JBC clone with good ground noise immunity

[blueskull]

I'm working on a JBC solder iron controller design which aims small size and light weight as well as USB control.

At the beginning I thought it would be easy -- nope.

I was able to get the power path sorted out, but the temperature sensing part annoys me.

As seen here:



The DC- of the heater is NOT grounded, which means if I ground the iron (which is USB ground), I need USB isolator, which is fine.

The second problem is the thermocouple connection. The TC doesn't have a Kelvin connection, which means during TC sensing, if the tip is connected to an electrical noise surface (say, a grounded case with high EMI emission), the TC readout would be a mess.

The second issue is exact the issue that brings the idea of building a controller -- my original JBC controller went crazy when I solder on ground planes connecting to my PXI box, which is noisy.

So, besides of building a custom JBC handle (which has Kelvin connection for GND), is there a solution that allows me to read thermocouple input while soldering on a noisy surface?

Also, the above mentioned diagram might be wrong -- it could have a Kelvin connection, just not I'm aware of. Does anyone have a cartridge pinout for C470 and C105?

[Ysjoelfir]

I'm pretty sure the pinout is not correct. IIRC the common connection is NOT the small contact, but the tip. So the heater would connect to the other side of the thermocouple.

edit: Just to clarify: I build a JBC controller myself and stumbled upon many many different pinouts and examples. So I measured it myself and found out why some pinouts are made up the way you showed. The Thermocouple showed a negative voltage at room temperature when I measured it, as it was pretty cold in the lab. So it seems natural to assume the shown connection. but as I measured it after heatin up the room over 25°C suddenly the polarity seemed to have changed. It seems to me that the TC has its zero point at 20°C, goint to a negative value below that, which could be accidentaly result in a erroneous connection diagram. However, I measured the series 245 cartridges, but I assume they are prety much the same as the 407, just lower power. The C105 however just is a series of TC and heater, no third connection, the complete heating current flows through the TC.

[T3sl4co1l]

Is there some rectifying junction in the iron or connector that's preventing a simple filter from solving this?  It's not like a thermocouple needs much bandwidth, even in a high performance thermal system like this.

Also I take it the cable is just wires, no shielding or anything.

Tim

[Ysjoelfir]

Is there some rectifying junction in the iron or connector that's preventing a simple filter from solving this? 

Also I take it the cable is just wires, no shielding or anything.

In my original JBC (just double checked as I have it on my workbench at hand) it is indeed just a single, 3 conductor cable, no shielding, and no rectifying junction in the hole connector-cable-holder-connector-cable-iron handle-tip chain.

[magic]

It seems to me that the TC has its zero point at 20°C, goint to a negative value below that, which could be accidentaly result in a erroneous connection diagram.

Thermocouples have their zero point when the tip is at equal temperature as the plug, and go negative below that

[KL27x]

Thermocouples have their zero point when the tip is at equal temperature as the plug, and go negative below that

By "plug" you mean the cold junction. This is where the bimetal wires end. This will be at the connector end of the cartridge. Not the plug on the soldering iron. This is one of the reasons the cartridges are so long.

That may or may not be what you meant. But the guy who made the unisolder apparently, IIRC, put a temp sensor for cold junction compensation inside the soldering station. So I think some people will make this mistake.

I agree, it sounds rather unlikely the thermocouple will produce a voltage when the entire tip is at the same temperature.

[SiliconWizard]

Thermocouples have their zero point when the tip is at equal temperature as the plug, and go negative below that

By "plug" you mean the cold junction. This is where the bimetal wires end. This will be at the connector end of the cartridge. Not the plug on the soldering iron. This is one of the reasons the cartridges are so long.

That may or may not be what you meant. But the guy who made the unisolder apparently, IIRC, put a temp sensor for cold junction compensation inside the soldering station. So I think some people will make this mistake.

Absolutely.

Thing is, many soldering stations do that. I'm almost sure JBC does that as well. There's no temp sensor inside the original JBC handles. If the cartridge is reasonably well designed, the temperature at the connector end of the cartridge is probably close to ambient temperature (although I'd expect it to be higher than ambient). JBC may have modeled this inside the controller's software to get something more accurate. OTOH, not doing any cold junction compensation would get you a significant error over the rated operating ambient temperature range.

[dom0]

JBC specifies +-5 % on the temperature, i.e. at 300 °C +- 15 °C with ambient temperatures from 10-50 °C. So compensating for ambient temperature in the station would be quite sufficient, because these specs will refer to the iron sitting in the holder and not someone with high fever clutching the handle for an hour.

[magic]

By "plug" you mean the cold junction. This is where the bimetal wires end. This will be at the connector end of the cartridge. Not the plug on the soldering iron. This is one of the reasons the cartridges are so long.

That may or may not be what you meant.

I surely meant the plug at the end of the cable, I didn't think much about it and about the implications of one thermal sensor wire being shared with heater power. Of course you are correct.

I agree, it sounds rather unlikely the thermocouple will produce a voltage when the entire tip is at the same temperature.

I would love a thermocouple which can do that. Scale it up and it's the last work I need to do in my life
Nevertheless, that's a big "IF". Or "WHEN"

[MSS]

I've made simple article about JBC C105, C210, C245, C470 pinouts http://adgd.ru/2021/01/04/jbc-soldering-cartridges-pinouts/