25 LED Open Ended Sequencer

Thread Starter

Wendy

Joined Mar 24, 2008
23,415
I've been helping another OP on another thread with his LED sequencer. Turns out what he wants/needs is a lot less than what I was originally thinking (no harm there). However, I've started a side project as part of my LEDs, 555s, Flashers, and Light Chasers to have a PCB that is easy to use for a project similar to this. Since this isn't going to be used in the other thread I'm splitting it off.

OK, to start off with, here is the base design...



I've just finished redrawing the schematic in Express in anticipation of laying out the PCB. However, time for bed, so I'll upload it here and proof it tomorrow.

.
 

Attachments

Audioguru

Joined Dec 20, 2007
11,248
1) Why use diodes to feed the bases of the transistors? The outputs of the CD4017 go all the way down to ground and will turn off the transistors much better without the diodes. Maybe you are using the diodes and the missing jumper wires as OR gates to make patterns of LEDs?
2) Why use transistors to drive the LEDs? The output current of a CD4017 driving 1.8V red LEDs with a 9V supply and without any current-limiting resistor is typically 16mA and is typically 13mA driving 3.5V blue LEDs. The max allowed current from an output is 25mA and since only a single output is high at a time then the CD4017 won't get even warm. The diodes can directly drive the LEDs except for the outputs that are used for logic.
 

Thread Starter

Wendy

Joined Mar 24, 2008
23,415
1. Notice the bases are open ended? You can tie more than one base together, to light a LED from more than one sequence position, hence the name open ended.

2. The LEDs are driven with 20ma, which a 4017 can never do.

One of the key concepts with the design is you can leave parts off. Want a 10 LED back and forth (it will require 18 sequences and need 16 base diodes)? Don't use the other 15 transistor/LED slots.

Which reminds me, I didn't leave a provision for adjusting the number of sequences (dagnabit). Back to the drawing board for a minor revision.

***************

OK, Paint version corrected, now for the ExpressSCH version.

+++++++++++++++

New ExpressSch uploaded. Done.
 
Last edited:

SgtWookie

Joined Jul 17, 2007
22,230
One item you may run into trouble with on U2 pin 3 is too many transistors' bases being driven by that output, loading the 4017 output so low that it's not recognized as a logic high by the reset inputs of the other two 4017's.

If you explicitly state that this particular output can't drive more than 1 transistor base, that'll be OK.
 

Thread Starter

Wendy

Joined Mar 24, 2008
23,415
Actually you can drop the base resistor to a lower value to add current to the base. Currently it is set for 2ma. Also, the 1/10 base current rule of thumb is just that, a rule of thumb. It guarantee's saturation, but is not a minimum for saturation.

The 4017 can handle 5ma and still provide a viable logic level. I have verified this.

The design is basically established, remember Fenris's Darlek sweeper?

It's funny, AG is questioning why I bother adding transistor and you're questioning whether there is enough drive for the transistors.
 

SgtWookie

Joined Jul 17, 2007
22,230
OK, what I was looking at months back was a project where the OP (OxboRene) was using a 4017 and a 12v supply; after looking at the datasheets and doing some calculations, I figured that a single base resistor of 3.9k Ohms was about the limit to ensure that you'd get a valid logic high level on the IC's output - which is why I mentioned it. If you had two or more such loads in parallel, the output might not rise high enough to be recognized as a logic high by the other 4017's reset inputs. Oxbo was having problems when he was loading the outputs more than that; once he increased the resistors to 3.9k, that particular problem cleared up. I don't recall who manufactured his IC's.

As I mentioned before, U2 pin 3 is the only output where this would be of any concern.
 
Last edited:

Audioguru

Joined Dec 20, 2007
11,248
It's funny, AG is questioning why I bother adding transistor and you're questioning whether there is enough drive for the transistors.
I asked why transistors and resistors are used when the typical output current of a CD4017 into a 1.8V red LED with no resistors is 16mA.
I stated that "The diodes can directly drive the LEDs except for the outputs that are used for logic".
 

Ron H

Joined Apr 14, 2005
7,063
If you can have more than one LED on at a time, don't you need a separate current limiting resistor for each LED?
Is that what you meant by "adjusting the number of sequences"?
 

Audioguru

Joined Dec 20, 2007
11,248
The display segments on my very cheap clock radio vary their brightness depending on the total amount of current becase the power supply voltage sags with high current.
It has no current-limiting resistors, just the resistance of the tiny power transformer.
 

Thread Starter

Wendy

Joined Mar 24, 2008
23,415
Not more than one at a time, though that could be a problem, but on more than one sequence. Lighting an LED on several sequences (which is needed for patterns such as back/forth lights) is why I included the diodes. If you want to drive multiple LEDs then you have several choices, up the power supply voltage, put them in series, or add a separate leg using the same transistor. Base current becomes an issue with the last choice

Wookie mentioned the drive issue for U2 Pin3. One OP had problems with it, another it worked perfectly the first time. I'm not so sure Oxio used my schematic, as he almost never responded to my posts, while Fenris used the design with mods that would have made the problem worse, and we collaborated closely on the design.

I will be taking measurements on the finished product though, to see how close it is.

Side thought, you can replace the BJTs with small MOSFETs though, I believe there are some with compatible pinouts.
 

nerdegutta

Joined Dec 15, 2009
2,684
Hi Bill.

I've spend some time in the weekend drawing fig 11.4 into Eagle. And now I see that the schematic in this thread is different.

You have added diodes and what seems to be jumper-wires.

Are you planning on having the LEDs and transistors on a different PCB, with the same power-source?

I got a bit confused with the altered design.:)
 

Thread Starter

Wendy

Joined Mar 24, 2008
23,415
The idea is an user can reconfigure it to what ever pattern they want. Back and forth involve connecting to transistors through their repective diodes to more than one sequence output, similar to Figure 11.1. There are several uses for sequencers, by leaving it open ended this design might fill most of the needs.

Wookie has brought up some loading issues it might have. There are small package MOSFETs out there that could drop in replace the BJT transistor (if needed). After I get a PCB I'll be trying various things out, to see what I can get by with.

If you come up with a PCB I may try making it. I'm stalled on the PCB design I'm working on, lack of time. Please include a PDF of the silk screen and PCB. If you go with an older design I'll probably try that too. If you remember I did mention I was going to redesign it a little to allow for more flexability.

If this works out I'll probably make a 2 4017 PCB, then a single 4017. It gets easier as we go down. :D

My boy likes blinky lights too, so I'll pass them on down when I'm finished.

.
 
Last edited:

Ron H

Joined Apr 14, 2005
7,063
How about adding an emitter follower in series with the first diode, to reduce the loading on pin 3? You would have to leave the diode in the circuit to prevent Vbe breakdown. If Vcc were less than 6V, you wouldn't need the diode.
 

SgtWookie

Joined Jul 17, 2007
22,230
If all of the LED driver transistors were replaced with 2N7000 (TO-92) or 2N7002 (SMD; SOT-23) N-ch MOSFETs, the 4017 output pin loading issues would pretty much go away. One could even eliminate the 3.9k Ohm resistors.
 

Thread Starter

Wendy

Joined Mar 24, 2008
23,415
If all of the LED driver transistors were replaced with 2N7000 (TO-92) or 2N7002 (SMD; SOT-23) N-ch MOSFETs, the 4017 output pin loading issues would pretty much go away. One could even eliminate the 3.9k Ohm resistors.
I know, so I will mention them as alternates. I don't know what the part numbers are, but I have a small stock of something similar. I like BJTs because they are available and cheap, and while some have had problems with this design (and I'm not sure they used this design), others have had no problems what so ever, and even increased the loading overall.

How about adding an emitter follower in series with the first diode, to reduce the loading on pin 3? You would have to leave the diode in the circuit to prevent Vbe breakdown. If Vcc were less than 6V, you wouldn't need the diode.
I thought of that, it is one of my favorite kludges. However, the diode in series with the base really needs a pull down resistor, which is why I initially rejected it.

The diodes are only required if you are doing patterns, as in more than one output is connected to the LED. It is a build in OR function. I suppose a couple of grounds could be added next to the open wires to allow for fading.
 
Last edited:
Top