ARCHIVED FORUM -- March 2012 to February 2022READ ONLY FORUM
This is the second Archived Forum which was active between 1st March 2012 and 23rd February 2022
Yes, it's happening! I've long had this project in my head, and now being ill and bored for a couple of days I started thinking about it.
I've been studying the schematics for the 1900 and the 2400 and I think it should be quite simple.
(In the following I will talk about FM1 but the concept should be the same for all functions, albeit some don't have the preallocated space on the pcb. The relevant part of the schematics is included below.)
I soldered in the R111 100K Ohm resistor and when shorting the joints for where D22 should be the unit starts up, as expected. My idea is simply to perform the shorting with a transistor activated by a Teensy uC.
So far so good, I suppose. My problem is powering the uC. Obviously it needs to be powered even when the BM is turned off, so I was thinking that the +15V standby rail could be used. And this is where I could use some help from the electronics wizards. The Teensy needs 5+-0.5V power. Can I just put in a voltage divider to get a stable enough 5V? And can I do that without disturbing anything else? The uC won't consume a lot of current (a few the tens of mA for a few milliseconds when an IR command is incoming, some tens of uA when idle which is the rest of the time).
Also, supposing a voltage divider would do the trick, which resistor values should I choose?
I admit that I'm a bit out of my comfort zone here, but that's the only way to learn new things, right! :-)
So, can the standby rail be used? Or should I tap somewhere else? And how should I get 5V?
Thanks!
/ Johan
A voltage divider as two series resistors should be more than adequate when the current flow isn't higher.Resistor values are not very critical with that little current flowing, something like 4,7K and 10K I should think would work.Put a zener diode (f.e. a 5,1V) across the 4,7K resistor and connect that resistor to ground.That will give you a good 5,1V supply and it will place a load of only 1mA on the 15V supply (= nothing).If the output voltage struggles to reach 5V you can up the 4,7K to f.e. 5,1K or 5,6K.
Martin
Awesome! I will give it a try.
Thanks a lot, Martin!
Nice project!
Martin's idea will of course work if you need 1 mA max. I would add a small electrolytic and a 100n or so ceramic cap across the zener to keep the +5V output stable during load peaks.
You could also try one of the infamous 3-pin voltage regulators (7805) and you'll only need one or two caps in addition to it (at least one 100n from the output to ground, needed for stability). The low current version (78L05, in a small TO-92 case) will be enough.
--mika
Ok. The voltage regulator solution sounds easy enough. I don't have a suitable zener lying around nor a voltage regulator, so either way I need to order it. I'll add it to my list of things to order.
I'll let you know how it works out once I receive the components.
Ok, I did some layout planning. I quickly realized that drawing by hand would get very tedious, so I scanned the web for a tool. I found one that did the trick. It's not pretty, it doesn't have a single fancy feature, it doesn't include a lot of different components. But, on the up-side it has zero learning-curve. It was perfect for this little project. It's written in Java so it runs on all platforms known to man (almost). You can find it here: http://code.google.com/p/diy-layout-creator/
Anyway, below is my layout. It's pretty straightforward (resistors are pictured standing up). I realized while placing the components that putting in all those 100K (R127, R111 etc in the BM2400 service manual) resistors to ground probably was unnecessary. I figured one should be enough, no? It's the yellow resistor in the image. Please tell me if this is wrong.
Does this seem reasonable?
I'm still waiting for components so I haven't started building anything yet.
I did however pick up a second BM1900 yesterday. Excellent condition. I'm looking forward to fix it up and hang it on the wall at my sister's. With BL1000 support! :-)
I'll have to take a look at that drawing tool later, it makes really nice pictures apparently!
I had to think a bit before I realized that the schematic snippet you first posted is from a 2400, and all that diode mess is missing from your 1900
As I see it, the switching with the transistors will have to change a bit and you unfortunately will need another resistor (the yellow 100k) for each of them. That resistor needs to be on the top side (from Beomaster switching point to the collector) and the emitter connected to ground. The Teensy digital outputs only provide +5V when high, so they won't be able to open the transistor otherwise (the base voltage needs to be higher than the emitter for an NPN transistor to conduct). Better test with a single transistor circuit before soldering the final product together!
Also, test the regulator circuitry before hooking up the Teensy; it will be instantly fried if you accidentally supply any more than +5V to it. The 100n cap I mentioned (C1 in your layout I suppose?) needs to be on the output of the regulator (from +5V to ground).
The volume control seems a bit different, but I believe it will work with the same control system, you just need to connect to the bases of 3TR8 and 3TR9.
Hi Johan,
If have some doubts if this will work or if it is the right way.
The first thiing is your ground lead (middle one of the three legs) to the voltage regulator. The one you drawed goes also through the 100K resistor
That has to go directly to the ground you drawed outside the board. But nevertheless, It dissipates/consumes some power and will probably get warm/hot. So a resistor before the input of the regulator will reduce that heat.
Another thing I'm worrying about. Does the upc board have active high or active low? My choice would be active low (with open collector) combined with pnp (BC557) resistors. Then you can work independent from the supplied voltages provided to the base and emittor of the transistors used in the diagram..
Now your using 5V and you don't know if that's high enough with the 100K resistor in the "emittorlead" to the ground.
The only thing you need to do is to get the corresponding input from IC5+6 via a resistor down to ground level. Just my thoughts.
While writing, Mika was faster
Thanks for your input, I'll try to address both your posts.
Sorry Mika, I should have been clearer with which service manual I was referring to.
First I'd like to correct my stupid mistake with the ground connection. Of course it needs to not go through the resistor. Thanks for pointing it out!
C1, the 100nF cap, is on the output. Perhaps it's not clear enough in the image. It's connected to +5V and ground.
Regarding the heat in the voltage regulator, should I put a resistor between the +15V lead and the input of the regulator? What would be a suitable value to get the desired effect?
Now to the transistors. Obviously you're right, Mika. I just didn't think, I guess. Just to make things clear, I have two options? Either I do what Mika suggests and throw in a 100K resistor (or whatever value it takes to make it work) on the collector of each of the NPN transistors, and keep active high. Or I can keep the current circuit design, but use PNP transistors and active low. Is this correct?
Beobuddy, I don't understand what you mean by this: "corresponding input from IC5+6 via a resistor down to ground level".
The AVR microcontroller in the Teensy has quite versatile I/O ports; you can program them to sink & source current, also open collector (open drain actually) if I remember right, and tri-state (high impendance) as well. So this could just as well be done with PNP transistors driven low. I would prefer NPN though - simply because it feels more natural to me when the switching transistors are on the ground side
(in principle, I believe you could do this with no transistors at all - just two resistors in series, with the junction normally floating at 5V and pulled to ground with an open drain I/O pin when active - but it might be sensitive to interference, and you risk damaging the Teensy if the control line for some reason ever gets higher than 5V).
BTW, remember you need to cut a trace on the Teensy PCB before powering it up - you can take its supply either from the USB connector or from an external supply, but not both at the same time. It's somewhere in the docs...
tournedos: The AVR microcontroller in the Teensy has quite versatile I/O ports; you can program them to sink & source current, also open collector (open drain actually) if I remember right, and tri-state (high impendance) as well. So this could just as well be done with PNP transistors driven low. I would prefer NPN though - simply because it feels more natural to me when the switching transistors are on the ground side
tournedos: (in principle, I believe you could do this with no transistors at all - just two resistors in series, with the junction normally floating at 5V and pulled to ground with an open drain I/O pin when active - but it might be sensitive to interference, and you risk damaging the Teensy if the control line for some reason ever gets higher than 5V).
tournedos: BTW, remember you need to cut a trace on the Teensy PCB before powering it up - you can take its supply either from the USB connector or from an external supply, but not both at the same time. It's somewhere in the docs...
tournedos: The AVR microcontroller in the Teensy has quite versatile I/O ports; you can program them to sink & source current, also open collector (open drain actually) if I remember right, and tri-state (high impendance) as well. So this could just as well be done with PNP transistors driven low. I would
The AVR microcontroller in the Teensy has quite versatile I/O ports; you can program them to sink & source current, also open collector (open drain actually) if I remember right, and tri-state (high impendance) as well. So this could just as well be done with PNP transistors driven low. I would
Couldn't remember the exact naming at that moment, Thanks.
This method I wouldn't prefer. As you can read in service manual there is used a 100khz oscillator and I would stay far away from such source of interference as you could. A transistor used as a switch (NPN or PNP) is the safest way I think.
If you go for the NPN version, I would use an extra resistor from the base to the emittor. That way the transistor doesn't leak or switch erractic.
Beobuddy:If you go for the NPN version, I would use an extra resistor from the base to the emittor. That way the transistor doesn't leak or switch erractic.
Not really needed for normal operation, you can (and should) program the uC outputs to do the exact same thing (drive the bases actively low). One caveat is that they all turn to high-impedance lines while the AVR is being reprogrammed. They are also that way after reset until the program reconfigures them, but that shouldn't take too many milliseconds after power on
I've finally found a little time to work on this. I decided to go for NPN and I put together a little proof-of-concept circuit:
It's not pretty, but it works perfectly! :-)
I still have to check the volume control, as that is slightly different, but I expect that to work too.
I also haven't built the 5V circuit to power the uC (I used a standard usb charger).
I'll let you know when I see some more progress.
Cheers
Hi Johan, nice job, I'm as always equally impressed by this type of work.
I notice the small IR receiver, presumably similar to a TSOP7000, but where did you get it?
Kind regards
Goran
Thanks, I have since finalized the project and the 1900 is now our kitchen unit.
It is indeed a TSOP7000. I googled and found them on some German site. They are not particularly cheap and I don't know if they still have them in stock.
I ended up not using a TSOP7000 but instead an MCL2 IR-transceiver. Mainly because there is no good place to mount the TSOP in the 1900.
What are you planning to use it for? I'm curious... :-)
Here a link... http://archivedforum2.beoworld.org/forums/t/8970.aspx to what I try to achieve, or the start at least.
Hmm German site you say, only found Hong Kong based on ebay... but then for about $5 and marketed as Vichay, since they are discontinued since years now I don't feel confident ordering.
so you put the MCL2 IR on the wall next to the 1900, or behind it somehow?
since you have used that in a project you might help me further on the link above.
//Goran
Oh. Ok, I'll have a look.
I can probably dig up the site for you. I actually don't know the manufacturer.. probably not original. But they work. Reception is not as good as you're used to with B&O stuff, though. More like other brands where you actually need to point the remote to the unit.
I put the transceiver on the wall next to the BM. Actually it's on the adjacent wall.
//Bo.A long list...