On that pic you haven’t put the wires in yet, so the ratsnests will still be visible. Only when things are connected do the ratsnests disappear.
The usual mistake is that you are probably not trusting the ratsnest, and are connecting parts differently from the SCH. The SCH is the master, so you have to trust the ratsnests in other views.
I do circuits with 150 parts and never have problems, that one only has 50, it just takes practice.
I understand that the rats-nest lines disappear when the components are connected via wires, but why would the algorithm show a rats-nest across the three pins of the transistor above ?
Do I have to connect the components to their respective +12V, GND, -12V rails as I add them so that these “false/confused” rats-nest lines disappear?
I observed that when I click a breadboard “hole” that is highlighted in green, it does seem to be “aware” of the power lines since the entire power line lights up in yellow “dots”. The issue is however, that almost any component seems to want to connect to a power rail, only a very few components want to connect to a pin isolated from +12V, GND or -12V, this was the original issue I had, it seemed like the only way to get rid of the rats nest line was to connect the components to one of the V rails, so I ended up with a lot of jumper cables on my breadboard, some even for connecting adjacent pins on a transistor.
I noticed one thing though: for one or two components, when I click the green hole connected to a component, BOTH the GRND and -12V light up: does this mean I can connect that hole with a wire to either GND or -12V?
No it shouldn’t be across three pins of the transistor.
No, there is some fault.
The green dots are an indication that it is connected. Big black dots in SCH are junctions, but be sure to grab the dot and move it to see if traces are actually connected.
NO, one pin of a part should not light up both 12V+, 12V-, and GND.
I don’t know how VAN did that .fzz, because it seams to be a modified original.
If you want to upload VAN’s .fzz that you are playing with I can try and see what’s wrong.
Here is a big .fzz that you can play with to see how it’s supposed to work.
NO, one pin of a part should not light up both 12V+, 12V-, and GND.
I wrote it highlights the -12V and the GND, not the +12V is that expected?
I don’t know how VAN did that .fzz, because it seams to be a modified original.
If you want to upload VAN’s .fzz that you are playing with I can try and see what’s wrong.
Sure, the fzz is under his post http://forum.fritzing.org/uploads/default/original/2X/a/a15f83b2e11e91f9c132e7103e6a73daef33ece2.fzz
No, not + and - at the same time.
I meant your mods to VAN’s .fzz
I had a play with VAN’s before and it looks fine. If you look at the parts he set out in BB view there is no rats between component pins, so it looks like something about when you put them on the BB.
When I see links between pins on a part and I can’t clear them, it looks like that part has corrupt associations that can’t be undone. Sometimes you can delete the part and replace it with a fresh one from CORE, but I couldn’t do that with your orig .fzz.
The way to do it is only work in one view until it’s perfect - SCH for you because you have it -, and then trust the rats in the others. Remember, it has to be perfect with every connecting verified.
EDIT
Grab the rats and pull it to see where it’s coming from. Sometimes they look connected when they are just running parallel.
It appears our usual advise that you can’t recover a corrupted sketch is incorrect. I just started with your original sketch and deleted all the wires in all the views (including the breadboards, because they create virtual wires when things are connected to them). That eliminated all the rats nest wires (which we didn’t think could be done). It wasn’t easy, as there were several wires that were very hard to see, I ended up using autoroute in pcb to create a wire so I could see what component had a connection, and then discovered a close to invisible wire in breadboard that I had missed! That sketch is here:
roland-tr-808-bass-kick2_no_wires_or_bbs.fzz (22.8 KB)
Then using the above sketch I added the two breadboards back in from core parts, and unlocked all the components and moved them so that they connected to the breadboards. That creates this sketch which has some problems already:
roland-tr-808-bass-kick2_bb_in.fzz (28.9 KB)
The only change on this one is that the 2 breadboards have been added back in and the components unlocked and moved so they connect on the breadboard. If we now switch to schematic and look at the rats nest lines the connections create we see why the problems occur in the original. Some of the rats nest lines match the wires on the original schematic, but some do not. For instance if we look at the base of transistor t5 on the left, we see the rats nest line from breadboard is connecting to the bottom of resistor R20. In the original schematic there is a connection from the bottom of R20 to the ground just below it. That however if done, will create a short between the base of T5 and ground. To correct that R20 in schematic needs to be rotated 180 degrees so the currently unconnected pin of R20 points towards ground, so when the connection to ground in schematic is made it will match the connections already made in breadboard rather than cause a short as it will now. R33 on the emitter of T5 should be eliminated and replaced with a wire (I did that on the schematic I made). One end of R21 is connected (in breadboard) to the ground end of D2, with a connection to ground on the cathode of D2 there will be a short on one end of R21 that shouldn’t be there. This indicates there is a placement / connection error in breadboard view that needs to be corrected. C9 connects to the anode of D2 in breadboard (which it shouldn’t) again indicating there is a connection error in breadboard view that needs correcting. Similarly the base of T2 connects to the emitter of T4 which should connect to ground in the schematic and which will cause a short. That needs to be corrected in the breadboard layout by moving the components so they don’t connect incorrectly. You need to go through the rest of the connections and move the components in breadboard until the rats nest connections in schematic match correctly (or start with the competed schematic and make sure the two views match. Starting from here should let you use most of the original layout in breadboard (once you correct the parts that have problems either in schematic or in breadboard which ever is easier). Note you probably want to replace the current op amp with the tl082 to get the sub parts to be able to move them around as well.
edit: as promised, I just uploaded a part for the ±12V converter I mentioned above to parts submit in thread:
While making it I discovered the input voltage range is actually 2.5 to 5.5V so it will run off 2 AA cells if you want your device to be battery powered.
Peter
Thanks Peter! Using your initial schematic, I laid out the breadboard. I think I understand the issue better now. The underlying issue with my assumption that the schematic view and that the breadboard view understand non-polar components such as resistors or non-electrolytic capacitors.
The main issue with my prior board that was causing all the addition odd ratsnest lines is that I often rotated resistors as I would normally BUT Fritzing is very strict about the pin numbers so flipping a resistor or capacitor causes the schematic to put a ratsnest line across the resistor because it wants the original “pin” orientation to be correct exactly like the schematic.
Once I followed the schematic and laid the components in the strict way “pin wise”, the schematic view stayed intact with “routing complete”.
These were excellent instructions and help, I am confident these changes along with the additional part you created will bring this project to life.
Yes that is one of the limitations (as well as the strength) of changes in one view reflecting in to the others. It isn’t necessarily obvious that orientation matters between views on non polarized components. Sounds like you have it straightened out though and that is good!
Peter
Would this be an example of what is needed to power this ?
Sorry to ask but why can I not power it using a wall/D.C. Adapter 12VDC ?
Not exactly, it will only provide one half of what you need. There are a variety of ways to do this, two 12V Wall adapters will work (one to provide +12V and the other to provide -12V) it needs 2 power sockets (which may not be an issue). That may be the easiest and it should work. One of the adapters provides +12V and the other has its +12V lead attached to ground and its ground lead connects to the -12V lead on your breadboard (providing -12V relative to ground). Another option is a regulated 12V wall adapter and a boost regulator module that will take 12V input and produce a 24V regulated output. Ground for the boost module connects to +12V and the 24V output provide the -12V output. This unit from pololu will do it but is quite
expensive at $11US,
an similar unit from ebay is $1 US (but long delivery) such as this one:
In either case you would need to adjust the supply output to be 24 V (as both boost modules are variable voltage). A 5V wall adapter and one of the ebay modules I referenced earlier will work as well with no voltage adjustment required (as the module takes 2.5 to 5V in and outputs ±12V already), the 5V adapter (or 2 AA batteries at 3V) will power the module and it will produce the needed + and - 12V to power your circuit. The downside to it is shipping from China for the module can take up to a month or so via mail, so it isn’t necessarily fast if you are in a hurry and as the module is only $2US overnight delivery would be probably expensive. You could also make a split power supply from a 24 volt center tapped transformer, rectifier, capacitors and 7812/7912 regulators (there are instructions on google for doing so), but you don’t seem to be able to buy a cheap +/-12V wall wart from any supplier I can find. You used to be able to find them surplus sometimes, but even my favorite surplus suppliers don’t have any.
Peter
I see, perhaps then this one!!?
Yes, good find! This is exactly what you want. Didn’t show up on any of the google searchs I did. I’ll have to remember this in case I need one sometime.
Peter
Hi J303
If you fancy building your own split supply you can take a look at this (I’ve designed this layout and built it. I use it for my modular synth):
:12V Rack Power_1.02.fzz (55.9 KB)
It is based on the MFOS project here:
http://musicfromouterspace.com/analogsynth_new/WALLWARTSUPPLY/WALLWARTSUPPLY.php
There’s also some nice schematics on Hexinverters site of all the 909 drums and a few 808 voices too.
http://www.hexinverter.net/neinohnein-kick/
You might also find these parts useful:
RE: spilt power: You can make a battery powered split supply. Here’s a guide on how to do it:
Good luck,
Kevin.
Is there no way to make this work from 9V batteries? How does Eric Archer make the Andromeda Space Rocker? It seems this whole circuit should be able to run off of two 9V batteries.
http://www.ericarcher.net/wp-content/uploads/2014/07/mk4-v2-kit-instructions-0.pdf
Same with this other clone, that is modeled on Eric Archer’s schematic
It runs on +/- power supply, you can run 2 9v batteries in series, or run it off a bipolar (+/-15v or +/-12v power supply). Or… experiment!
As long as it will run on +/-9V (which the second article seems to indicate it will) then 2 9V batteries connected as in noodleDriver’s Split Supply Concept drawing should work just fine. Won’t hurt anything to try it, low voltage won’'t damage anything in the circuit, at most it won’t work if the transistor bias is off far enough (but that’s somewhat unlikely).
Peter
2 Batteries will work just fine and will not hurt anything in your circuit but as they loose charge the voltage will drop and maybe make it sound a bit different ie drop in volume or some minor thing. At some point the voltage of your batteries will droip below (3 or 4 volts) the minimum voltage for some component. I think op amps like TL072 work right down to 3 or 4V then they just stop working (ie you will hear no sounds) at which point you just replace the batteries! This method will work for any circuit that requires a split supply. Of course you could use 2 x 12V batteries but they are not so common unless you use 2 car batteries or something crazy! I have used both methods to test modules (2 x 12V car batteries and also 2 x 9V square batteries. However, it is quite easy to find an old 24VDC wall wart in your attic and use a resistor divider to make a virtual ground then 2 big caps for smoothing.
This page shows the various methods described above:
https://tangentsoft.net/elec/vgrounds.html
Kevin.
Thanks for that link, I really had no idea dual power was a requirement for anything, ever. This all must be done internally in the TR-808 or TR-909s.
I am glad you sent me that page, because I was quite confused by the schematic from Eric Archer, his schematic has some generic capacitors across to ground. I then looked at a PCB built by jmej and the board has two electrolytic caps to ground on the dual voltage.
I see the same pattern from the link you pointed to -> using electrolytic caps
Jmej
https://www.muffwiggler.com/forum/userpix/1043_808bs_sc_2.jpg
In modular synths the power supply will be fixed to the case and shared between all the modules via a distribution bus. There’s no point making a seperate power supply for each module.
The 808 is MAINS powered and has an internal (center-tapped) transformer which converts this down to some useful voltages.
I strongly advise against using a VCT transformer as this is potentially LETHAL! (You could die if you make an error!!!)
Stick with the Adafruit split supply as this takes care of all the deadly bits for you…
When you build enough modules to exceed the 500mA per rail provided by the Adafruit PSU you can either buy another Adafruit supply and keep doing that or buy a nice modular synth case/psu like the Doepfer LC9 or a microZeus by tiptop audio http://www.tiptopaudio.com/zeusmicro.php.
There are a lot of cases to choose from that are factory made but obviously they cost a lot more than building your own.
I built mine for under £20 and it can provide 750mA on each rail (+12V rail and -12V rail)
Hey thank you Noodle, Peter, Oldgrey.
I tinkered a while with this circuit using the 9V batteries in series and an Arduino to provide a trigger from pin 13 so I can see the LED flash when it triggers. It currently produces a loud “click”. Definitely sounds like the initial fast attack pop at the start of the 808 bass kick. From what I read, using 9V batteries will only produce this kind of click. So I am waiting for the Adafruit power supply although the DC/DC power supply from Murata looks very good.
I noticed the batteries got very hot rather quickly and lost charge fast.
From what Eric Archer mentioned, the Accent needs to be tied to (+), I am sending a short 5V pulse from the Arduino.
Here are some references for what is necessary for this circuit to work:
Electro-music.com forum thread about this circuit
Original Web Site (archive of Ericarcher.net)
Here is what Eric Archer wrote about the device:
Trigger and Accent inputs - Each drum has a TRIGGER and ACCENT input. In the 808, the accent input is a global signal that boosts the level of all drums when requested (TTBOMK). In my clone, I just have a potentiometer set up as a voltage divider to deliver a constant voltage to the ACC input. It is basically a volume control in this arrangement. The trigger inputs are a little more tricky; if you are driving the drums with a logic circuit, you’ll need a little bit of trigger generator glue inbetween. Basically this lets the energy of the positive transition thru to trigger the drums, and shaves off the negative edge. If you omit the trigger generator, the drums sound weird b/c they will also trigger on the negative edge, with a different and weaker tone, reduced decay time weird.
Youll need a split power supply to make these circuits work. You can use +/-12VDC or +/-15VDC. In a pinch you can make a +/-9VDC supply with two 9V batteries and a pair of battery snaps. Or you can adapt an old PC power supply to make a +/-12VDC bench power supply. Look for instructions on how to hack a PC supply on the net.
Or do what I did: I found that the miniature muRata NDTD0515C DC-DC converter works nicely, boosting juice from a cheap wallwart up to nice +/- 15VDC bipolar power.
From a forum entry:
i’ve just finished building this circuit and have some results to share, for what they’re worth.
i’ve run the circuit at 9V, 12V, and 15V and can say that the sine/click balance is directly proportional to the source voltage.
at 9V the sound is exactly like the samples posted above where the click was by far the most predominant element.
at 12V the sound is very, very close to authentic.
at 15V the sound is as close to owning an 808 as i’ll ever get!
I will post results once I have the circuit connected to the 12V+/- power supply and I’ll post the breadboard.
I don’t think that should be happening. If you have a multimeter or amp meter measure the current the circuit is taking. Hot and fast charge loss likely indicates you have a short across the power supply somewhere. Op amps don’t usually draw all that much current unless you they are driving a low impedance load and I don’t think this one is so you are likely looking at a wiring error of some kind.
If it likes 15V best, then the Murata (although a bit pricey) is a good bet. A cheaper alternative would be 2 15V regulated wall warts (with the disadvantage of needing 2 power plugs)
Peter
