DiY AMI U47

[Melodeath00]

jb, are you sure that's carbon comp? Specifically I'm looking for 60M, 100M, and 150M. Everything else I can find pretty easily.

 

[gary o]

Melodeath00 I have used 5X 50M metal resistors in series in a Elam250 circuit seems to make no difference to my humble ears anyway...... also not everyone believes the carbon resistor mojo, my ears believe it but they are said to distort in a nice way under high voltages so only worth putting in certain positions in certain circuits and its known that they are more noisey... & in low level of a mic may be a bad thing...have a google about carbon resistor mojo its very interesting but I dont think much use for mics nowdays.

Your capacitor choices will make the mic sound different for sure tho....

All the best 

 

[Melodeath00]

Yes, I've done a bit of googling on this "mojo" thing. I don't know if it's true or not (I tend to be a cynic and not believe that kind of stuff), but I decided that since I am planning to swap out caps and resistor values anyway, I might as well try carbon comp as well. I am swapping my 1G resistors to the Neumann-correct values, and swapping the output coupling cap to PIO and .5uF

I'd also like to install this switch or relay so I can get true cardioid. Looking for more advice on that, too.

 

[gary o]

I think its great that we can tweak our DIY builds, I like PIO caps & others too ..... I have switches on my DIY so I can swap capacitor values I am also going thru the change the 1 gigs back down to original or at least lower than 1 gig values still learning lots great fun....

 

[Melodeath00]

Let me know what you think of those changes. If my 100M/60M sounds worse than my 1G/1G edition, I will be upset! haha

It's not going to be all that easy to make these mods. Desoldering with a wick is a nightmare for a newbie like me.

 

[gary o]

Well you  may prefer the 1 gig its subjective......is the AMI a circuit board if so be careful not to ruin it by de soldering..... are you seeking the U47 sound if so original resistor values would be best but then as you say you dont have neumann capsule tran & V14 so you gonna end up with a U47esque mic should be a dam good one & you can tweak to taste..... I built many mics now & it really does come down to the capsule long as transformer is good i find the grid resistor is important but to my ears the resistors caps grill & everything else is like seasoning salt & pepper maybe a bit of sauce on your main meal (capsule) have fun.

 

[Melodeath00]

The AMI board is just a board with metal pins on it. I'm not sure what you call it, but it's not a PCB. It's point to point. There's two of them, and they are placed on the two sides of the transformer.

When I built the mic I was shooting for as close to U47 as possible. Of course, that is extremely hard to judge since I have never used a real U47. I am not at all displeased with the mic as is - It sounds very, very good. It is more that I am curious how these historic value changes will impact the sound. In theory, lowering the high impedance resistors and the output cap will change the sound by lessening the bass response, but I'd sort of like to see/hear it for myself.

 

[o3misha]

I have two microphones U47 with VF14. When I tried different types of resistors, I found that only the size of 60 to 100 megohms satisfied me. At a value of more than 100 megohm resistor amount of bass was so much that I felt lack of intelligibility in the upper part of the spectrum . And changing the capsule does not change anything, feeling too unwieldy sound remains. I can tell you that in U47 there are no unimportant components . Influence of grid resistor is huge.

 

[gary o]

Two U47s two V14s !! ..... green with envy, are the grid resistors carbon ? what type of cap is C2 .... i read that makes a huge difference to the sound tho what one person calls huge may not be same for others.... Its seems a trend with DIY and china mics copying classic circuits to use 1gig resitors plenty of bass too much kinda muddies up the higher freqencies, prob a good thing in china mic with peaky capsule..... I only had 1gig resistors for a while for my DIY mics I have been re trying the circuits that I made with orignal values, its better more clarity I can still easily pick out capsule Im using tho but its better.... I have Maxes MK47 kit but its always been a little noisy ( white noise) I re built it point to point and that made it quieter ( prob had dirty board) but its still noisier than all my other DIY mics I know the circuit its sensitive to PSU noise so I guess could be that, tried loads of tubes, I read that the higher value grid resistors eefect noise but I have the 100 Megs in there.

All the best

 

[Melodeath00]

Thanks guys. We will find out if the 1G is "necessary" when using EF800/EF802 instead of the VF14. Perhaps the VF14 has more bass to begin with?

Any opinions on using 1/2 Watt versus 1/4 Watt resistors? The AMI schematic uses 1/2 W for everything except the pattern control line, but typical Neumann Schematics vary between 1/4 Watt and 0.1(?) Watt.

 

[o3misha]

When the tube is underheated, the frequency response is not linear. You have sluggish and sedentary bottom. And the amount of bass can be thus perfectly normal , and a completely different dynamic than when the lamp is operating normally . Large grid resistor can flatten the frequency range , but with sluggish bottom middle it can be an unpleasant surprise. That's what I did not like. I believe that the grid resistor also has an impact on the grid current , which for small values ​​of the bias on the grid ( -1.1VDC) may increase with the big grid resistor. With increasing bias (  -1.6  -2 VDC) influence of the grid resistor decreases. For example, in U67 we can use larger grid resistor. So , with 60M-100M U47-ish mic with underheated tube works more stable and has less distortion, IMO.

 

[Melodeath00]

Interesting points....

Coincidentally, I just saw that Oliver posted yesterday some information.

"the tube with the least amount of gain has the highest input impedance, at least that is the theoretical side"

and

"Because nearly all tubes for microphone application are operated outside their factory specs. and yes, each and every tube can vary, it is mainly the grid current that has minute differences
in tone."

Now, thinking out loud here...

I'm not sure exactly what he's saying, but it's worth noting the EF80-type tubes have more gain than a VF14, and thus, apparently a lower input impedance. Maybe if you're using a higher gain tube like the EF802 it makes sense to use a grid resistor higher than 60M to lower the grid current? I'm not sure. Maybe I'm misunderstanding and that's not what a grid resistor does. The grid resistor does seem to be in parallel with the input of the tube, so that would increase tube input impedance. Would that also lower the grid current?

 

[AusTex64]

Two advantages of tube underheating are lower noise and grid higher impedance.

 

[o3misha]

You are right, but underheating may cause increasing noise as time goes by. Its obvious problem for VF14s for example. Also, the emission is less (not good, BTW). Increasing of  impedance of the grid affects on changes in low frequencies . I'd descrbe it as: low freqs became sluggish .

 

[chaserose1]

Hello,

I am happy to report that I now have a AMI U47 that is working well after such a bad outcome with the first attempt, it did require a complete do over/rebuild.  Voltages are correct using the alternative circuit component selection.  It is a little noisy however so I will attempt using different tubes to see if that will improve.  It does have a nice low end which is what I was hoping to achieve.  Thanks for all the info on this forum subject, I doubt if I would have ever completed the build successfully without information shared here. Thanks guys/gals.

Sorry for the topic change- just wanted to report and say thanks

Chase   

 

[AusTex64]

Chase,

Did you clean everything possible with 90% isopropyl alcohol? PCB, resistors, acrylic decks, etc? Any bit of dirt or skin oil will manifest as noise.

And congrats on a successful build!

Robert in Austin

 

[Melodeath00]

Hey Robert, is that Relay you linked to going to work with a heater of 5.6V? It was rated for 6V. I looked inside the AMI PSU and I saw no easy way to adjust the filament voltage (wanted to try going down to 5V, but also not sure fi I need to go up to 6V for using the relay). Is it a "latching" type?

 

[AusTex64]

Not sure, but usually there is a range of voltage that will actuate those relays. I think there's a data sheet on the Mouser site, have a look at that. It is not a latching relay.

 

[Melodeath00]

Thanks for the reply. Is your AMI PSU also at 5.6V heater?

 

[AusTex64]

I believe so. It's been a while since my build, it's buttoned up and I'm using it regularly. I think Oliver is a fan of underheating slightly, for the reasons stated in my previous post.