AKAI CM-15 Tube Lipstick style Tube Condensors 6AU6A NOT 6AU5A as previous

GroupDIY Audio Forum

Help Support GroupDIY Audio Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.

tardishead

Well-known member
Joined
Aug 11, 2004
Messages
623
I have had a pair of these with PSU for a couple of years now and I have struggled for a long time to get them properly working. I decided to fully learn what was going on with the circuit so took a couple of hours to investigate. There is absolutely no information on the web so I thought I would post my findings maybe to help others.

They are a very simple design triode strapped 6AU5a (SORRY THIS IS A MISTAKE - THE TUBE IS 6AU6A ) cathode follower output like Altec Coke Bottle. They have no output transformer - I will install a couple of 1:1 600r:600r transformers to provide balanced outs.

IMGP1439.jpg

IMGP1438.jpg


The cables needed resoldering and a couple of connections in the mics themselves were bad.
The power supply is very neat and has really good filtering for HT and DC heaters. Very good quality components. It says PREAMPLIFIER but it is a power supply.

I traced out the basic audio circuit.
The output connections are very strange - has anyone seen this before?? The output is totally unbalanced. I presume the only way to balance it is to get rid of the 30k and 1m resistors and add a transformer.

sc007d6a79.jpg
 
Good stuff! Can we see some inside photos?

Are you sure about the output connection? I've seen plenty of unbalanced tube mics, but nothing quite like that. Do + and - relate to different pins on those connectors? Presumably the connectors have 0V as well.

You could balance these - something like 2:1 to 4:1. I'd start with whatever you have in the spares box!
 
Slight step-down might be preferable like zebra suggests. I seem to remember discussing how this was beneficial for the C37 design - it's well documented how cathode-followers don't actually work that well driving low impedance loads. 
 
That is how both of the output connectors were wired - very strange I dont get it. They are not xlr but some other 3 pin.

I did have step down 10k:600r wired to the output connectors but I was losing a lot of HF when connected to mic pre. I took apart the chassis and thats when I saw those strange resistors across the outputs.

Originally I also tried to connect straight to my desk - no transformer. I had a big hum. Worked best on line input but needed + 20db and was not enough gain. At this point the response was good enough bass and treble but unusable because of the hum etc.

Now I know about them I will get rid of these resistors and try the 4:1 transformer again straight to mic pre. I cant see why I would need to keep them. Maybe they were for another purpose???? I bought the mics from an ethnographer who recorded with a reel to reel in the late 60s 70s.

With such a simple circuit there is no reason I cant get them to sound good. Thats why I took the time to trace the circuit.
 
You don't want to load these with less than 10K, from "+" to "Ground".

I'd really consider building a higher-Z buffer, like these are surely intended to drive (~~100K hi-Z mike in). 5532 into 600:600 is a thought. Unity-gain may be ample.

They may just not be great mikes.
 
Thanks PRR

Can someone explain to me what is the difference between this and the AKG C60 circuit.
C60 has a AC701 tube - 180m grid resistor - 1k5 connected to cathode and 82k connected between grid resistor and cathode resistor to ground. Very similar circuit. Main difference seems to be the grid resistor C61 has 180M where as Akai have 10m.

C60 has a 1:1 output transformer I think.

What is the difference between the tubes AC701 has a lower output impedance????
 
tardishead said:
C60 has a 1:1 output transformer I think.

Not necessarily. I would have guessed they would have factored in some step-down.

I think something between 2:1 and 4:1 would suit; that'll get 1K preamps loads up to around 10K if it were a 3:1.

The output impedance is driven down by the fact that it's a cathode-follower configuration. However, the principle itself fails if you try and drive low impedances since this affects the process of negative feedback.

I think these links may help:

http://www.tubes.mynetcologne.de/roehren/daten/as_triode_overview_e.pdf

http://www.tubes.mynetcologne.de/roehren/daten/ef94pentode_as_triode.gif

Page 235 here:  http://www.pmillett.com/Books/Atwood/Crowhurst%20Cooper%201956%20High%20Fidelity%20Circuit%20Design.pdf
 
Yeh I took away those resistors on the output and added a 10k:600 transformer. They sound much better now. In fact they sound great and compare well to my KM54 and Gefell M582s. Not as smooth but a bit less coloured and very punchy and direct.
They suffer from a bit of noise but they are definitely usable especially on loud sources. No hum whatsoever.

The grid resistors have gone out of spec a little. One reads 18MEG and the other 13MEG. The one which has drifted more is a bit louder. I will put in some new 10 MEG resistors - will this potentially reduce noise?
The other resistors in the audio circuit are wirewound and seem very close to spec.

I presume someone would have noticed that this circuit is VERY similar to Sony C37 - same tube - only real difference is the much larger grid resistor 180M instead of 10M on the Akai mics. I presume this is to tailor to a different capsule - is that right????
 
> TUBE IS A 6AU6A

Well, yeah, I assumed that.... 6AU5 is a small truck, would not even fit in the case.

> AC701

Has VERY low leakage and VERY low self-noise.

6AU6 is your all-purpose pentode. FM radio and low-cost TV filler. Excellent tube, many millions were made, readily available and affordable.

> lower output impedance?

'701 may be higher. Smaller tube, and optimized for low leakage rather than high conductance.

Not really important. Cathode impedance is bound to be around 1K; best-load considerably higher. So we can't drive low-Z lines without a transformer. Transformerless, any CF will happily drive hi-Z (>20K) loads fine.

Roberts/Akai was selling semi-pro tape recorders for advanced amateurs and small professionals. Think TEAC/Tascam reel decks. Mike inputs were 1/4" hi-Z (>50K 5mV). These work with crystal mikes, dynamics with inboard or in-line transformers, or if you had more money, Akai offered these condensers.

> difference between this and the AKG C60
> this circuit is VERY similar to Sony C37


It's a basic self-bias cathode follower. There's several ways to skin this cat; this form is simple and does the job, whether the goal is quality or price.

> much larger grid resistor 180M instead of 10M on the Akai

That's one thing which makes me think this mike is not the very-very best. Any condenser head I've seen, 10Meg will limit bass extension. Possibly not reach 100Hz! Yes, maybe the capsule booms and they wanted bass-cut. But also the capsule bypasses the grid resistor and the grid-current hiss. "Usually" the optimum for a low-current grid is 100Meg-200Meg. The 6AU6 is not particularly low grid current, so it may like lower. But probably not as low as 10Meg. Where did that come from? Back in the day, low-price resistors did not come higher than 10Meg. You could get higher; apparently AKAI did not go to the effort.

Before you go crazy with part-tossing, do some basic sanity-check. What are the plate and cathode voltages? The ratio may suggest that one mike has a problem. The cathode voltage is also the capsule bias, and a large difference between mikes will give sensitivity difference, possibly getting into noise.

The resistors may have drifted, which suggests Carbon Composition, which has excess noise (and cheap production). Now that initial price is forgotten and you've broken the long-expired warranty, you may as well use something better.

DigiKey has metal-film half-hundred Meg resistors 5 for 17 bucks: VR68J68MTB ...Get these, plus a handful of 10Meg just in case AKAI did know what they were doing.

Get a couple spare 6AU6. Burn them in for several days before you make any tests or judgements: tubes idle for 40 years need to have the gas stirred-up and gettered-out. You may, in the end, come back to the AKAI 6AU6... they are probably good, and AKAI may have had access to some especially low-leakage 6AU6 (maybe "rejects" from a high-current product). But we need to know if tube-swap makes any difference.

I suppose you have idled these mikes for many hours. Capsule hiss may be moisture. 6AU6 heat will drive it out. You may have to *gently* wash capsule insulation, but get an expert eye to tell you, don't ruin these capsules with untutored cleaning.
 
> much larger grid resistor 180M instead of 10M on the Akai

That's one thing which makes me think this mike is not the very-very best. Any condenser head I've seen, 10Meg will limit bass extension. Possibly not reach 100Hz! Yes, maybe the capsule booms and they wanted bass-cut. But also the capsule bypasses the grid resistor and the grid-current hiss. "Usually" the optimum for a low-current grid is 100Meg-200Meg. The 6AU6 is not particularly low grid current, so it may like lower. But probably not as low as 10Meg. Where did that come from? Back in the day, low-price resistors did not come higher than 10Meg. You could get higher; apparently AKAI did not go to the effort.

There is definitely no lack of bass - well below 100hz. In fact you might call the sound a little boomy - the one mic whose grid resistor has drifted more sounds slightly boomier than the other. It would seem as the resistors have grown bigger the capsules have become a bit boomier. I will get 10m resistors and some bigger to try and tailor the frequency response.

Sure enough the hiss gets better the longer the mics were powered.
It is very humid here in Sydney.

 
RDH4, chapter 7, page 322, fig 7.10
The input resistor value is multiplied

I like to cycle 6.3VDC tube heaters at 6.5VDC for a few hours then at a underheated setting like 5.7VDC for a few hours then at 6.5VDC again for a few hours and 5.7VDC for a few hours.  I read about this at Steve Bench's site IIRC.

Check you heater voltage for VDC and ripple.  You might try running the 6au6 tubes at 5.8VDC(lowers the grid current but you still need some tube current to drive the 4:1) after you run them for more hours.

Measure the B+ and the nodes of the cathode string to ground(cathode and 1.8K,100K node) this will give more information that might help.  Things like cathode current and charge voltage to the tube.

I like this type of circuit in tube microphones(I also like plate out circuits (taste)) it is direct coupled to the capsule and one can adjust the charge voltage to the capsule.  A good capsule and a nice transformer this type circuit can sound good.  Say you want a microphone that will be used at a high SPL adjust the capsule voltage and tube operation points maybe even use a higher than 4:1 output transformer(think of it like a pad).
 
> input resistor value is multiplied

D'oh! Well, then bass is fine. Gonna take more coffee to think if grid current noise is multiplied or bypassed.
 

Latest posts

Back
Top