only things I fried and the rest of the heater circuit still is intact.

Thanks for all the help guys. You guys are awesome
 

zhong8051

Member
yo - but when you have a session then you maybe heat up repeatetly over lets say 3-5 minutes.

can you simulate a "session" and monitor the transistor temp?

regards poly
after heating, temperature of FET about 55*C and it cool down fast so I think when we heating again after 1-2 mins it OK ^^

Just realised I completely missed this post and I am back to report that after careful analysis, the FETs on the heater seem to have cracked and completely busted.
I am checking for the ones you linked earlier and also for the ones @rz linked as well in the open market for a quick replacement. Hopefully the FETs are the only things I fried and the rest of the heater circuit still is intact.

Thanks for all the help guys. You guys are awesome
I think you can DIY 1 heater like me and forget the module you buy on EBAY ^^
 
Last edited by a moderator:
zhong8051,
Top Bottom