Do-it-yourself pedal building
The DIY forum is for personal projects (things that are not for sale, not in production), info sharing, peer to peer assistance. No backdoor spamming (DIY posts that are actually advertisements for your business). No clones of in-production pedals. If you have concerns or questions, feel free to PM admin. Thanks so much!
Thu Apr 25, 2019 6:37 pm
I typically use a GL5549 LDR (10M off like the 5C3) and then make sure I have a trimmer on the LED and test a few ..
As you can see it's 5$ for one of the 5C3 models .. you can get about 200 5549's for that.. ymmv..
Thu Apr 25, 2019 7:03 pm
Is this the method where you stick the LDR next to a super bright LED in a heat shrink sleeve?
I wouldn't know how to go about testing them, any instructions on how to do so?
Thu Apr 25, 2019 11:15 pm
Yes this is that method. It kind of depends on the circuit from my experiences. I usually start with a diffused green or yellow. Get some 6mm heat shrink, cut off about 0.5 inch and stick them together with it over, heat with a lighter or heatgun etc .. then stick it in the circuit and test. Ultraclear will have a much sharper response, but most circuits are designed so that you won't notice. I've used these as direct replacement for 5C3 in every circuit I've ever built. I've never bought a 5C3 .. I got a couple of those NSL-32s where a circuit said they sounded good but ended up using a 5537 and led as it sounded better. You can still get all the GL 55 series on ebay cheap AF but they aren't all super precise.. I usually throw it in my resistance tester and put it under the edge of my table which is pretty dark and test .. if its like 7-11M .. good enough.
If this is for the flange with no name board, buy that LDR for the daughter board Dino specifies.. . I think it's like 20M dark or something. I bought a handful, but I used 5549 and diffused green for the rest. Sounds great.
Fri Apr 26, 2019 8:14 am
Nice, thank you!
Fri Apr 26, 2019 11:51 am
I've used those (cant remember if the logo was the same but the case was). If it didn't match the characteristics I just found it easier to alter the circuit slightly than to re-source different vactrols.
Sun Apr 28, 2019 9:36 am
NSL-32 are not a perfect replacement for VTL5C3. the VTL5C3 is a bit more consistent. also maybe less obvious is that VTL5C3 is the best replacement for VTL5C3. you can collect 3 dimensional data with the LDR response over time vs temperature or some other data point. this is a hard thing to fake exactly but it might not be something you care about for a compressor. for a buchla LPG clone there is no substitute. I have had usable results by building something like an in-circuit testing jig with trimmers to calibrate both current to the LED and minimum resistance as well as linear vs log curve on the LDR side. minimum ON resistance is achieved by putting a resistor in series to the LDR. the minimum resistance of the LDR is first measured then that value is subtracted from the desired minimum resistance to obtain the value needed for adding a series resistor. the linear vs log curve is calculated by using a parallel resistor calculator like spice to graph all the possible values of the parallel resistor or some more complicated resistor network. we already know the min and max values. all we really do here is find the linear or log curvature that gives us the most ideal real world conditions for the middle value when the LDR is somewhere between OFF and ON. you can also use a trimmer and tune your filter or LFO by ear. the LDR is not perfectly linear with respect to LED control current. a lot of this is just trial and error testing the end user product. I'm pretty good doing all this in my head by now so I don't even waste my time with spice to figure out a parallel resistor on an LDR. best advice to the use of a NSL-32 is to throw half of them in the garbage after testing and sorting.
I have almost zero experience rolling my own from scratch with black electrical tape. YMMV
Sun Apr 28, 2019 4:25 pm
What sort of range do you experiment with in dealing with the LDRs? I have a Doppleganger clone I built, I am trying to get some more intensity out of the HF Span which sounds pretty subtle and I'm curious if I can get some more out of it. The resistors connected directly to the LDR are 220k, I tried 100k and 470k in those positions and heard no difference.
Mon Apr 29, 2019 11:06 pm
there a a few different things you can do. consider what you need and then start with the easiest mod that will get you to that result. adding a resistor in series with the LDR side will increase the minimum resistance. adding current to the LED side will decrease the minimum resistance up to a point until you actually destroy the LED beyond repair. it is also ill advised to sell someone a pedal with an unreliable LED since the current is way too high for reliable operation. actual numbers on this are defined by destroying 10 samples and multiplying the average result by 0.7 (%30 below average over-current failure mode). you must do these failure tests with a high degree of granularity to ensure your over-current failure current numbers are not higher than they actually are. if that happens then your FM*0.7 is also too high as a result. max off resistance is mainly a result of a sorted part. some of them can get to 10M and others only 700K off. the majority are 1.3M to 3M. if your LED driver is leaking any current more than 0 current then you should address that also.
now that you have exhausted everything you can do with all that you move on to examining the circuit that is being controlled. you can usually work backwards from a desired result such as LFO rate, LFO amplitude, modulation depth, filter cutoff frequency etc...
in your case the there is a 10K feedback resistor on the opamp that drives current to the HF span LED drivers through a 1K HF span pot. you only need to increase this with a 10K trimmer in series to the 10K resistor. you lift one leg of that resistor and you insert your 10K trimmer. then the other side of the 10K trimmer goes into the empty pad on the PCB. you just use one outside trimmer pin and one middle trimmer pin. it makes no difference what side connects to what since a resistor is not polarized. keep in mind that you can totally poop your LED's beyond repair. VTL5C3 are $10 each so this is a dangerous game you are playing. experts only.
I will also say that LDR's and obviously vactrols that contain LDR's have a lag time and the depth of response will decrease as the LFO frequency increases. a vactrol will respond best to a LFO frequency that is close to the ideal theoretical 0Hz+(1/infinity). there are also other resistors to play with in that circuit but there may be negative side affects with sound, usability, reliability etc.. YMMV
Powered by phpBB © phpBB Group.
phpBB Mobile / SEO by Artodia.