Adding in a boost for full power

[Aurical]

For future revisions, perhaps you could consider integrating in a boost converter to supply 24V to the heating element, such that the full 65W of power can be used. Perhaps the space freed by the barrel jack can be used for this.

[Jana-Marie]

Hey, I already thought about that and I will try! A boost would also allow one to draw 15W with a 5V only supply (currently ~4.2W)

[julidau]

I mean technically this would be all the change needed, right? (although you might need a beefier diode...)

EDIT: also (for apparent reasons) an n-fet instead of a pfet, silly me

[julidau]

If you really want 65W@24V you might end up with these values.

Source: http://schmidt-walter-schaltnetzteile.de/smps_e/aww_smps_e.html

it might be possible to fit this inductor in there. It was the smallest one i could find that tolerated the current@100khz with 33uH.

PS: sorry for the spam

[Jana-Marie]

It's not that easy and your design cannot work, multiple problems:

• with the schematic displayed the tip will always be powered, a boost converter can not output a voltage < Vin
• you probably won't find a diode that can dissipate the power needed and that fits into the case
• you are measuring the inductor current, not the input current
• you cannot measure the tip-temperature while having a buck-boosts output directly connected to the tip. To measure the temperature one of the tip-connections has to be floating

To actually get this to work you would need at least a synchronous boost with a cascaded PWM controller for the tip. A buck-boost is not much more complicated then a boost but allows you to have a voltage controlled tip (with a static pwm controller for temperature measurement).

[julidau]

hm yes i have not considered the case of turning the iron off so using a synchronous boost converter would indeed be needed. But why would you need a cascaded boost controller ?

[Jana-Marie]

A synchronous boost can also not turn the Iron off, the output fet will always conduct. You would need to PWM the output of the boost to disconnect the tip

[puzrin]

@Jan--Henrik i think, in current time, you could just switch to USB-PD PPS mode only:

• up to 21v out
• no need for PWM in iron (low EMI, more simple HW)

See Bazeus 65W GaN. I guess, if people purchased TS100, puchasing USB PD with PPS for 20-30$ will not be a problem.

PS. stusb4500 you use supports only fixed profiles. Need something different for PPS, fusb302 and so on.

[Jana-Marie]

Hey @puzrin, Thanks for the input, however that won't work. As discussed above a "Off-Time" is needed to sample the tips temperature. The first version of the Otter-Iron features a FUSB302B and is otherwise fully functional, It didn't succeed because I couldn't get the software PD stack to run.

The mentioned power supply looks nice!

[puzrin]

It didn't succeed because I couldn't get the software PD stack to run.

I understand you very well :). I do now micro reflow table, and temporary switched from USB PD to AC version to minimize risks. May be, will try to do USB-PD sink lib for all, but can't promiss anything.

One more note - you could drop DFU jumper and use existing button instead. See https://easyeda.com/puzrin/dispenser, this approach works well.

[Jana-Marie]

I use this PD lib: https://github.com/Jan--Henrik/USB-PD-Firmware It works, but other peripherals are quite buggy in chibiOS.

I did that here: https://github.com/Jan--Henrik/Otter-Iron/tree/master/KiCAD/Otter-Iron_Devkit I may also Implement that in the main Iron, if I do a new revision.

[designingSparks]

What is the resistance of the tip when it is cold e.g. 20degC and hot, e.g. 350degC?

[julidau]

@designingSparks do you mean heater resistance? Or the "resistance" of the thermocouple?

[designingSparks]

@julidau I mean the resistance of the heating element. I guess it increases slightly when hot.

[julidau]

@designingSparks

i hope this can satisfy your curiosity. my measurement leads unfortunally do not survive > 200°C 😉

EDIT: raw data

[designingSparks]

@Jan--Henrik and @julidau I believe there would be a way of making this work with a boost converter: Configure the duty cycle of the boost converter such that when it is on, approximately 24V is generated. Then just pulse the boost converter on and off at a low frequency of about 100Hz. This could be done by pulling the gate of the N Mosfet low using an npn (or in SW by disabling the boost timer) Use the duty cycle of the LF pulse to regulate the temperature. During the off phase of the LF pulse, the temperature can be measured.

[designingSparks]

@Jan--Henrik Is there space for an SMC diode? e.g. B530C-13-F, 30V, 5A. That should be capable of handling the required current.

Update: I checked this on the PCB. An SMC is around 6.2x7.1mm max. This would just fit in where the existing diode is.

[puzrin]

FYI: T80P with USB PD available. But... only 30W (12v 2.5A) and only 5 types of tip.

To be honest, i would prefer T12 with USB PD (can be done 40W). Because choice of good tips is nice, and "expencive" Hacko tips cost only 10$.

[Jana-Marie]

@puzrin Thanks for the information! Also check out my Otter-Iron-PRO, it can be modified to accept Hakko T12 Handles by only changing the connector-pcb. https://github.com/Jan--Henrik/Otter-Iron-PRO

[Jana-Marie]

@designingSparks That Idea can/will work! I already added the inductor-footprint to my Otter-Iron-Pro design, but I currently don't have enough free-time to implement and test that topology :/ Please ping me in ~2-3 weeks, if I've done nothing until then!

[Jana-Marie]

@julidau Nice chart, thank you for creating it! :)

[puzrin]

Also check out my Otter-Iron-PRO, it can be modified to accept Hakko T12 Handles by only changing the connector-pcb. Jan--Henrik/Otter-Iron-PRO

With usability in mind, it whould be preferable to have everything inside stick. Something like 9501 handle with custom 3d-printed tail.