mgronroos said:
Hey Taliska,
I understand your frustration in not liking your mic! :-[
However, what you describe does not sound like my build at all, and there are several others here who have liked their U47fet builds.
I have compared it to several other good mics I have (Neumann, Manley and my C12, M251 builds etc.), and I think I would have noticed if it would have "horrible mids".
Maybe you got a bad RK-47 capsule? Or perhaps the wrong value in a component? Maybe something got overheated? Contamination? Ground issue?
These are just suggestions that you might go over the circuit again and double check for errors before you discard this (in my opinion) wonderful mic!
I am actually so fond of it that I am building another one right now, so I strongly suggest that you give it another try..
BTW: Did you make the self bias (I did) or the a bit more tricky one?
Let us know what you find!
Best,
Michael
Hi Michael,
Thanks for the support!
I think it probably comes down to expectations vs personal preference quite a bit. I'm not saying it's completely unusable, but just falls outside what I would call useful based on the way I would like to work which is as follows (I'm a hobbyist so my techniques are a work in progress):
- Try to get the source sounding roughly correct on the way in (we're not talking about nailing it, just ballpark, generally with a max of a couple of tweaked eq bands on the way in).
- Once ITB, "nudge" the sound in the right direction with a little more eq to make it work in the context of the arrangement.
- The idea is basically sound refinement by a few minor tweaks, rather than taking big leaps.
Now I have nothing against "coloured" mics, but super coloured mics are almost always going to fall down on that first step. I'm probably always going to choose something else if the frequency response is too biased, which is where I'm at with the RK47.
I don't have a huge mic collection, but I'm trying to cover the bases slowly. All my points of reference are first-hand store bought specimens. I've got a K2, C214, RE20, and a SM58 to compare it to. I like to experiment with various styles, and I've got a selection of (cover) tracks that I've set up projects for in Logic that I try all those mics on for the sake of comparison and they (my other mics) all basically work without having to try too hard (lost kitten-metric, since i fell for you-nina simone, walkaways-counting crows, wicked game-chris isaac). I can honestly say that the K2/C214/RE20 are all awesome and are all equally likely to be chosen as the best fit for a particular song even though they're quite different sounding (K2 I would say is dark-ish, C214 is crisp, RE20 is somewhere in the middle plus has that whole different sound due to how dynamics handle transients).
I also did my first power up with the stock capsule and the sound was actually quite usable apart from the slight raspiness (I'm referring to sound quality here, not frequency response) at the top end which I expect to hear with cheap capsules.
Today I've tried the RK47 on acoustic guitar and the problem I have with it is this: I can actually make it sound quite nice once I'm ITB...where it fails is just at a common sense level. If I'm having to try too hard to make it sound good, then it's probably not a good choice. Again, I managed to get it sounding pretty nice, but I don't expect to have to be so surgical with my eq for a mic that's capturing ordinary sources in ordinary situations.
Oh and yeah, I suppose I should elaborate on horrible mids. That's what I thought I was hearing, but after more investigation today, I'd say it's got more to do with some pretty strong roll-off from 5khz upwards. In fact, what I'm hearing is pretty much exactly the graph they have on their site. I've attached it for reference. The rk47 might work better in a circuit that exaggerates the high frequencies, but I don't feel that the d-47 is that kind of circuit (just because I've never read that complaint in this thread).
At this point I'm pretty confident in my building abilities and the reasonable response from the stock capsule makes me think that things are working as they should. Check my post history for context, but I've been getting my hands fairly dirty for a while now, and while I'm lacking a bit in terms of the theoretical aspects of being a solder monkey, I don't think I've made a mistake in the build, especially as things tally with the mparts graph.
Anyway, I'm glad that you're getting on with your RK47 ok.
Apologies if I came off a little bit harsh in my previous post...I'm not trying to get anyone in trouble, but I also can't see other people finding this capsule particularly usable in this circuit and I wanted to provide another viewpoint to balance out your previous enthusiastic post (it was actually your post that motivated me to try it out...before that I was going to avoid it as that guy posted samples compared with a dale m7, which had put me off getting it).
To sum up, the RK47 didn't sound like I expected it to, but it's entirely my own fault. It matches the graph they publish, but I just didn't think it would be that extreme. As always, this is just my opinion and other people should feel free to make up their own minds!
Cheers,
Kaz
P.S. Here's a link to some screenshots of eq settings in logic that I had to use to push the raw recordings where I wanted them. For these I didn't use any eq on the way in, as the rk47's freq response was something of a mystery at the time. For anyone not familiar with the UI, for each track (vocal, guitar) I've included two images (pre-eq-curve, post-eq-curve). The eq settings are the same between pre and post images, but different between vocal and guitar images for obvious reasons. It's the frequency distributions at the bottom of the images that you should compare. The pre and post screenshots weren't taken at corresponding moments in time, so shouldn't be compared precisely, but rather just used to examine the trends.
https://www.dropbox.com/sh/9878lr9aa22tfvy/AADBsy8c-hhuhHhLcQjzhlN4a?dl=0