Multichannel JBC Thoughts

[TheSchilk]

Hey everyone!

First of all - sorry for yet *another* thread about diy JBC stations.... I know....

I have been playing around with the idea of designing and building a multi-tool  JBC station: If I am going to open this rabbit hole, I might as well get it done once and for all!

I have done some basic prototyping and testing, and feel like I have a handle on most aspects.

There are a however a few points I would like to discuss, I decided to bunch them into one post to keep the number of JBC topics down.

1) Tool Selection

I would love to hear from people who have had the opportunity to use multiple JBC tools, to pick a good combination of hand pieces.

Right now I was planning of having a channel each for:
    - T470 high power iron
    - T240 standard iron
    - T210 precision iron
    - AM120 precision tweezers
    - NT115 nano handle
    - AN115/NP115 nano tweezers

I realize that this is very excessive, but I don't plan on buying all tools right from the get-go. If I am going to go through the trouble of designing this station I might was well keep my options open.

Specifically:

- How different are the precision iron (T210) and tweezers (AM120) from the Nano iron/tweezers? Is having both a benefit? Or do their use cases overlap a lot? The feature creep enthusiast in me wants to support both, but I would like to get a sense of just how over-kill that would be.

- Does anybody have experience with both the NP115 and AN115 nano tweezers? While the AN115 is adjustablem the NP115 looks very ergonomic.

- right now I did not plan for any higher-power tweezers than the AM120s (40w per side). Am I right in thinking that something like the AT420 would not provide that much utility?

2) Channel Isolation

Based on this thread and this reverse-engineered schematic, JBC stations power the heater and monitor the thermocouple as such:



Plus, that is how Marco Reps wired his iron, so it *must* be correct (;P).

The upside here is that, as shown above, both C245-style (with seperate TC and heater, also including the C470) and C210 stylecartridges can be connected in the same fashion.

However, by not having the AC supply for the heater referenced to mains earth directly, there are some challenges when trying to create a multiple channel unit:

If 2 C245 cartridges are connected to the same supply, this will parallel their thermocouples, which wouldinterfere with readings:



IF a C210 is thrown into the mix, it would short out a thermocouple through the two heaters:



Both are of course no good.

The easiest solution is of course isolated AC suppplies, but that makes sourcing a transformer a pain. If I do want to have the 6 channels proposed above, I would end up needing the following 8 transformer  secondaries:

42V / 250W : T470
24V / 130W : T240
3x 24V / 40W  : T210, AM120 Tweezers
3x 9V  / 14W  : Nano iron + tweezers

That would require a custom transformer (or a few transformers) and seems like a brute-force solution.

Another idea would be to introduce two switches per heater:



However this only works as long as C240-style and C210 style cartridges are not mixed. It would allow heater current to pass through the TC in parrallel.



It is worth noting that powering multiple C210-style cartridges from the same source should not be a problem, as the  AC source would simply end up being earth-referenced through the shorted parts of the cartridge:



(but I have not tested this)

Another option would be powering the cartridges in a non-standard way, and earth referncing all sources:



I think this is what the 300w 2 channel station from bavaria is doing, but I feel a bit uneasy about doing this. I would rather stick to the 'official' setup.

Right now I am tending towards the following solution using two transformers.

The first transformer would be a 2x24V transformer and could be used to power both the C245 and C470 using the double-switch technique:



The second transformer would feature a 24V and 9V winding and would be used to power all precision  and nano tools.

By not having each channel completely isolated, I can't have combined 245/210 channels like the original JBC station does, but it allows for simpler transformers.

Has anyone else gone through this? Am I way off base? Is there a simpler solution?

Pineapples?

Thank you in advance for your input

[KrudyZ]

For tweezers I have an NP105 nano, a PA1200 micro and also a PA4200.
The main difference between the nano and micro is the distance of your hands to the work. The tip size is basically the same, but you would need to get curved tips for the micro to close parallel on the smaller SMDs like 0201. The shorter hand distance of the nano reduces shaking, but makes working a bit more crowded. I don't really see a good reason to have both.
The PA4200 is a completely different beast. It outputs a serious amount of heat and is great for taking parts off or soldering the heat sinks of SMD power transistors. I rarely use them as actual tweezers, more like a dual iron you can hold in one hand.

I also have these single irons, NT105 nano, T210 and T245.
I like the nano as it is much shorter than the other ones. Between the 210 and 245 I almost always prefer the 245, it just feels more robust and the tips take more abuse such as applying some wedging force to bend a pin, not that you should ever do that to your iron.

I don't like the JBC desolder and air tools. I prefer my Madell QK857D over the JBC station I had and for desoldering I'm back to using a manual solder sucker.

[AccountRemovedPerUsersRequest]

First of all, I seldom share any ideas here since I am just a random hobbyist putting some basic stuff together and there are real professionals out there. But since I was managed to get a DIY C245 warm, I may say something.

Having more than one transformer sounds like something I would definitely not do.

Could you put all channels to "sleep mode" and right after you pick one you connect the channel to ONE transformer/TC input the way that the exact channel needs to using triac, relays, fets or what ever you're planning to. On SW level calibration and power settings are set accordingly. Maybe some kind of SW driven channel -> TR/TC matrix gives you all the options you need now and in the future,

BTW: I made my 245 for fun and at some point I realized that has become my primary soldering iron mainly because of advanced sleep functionality.

A

[TheSchilk]

First of all, I seldom share any ideas here since I am just a random hobbyist putting some basic stuff together and there are real professionals out there. But since I was managed to get a DIY C245 warm, I may say something.

Im always happy to hear from someone with experience with a certain topic, even if they are "just a hobbyist"

I'm also mostly just shooting from the hip   

Having more than one transformer sounds like something I would definitely not do.

Could you put all channels to "sleep mode" and right after you pick one you connect the channel to ONE transformer/TC input the way that the exact channel needs to using triac, relays, fets or what ever you're planning to. On SW level calibration and power settings are set accordingly. Maybe some kind of SW driven channel -> TR/TC matrix gives you all the options you need now and in the future,

Well I definetly will need at least 3 voltages:
The T470 iron runs at 42V
The T245/T210/AT120 usw run at 24V
The Nano stuff runs at 9V

So there is no way around having at least 3 secondaries, and without getting a custom transformer made, the only realistic method would be to have multiple transformers.

That is annoying thought.

Could you put all channels to "sleep mode" and right after you pick one you connect the channel to ONE transformer/TC input the way that the exact channel needs to using triac, relays, fets or what ever you're planning to. On SW level calibration and power settings are set accordingly. Maybe some kind of SW driven channel -> TR/TC matrix gives you all the options you need now and in the future,

Yes I definetly considered this.

My plan right now is to design a 'generic' channel layout that can support any iron I want by populating different parts/connecting different transformers. Then I can just put a few of those onto a pcb and configure it for the tools I will end up using during assembly.

I don't think having a single channel that is multiplexed is realisitc, mostly because there are 3 different operating voltages and the switching would be more complex than simply implementing multiple parallel channels.

That does not mean I can't do what you suggest: Software can keep track of how much power is being drawn, and only activate so many channels at once as to stay within the power limits.

[TheSchilk]

For tweezers I have an NP105 nano, a PA1200 micro and also a PA4200.

Perfect! I was hoping that there was someone with experience with all of them

I was having trouble finding info about the PA1200 and PA4200 - I guess they are older models?
But I would assume they are comparable to the AM/PA120 and AT420?

The main difference between the nano and micro is the distance of your hands to the work.

The tip size is basically the same, but you would need to get curved tips for the micro to close parallel on the smaller SMDs like 0201. The shorter hand distance of the nano reduces shaking, but makes working a bit more crowded. I don't really see a good reason to have both.

The PA4200 is a completely different beast. It outputs a serious amount of heat and is great for taking parts off or soldering the heat sinks of SMD power transistors. I rarely use them as actual tweezers, more like a dual iron you can hold in one hand.

Yeah I was starting to suspect as much. I guess having an AT420 and nano tweezers would make much more sense and cover
much more ground.

I also have these single irons, NT105 nano, T210 and T245.

I like the nano as it is much shorter than the other ones. Between the 210 and 245 I almost always prefer the 245, it just feels more robust and the tips take more abuse such as applying some wedging force to bend a pin, not that you should ever do that to your iron.

I don't like the JBC desolder and air tools. I prefer my Madell QK857D over the JBC station I had and for desoldering I'm back to using a manual solder sucker.

I have never tried them myself but I wasn't planning on going down that route for this project anyway.
That would be too much feature creep - even for me.

But thank you very much for sharing your experience!

Maybe a much more sensical lineup would see me skipping the precision/micro (T210/AT120) line all together:

- T470 high power iron
- T240 standard iron
- T420 'standard' tweezers
- NT115 nano handle
- AN115/NP115 nano tweezers

That should really cover 99% of all the soldering I would ever have to do.

[Shock]

Have you heard of the Unisolder? It runs many different JBC handpieces as well as other brands. The populated PCBs are $100, just need to add transformer and case. This youtubers implementation of it seems to have desoldering, precision hot air and C470. The firmware is fully developed as well.

A multichannel version using relay switching has been done (as the station has handpiece detection). An add on module would probably be the best way to implement this otherwise it becomes a rats nest.

http://dangerousprototypes.com/blog/2016/02/03/unisolder-5-2/
https://github.com/sparkybg/UniSolder-5.2/blob/main/README.md
https://www.tindie.com/products/gitdiy/unisolder-52c-finished-circuit-board/

[TheSchilk]

Have you heard of the Unisolder? It runs many different JBC handpieces as well as other brands. The populated PCBs are $100, just need to add transformer and case.

I have definetly heard of the unisolder - that project is amazing!

This youtubers implementation of it seems to have desoldering, precision hot air and C470. The firmware is fully developed as well.

A multichannel version using relay switching has been done (as the station has handpiece detection). An add on module would probably be the best way to implement this otherwise it becomes a rats nest.

http://dangerousprototypes.com/blog/2016/02/03/unisolder-5-2/
https://github.com/sparkybg/UniSolder-5.2/blob/main/README.md
https://www.tindie.com/products/gitdiy/unisolder-52c-finished-circuit-board/

I had not seen these! Thank you very much for sharing!

I did not see any schematics/details for the switching. For example, does it actually change the AC voltage based on the handpiece selected? Or does it also just power the 470 with 24V?

[TheSchilk]

Having slept on it for a night, here are some more thoughts:

I don't see myself using more than one iron at the same time pretty much ever, so it makes little sense to size the transformer(s) for continuous operation of all tools. I don't, however, like the idea of a single channel that is reconfigurerd for the different tools by some relays switches, because while I won't have more than one tool in my hand at a time, I would still like to be able to provide some small amount of power to the tools in sleep/hibernation mode to keep them at their standby temperature. That would require a constant relay switching.

Because JBC stations always pass complete half-cycles, I think it makes much more sense to spec the station to be able to supply any half cycle to any one tool. This way the software can heat and prioritize the tool that is currently
being used, but can keep powering other tools in standby mode on cycles where the main tool is not being powered (due to temperature regulation, or temperature measurement).

If I go for the revised list of tools (T470, T240, AT420, Nano Iron, Nano tweezers), I could get by with the following setup:

One 300VA 2x24V transformer to supply the T470, T240 and AT420, and one 50VA 9V transformer to power all the nano tools.

I particularly like this because I already have the 300VA 2x24V transformer at home! *grin*

The C470 cartridge would use both 24V windings in series to get 48V, while the other tools would use a single winding per cartridge. By having low and high side switches like I talked about in my first post should mean I can use any combination of windings on a channel, and as long as software makes sure to not turn them on at the same time, they should not cause any problems.

As far as I know, the C115 nano cartridges are wired like the C210 ones, so they could all be hooked up to the 9Vac rail in parallel and also work at the same time. Those would also not require a second low-side ac switch.

That seems like a fairly efficient solution:

The larger 300VA transformer can't really be any smaller: The C470s need 250W, an AT420 would need 260W. So it makes sense to combine as many tools onto this transformer as possible.

A second, much lower power, 9V transformer would not be that expensive and can also be used to power the control electronics.

Plus that means I can get away with two transformers that should be fairly easy to find: both only have one secondary voltage.

[KrudyZ]

I personally would be OK with just having the last used tool go to sleep temperature and all others not powered. Even from cold they get up to heat within a couple of seconds.
If you are building this from scratch then I see little reason to power the iron with AC. It's really just a remnant of the early design decisions that were made back in the days.
It's straight forward to design a programmable DC power supply and add a power switch to isolate it from the tip during the temperature measurements. The buck regulator could serve any of the tips at any power level. I also would use old school mechanical DPST relays to select which iron is active. This greatly simplifies the thermocouple readings.

[TheSchilk]

Thank you for your insight and sorry for the somewhat delayed response.

I am in the middle of exams...

I personally would be OK with just having the last used tool go to sleep temperature and all others not powered. Even from cold they get up to heat within a couple of seconds.

You are probably right. If I stick with AC, this would come 'for free' so I would probably implement it. It is however not a priority.

If you are building this from scratch then I see little reason to power the iron with AC. It's really just a remnant of the early design decisions that were made back in the days. It's straight forward to design a programmable DC power supply and add a power switch to isolate it from the tip during the temperature measurements. The buck regulator could serve any of the tips at any power level.

I looked into this a bit after your reply, and it definetly seems like a valid approach. There are some stray comments floating around the internet that AC is better to avoid electron migration, but I haven't seen any evidence to support that. (And I don't quite understand how a resisitive heater is supposed to be damaged by AC?)

A proper 300W meanwell powersupply would probably cost a little less than the two transformers I proposed above, so I don't think the cost saving would be  too much at a scale of 1 unit.

Besides being able to power any iron with a single supply and the weight (and probably cost) benefits, is there any other advantage to a DC supply?

I can't imagine the efficieny would be that different from transformer and SSR switch, but I guess it would allow the control to be decoupled from mains frequency. Would a faster update rate and finer PWM control provide such a big benefit if JBC can get by with half-cycle control?

I may still stick with an AC system because I am not very familiar with high current buck converters, and with an AC system I understand every component of my build. That would make future repairs easier. I am however very tempted to start learning about this. Besides, I already have a fitting transformer at home.

I also would use old school mechanical DPST relays to select which iron is active. This greatly simplifies the thermocouple readings.

I am not sure I like the idea of relays clicking away every time I pick up a different iron. If I will have to do power switching anyway for heater control, I may as well just use those power switches to select the heater. I think there is little to be gained by having a single channel of heater control and thermocouple sensing and switching it to the different outputs. The relays to switch all those signals would probably be more expensive than a few power-transistors and thermocouple amps per channel.

Thank you again for your insight.