i did not! lol - fuck. this poor board…
1 Like
this is why we prototype things… sometimes things aren’t as successful right away haha.
Really glad you’res is working @ChristianBloch
1 Like
Still weird – I mean, for a red LED, that’s almost 40 mA. They’re usually rated for half that.
That’s maybe because of the huge base resistor on the transistor: let’s say the opamp output is at 12V. Minus the led+resistor+junction voltage (~2V), we have 10V across the 100k resistor, so that’s 100uA in the base. Even with a big Hfe of 100, that’s only 10mA in the LED, so the 270 is useless here (but that’s 100mW in that poor transistor).
By the way, why on every circuit I see the NPN transistors are on top, collector on the supply rail, and not on the bottom, emitter on the ground? Is it to have some sort of threshold before the led lights up?
1 Like
Well, even if we ignore that relying on precise hFE is a bad design pattern, he’s specifying a 550C which is the max gain type (hFE 420-800). But it’s also also a low-noise selection, which seems a bit pointless here, so not sure there’s a lot of thought put into the component choices here 
But I suspect this is more that it’s just a pulse, and he ramped up the the mAs to make it more visible, instead of doing a bit of pulse stretching.
1 Like
Yes of course, I clearly do not recommend to rely on the Hfe, I was just trying to explain what’s going on.
The datasheet specify a Hfe up to 800 indeed, but also shows this:
So I don’t really know which one to trust…
Yep, that must be that!
2 Likes
Given the complexity and number of components for the S&H section of the schematic I wonder why Niklas didn’t go for the S&H published by R. Schmiz:
4 Likes
I’m at a point where it’s working well, not just working . Here are the changes I’ve made and some notes.
- You have the rate and range pots labeled wrong. Rate is the top one, range is the bottom. Luckily, I’m using big and huge knobs, so the labels are obscured.
- The bottom pot is wired backwards, so I switched pin one and three using wire. That makes it go from no range to max range
- A C1M pot for the top works much better and feels natural
- I only get enough voltage for a red LED, all others don’t or only barely light up
Pics:
5 Likes
What is a C pot? Exponential?
3 Likes
I’m getting nowhere near 12V from the transistor. 12 V is going into it, though.
2 Likes
anti-log Potentiometer
1 Like
Reverse Log. So it’s easier to fine tune at the high end. Before the hold was either too fast, one good spot, then too slow. At least for the way I like using it where it’s like 16th notes at 130-ish BPM.
1 Like
lmao that was like at the exact same time.
1 Like
Log is fine tune at the start.
2 Likes
BTW, they’re really exponential, not logarithmic, but oh well.
2 Likes
What transistor are you using?
2N3904… Body seems unclear, is it a complete sentence?
1 Like
The 3904 has a relatively moderate gain, at least 2-3 times lower than the 550C used in the schematics (which as noted above is the highest gain class for the 54x series). Which is probably a good thing here
(Transistors vary enough that you usually want to design for the minimum gain, rather than hoping that you didn’t get one with maximum gain. This is done e.g. by using feedback, or for transistors used as switches by making sure they saturate also at minimum gain.)
2 Likes
Makes sense. BTW, I realized that if I wired the rate (top) potentiometer in reverse, I could just use a regular A taper AND go from slow to fast rather than the other way around. Probably the way to go, but I’m fine with the way it’s wired now. May decide to switch out the LED transistor and go back to my beloved yellow flat top, though 
.
1 Like