acm 1200 circuit and mods (similar to apex 460)

[alexc]

So I finally began to do my tnc  acm-1200 mic mods as the mic fever finally got the better of me.

Actually, I subbed the sino tube it had for a sovtek ax7 and found so much noise as to make it unusable.
So I decided to take a closer look.

Most of what I have read says this mic is the same as the apex 460 with some switches for 10dB pad and a HP filter.

----------

As I worked through the circuit, I confirmed the reported similarity to the apex 460, with a caveat that the inclusion of the HP filter changes the nature of the 'constant current draw amplifier' as most people name the apex 460 amp circuit.

In the tnc acm-1200 mic, the HP filter is implemented as a 100nF poly cap inline between the plate output of the 'grounded cathode amplifier' first stage and the grid of the 'cathode follower' second stage. It also has a 510K resistor to ground at the grid of the CF. The HP filter switch adds a 22nF cap in parallel or not.

This arrangment of course blocks the dc to the grid of the CF changing the bias of the CF stage completely, with respect to the apex 460 circuit. I also found the CF in the acm-1200 has a cathode resistance of 47K vs the 270K in the apex 460 circuit.

Which explains what I see when doing general testing of the acm-1200 : the CF stage is biased real bad

I was seeing something like 146Vdc on the CF plate and some 10Vdc on the cathode and 0Vdc at the grid.

Meaning some 0.2mA current or so and an output signal swing of some 18vpp before severe cutoff of negative peaks.
The thing is almost switched off!

Unless I am seeing something wrong, it seems to me that the biasing is totally crap.

note also the grounding in these mics is very poor, being propagated about the mic via the internal rails and very hit and miss tinned screw pads with easily stripped screws making a poor contact (high resistance to ground) guarranteed !

The pad is implemented as a switched shorting of a 68K resistance in series with a 30K resistance in the plate circuit of the 'ccda' first stage. With the pad switched, the plate resistance becomes 30K vs 96K when not switched.

--------

I think the apex 460 'ccda' circuit is quite a good one, even though many complain it is not.
I've used that topology many times in other amp circuits and it is a favourite of mine.

But the inclusion of the HP filter, implemented as a series capacitance/shunt resistance in between the 2 stages of the 'ccda' renders the whole thing stupid!

I know the mic head amp is not required to provide large  voltage swings but 18vpp before clipping into a 11:1 stepdown transformer does not seem right to me.

Especially when the modified 'ccda' in the apex 460 is  such a high performance amp - high input Z, very low output Z, reasonably high gain (x25 or so), good drive capability with flat frequency response from 20Hz  and -3dB at >20KHz.

------

So in addition to the other mods usually discussed for the apex 460, I'm restoring the amp circuit to the apex config as well : complete removal of the pad and HP filter circuits and restore the CF to dc bias.(leaving the pad would cause too high a grid voltage at the CF, when the pad is engaged - not safe)

Once I do that, I'll check it all again and decide whether to lift the B+ and stage currents some. At present, I have a 12AY7 in there and I think it is quite starved of current the way it's biased right now.

-----

Common apex mods mods that I am doing are :

- change tube from sino 12AX7 with non sino 12AY7
- replace capsule with the best you can afford
- remove output transformer secondary 'rf caps' to prevent peaking response/distortion
- replace 'ccda' input grid cap with 1000pF polystyrene, 'ccda' output cap with 1uF metal poly
- replace 'ccda' first stage cathode bypass cap with 22uF electro (down from 100uF)
- rebias 'ccda' with first stage cathode resistance of 1K8 (down from 2K7)

and possibly

- add a 250pF 'eq cap' from 'ccda' first stage plate to ground
- add a 220K shunt resistance across output transformer primary to damp resonance of stock traffo

I won't attempt the head basket grill removal - I'm sure I'll ruin it if I try

I am keeping the stock apex output transformer - I'm seeing it perform quite well in my measurements. Should I see anything really bad in phase response or distortion, I'll reconsider. 

So far it is just fine.

------------

I'll report back when I do all the mods.

Cheers

 

[HellfireStudios]

You can increase the 1uF transformer coupling cap to allow more bass through. Some of the original C12's had higher values in this position most likely to tune each mic properly. Go with your ears and instincts on this one.
Depending on the capsule you intend to use (C12 pr K67) the EQ cap can be any where from 100pF to 500 pF. The ELA M 251 uses a 100pF to smooth the high end with the C12 capsule. I've also seen a 500pF used to smooth K-67 style capsules. Again it is a taste issue. So, smooth to taste.
The stock transformer, from what I've read, isn't all that bad. Not the best, but not bad either. How did you determine it had an unwanted resonance?

-James-

 

[alexc]

Comment about unwanted resonance in the stock apex 460 transformer created by the RF prevention caps was from 'net research'  - a lot of info out there on the subject.

I did measure the freq response of the stock apex transformer with rf caps removed and a 2.2uF poly output cap, when operated with very low input signals, before the incorrectly biased CF clips : ruler flat 20Hz - 15KHz then rolling off -3dB to 20KHz. (-10dBu input, capsule removed)

I'm not too concerned with the C12 configuration - never used one personally.
I don't see the single stage 'grounded cathode amp' with second stage unused as a very good idea for this application, so I'm just focussed on getting a circuit which works as I expect.


Just finished removing all the 'ccda trash'. Testing again now.

 

[HellfireStudios]

I wonder why Advanced Audio and Fox Audio Research offer options to keep the original transformer?In fact, they seem to find them very usable (not that it matters much to me, I upgrade my transformers). It seems like your test didn't show any resonances in the transformer either

BTW, the Apex 460 circuit is very similar to one side of the stereo AKG C24 (with the addition of CCDA). So the C12 style capsule is a good match for the circuit, unlike the harsh K67 style capsule that resides in it stock. Also, changing the transformer to a higher ratio (10:1-14:1) negates the need for the CCDA part of the circuit.

-James-

 

[alexc]

I have my next baseline in the rebias of the apex 460 'ccda' amp with 12AY7 tube :

(capsule removed, feeding signal to grounded cathode amp grid)

I adjusted the psu to provide the higher currents - went from 60.4K of hv resistance to 27.1K
for a total final current of 4.4mA at a B+ 147.2Vdc  (should be 150V with the 2 series zeners)

grounded cathode amp : B+ 147.2V, plate 72.7V, bias 1.26V, current 0.7mA, gain x18.5
cathode follower : B+ 147.2V, cathode 73.1V, cathode resistance 41K, current 1.78mA, gain x0.95

Still tweaking the bias at these B+ settings. Seems to me the CF is running a little off kilter still, but in the ball park.

I can now get the CF to swing fully without clipping, with the gca stage the limiting point due to it's bias of 1.26V.

Not sure what the voltage swing of a 34mm K67 style capsule is, but I'll figure that part out shortly.
Along with polarisation issues - the variable pattern attenuator goes from 0V..B+with no capsule connected.

I'm connecting up the original capsule for further testing. Then I'll decide on my next baseline bias points and go from there.

I think it will be necessary to provide more headroom at the first stage by increasing the gca bias to around 2V or so, cooler than the current 1.26V.

Easiest way to do this is to increase B+ to around 200V. Will also have the advantage of allowing a cooler bias on the CF which should decrease distortion. It's all a little uncomfortably low headroom right now.

Cheers

 

[alexc]

I only tested the stock transformer *after* removing those rf caps, finding it very flat.

Other issues people have noted with the transformer are higher distortion
(particularly in the region of the resonant peak caused by those rf caps)

compared to other well regarded models.

It's all a question of price and performance. I just want to keep costs down on this one and save my bucks for a full build from scratch.

As for the question, 'to ccda or not to ccda' - everyone has an opinion, I guess.
I like. Many don't. I see it as an elegant solution to the application, not something to negate.

But this is my first mic mod - so what do I know.

 

[alexc]

With the original capsule in, some quick measurements of voltages at the plate of the grounded cathode amp stage, gives around 6vpp quite easily and strong peaks at maybe 10vpp

At x18.5 gain, that would indicate a capsule output level of around 320mVpp easy, 540mVpp peak.
For a first approximation, that is

So, that would indicate that the current bias of 1.26V is fine.

It would also indicate that the CF biasing does work as I found it - with a first stage output of 10vpp, the CF biased as was will work. But I don't like it. No sir!

---------------

Capsule in polarisation voltages seem to be variable from 23.2Vdc (fig 8) down to 12.0Vdc (card) down to 0Vdc (omni). Not sure if that is correct. This is measured at the wire to the front diaphragm of the capsule.

--------

First  listening test with the mic pre in my Motu 828MkII test bench interface is :

G O O D  ;D ;D ;D

A lot of  the noise and hiss is gone. I mean it was completely unusable before.
The first time I put in front of someone, I had to take it down asap and do all of this.
Could have been due to a noisy ax7 - not sure. But as before - 'no sir - I don't like it!'

So - now to think about what I've done, do some more tests and await the new capsule.

Cheers

 

[pasarski]

---------------

Capsule in polarisation voltages seem to be variable from 23.2Vdc (fig 8) down to 12.0Vdc (card) down to 0Vdc (omni). Not sure if that is correct. This is measured at the wire to the front diaphragm of the capsule.

--------

You can't measure there at he high impedance area with a normal DMM. But you can calculate it easily (capsule polarization draws about 0 A). 147 fig 8, 74 cardioid, 0 omni.

 

[tchgtr]

Nice thread AlexC,
  I have a Nady 1050 in which I have kept the original circuit, and transformer, and installed a C12 clone. It's a nice mic. Could use  a better capsule, but I like to keep costs down also, and am enjoying your info.
Following with interest.

 

[alexc]

Thanks  pasarki for the info and tchgtr with the interest

I'm doing some reading regarding capsule polarisation schemes and so on now.

I spent the evening doing some acoustic gtr and voice comparisons with the mic+mods+original capsule vs a sdc mic I have.

Sounds pretty good - the acm-1200 is now definately on the darker side rather than overly bright.
Perhaps due to the 330pf 'eq cap' I have in there. It's good actually - I'm not a fan of overly bright mics!

The capsule I'm awaiting is a beezneez selected china k67. Seemed like a reasonable way to go in this learner mic.

If I get a good result, I'll go on to try a 47 style with a single pentode stage to transformer type of thing. Trying to get some different styles going. My other ldc is a cheap mxl v67 fet which I've always liked a lot. Way better than an early rode nt1 (first version) I had and suffered with for years. !

-----

The other thing I am thinking about is  the remaining hiss and if I can improve that any.
It's a little difficult to assess because I don't have any really, really quiet spaces here. 

 

[kong]

Hi Alex, great thread and I'm also in the process of modding my Apex460. I have a question. What pins on the tube socket is the 330pf eq cap attach to (ground is #9?)?

 

[alexc]

Attach to first stage plate - pin 1 to ground.
I wouldn't use pin 9 as ground for this - that's the heater centre tap. 
I put a wire to the pcb ground trace.
Might make no difference but that's what I did.

I'm no expert at 'voicing' mics, but it's definately not overly bright with this and the stock capsule.

I've compared it to my (cheap) small diagphram condensor which is much brighter and also to a large diagphram fet condensor which is quite similar. Not bad so far. Tube circuit sans capsule very flat and wide response.

I should have my capsule this week. Assuming I can get the mount to work (bets?) I'll have a decently qc'd sino k67 capsule in there. So it should be better, in the case where my current cheap sino is not up to scratch.

Anyhow, this is a good one to learn on and $100 or so in mod cost.

I think the tnc acm81 stock is not setup as well as a stock apex460.

Mine worked but was truly woeful - low output, thin and crappy but with excellent reproduction of hiss

The next major step forward is the Chunger/Matador Apex 460/C12 rebuild.



Cheers

 

[kong]

Thanks Alex, I'll try that. I'm thinking of getting an RK-12 for the capsule and I'm following that 460/C12 as well as I have another apex460 that I haven't done anything to it.

 

[hazmatstrat]

Thanks  pasarki for the info and tchgtr with the interest

I'm doing some reading regarding capsule polarisation schemes and so on now.

I spent the evening doing some acoustic gtr and voice comparisons with the mic+mods+original capsule vs a sdc mic I have.

Sounds pretty good - the acm-1200 is now definately on the darker side rather than overly bright.
Perhaps due to the 330pf 'eq cap' I have in there. It's good actually - I'm not a fan of overly bright mics!

The capsule I'm awaiting is a beezneez selected china k67. Seemed like a reasonable way to go in this learner mic.

If I get a good result, I'll go on to try a 47 style with a single pentode stage to transformer type of thing. Trying to get some different styles going. My other ldc is a cheap mxl v67 fet which I've always liked a lot. Way better than an early rode nt1 (first version) I had and suffered with for years. !

-----

The other thing I am thinking about is  the remaining hiss and if I can improve that any.
It's a little difficult to assess because I don't have any really, really quiet spaces here. 

I found with these mics that 330pf is a little much on the plate to ground capacitor. In the ELAM 251 that capacitor is on a plate follower circuit. In those circuits the capacitor range can be form 100pf to 500pf while still maintaining the high frequency balance.

The cathode follower circuit reacts more dramatically to the plate to ground capacitor thus less is better. I use a maximum of about 100pf to tame a K67 style capsule.

What I do for the bass roll off is to use a 1uf polypropylene bypassed with a .022 polypropylene capacitor. I use the switch to engage and disengage the 1uf cap while leaving the .022 in place. I also get rid of the dual plate load resistors (part of the roll off) and replace them with a single 100k plate load resistor.

 

[alexc]

Thanks kindly for the info regarding the use of an 'eq cap' from first stage plate to ground, particularly in a ccda config.

What you say makes a lot of sense.

I've just completed changing the capsule to a sino qc'd k67 capsule from beezneez. It all works and I'm just starting to listen and compare now, so I'll have a better idea soon.

I ended up with a 1uF output cap as from my measurements, the ccda response is quite flat with that value.
I have the 330pf in there for now and will do the listening.

One thing I did notice was, I thought I replaced the capsule correctly, with the 'front diaphragm connecting to the variable polarisation voltage and the rear connecting to ground.

I did notice in cardioid, the back of the mic seems significantly higher sensitivity than the front.

So I swapped the diaphragm leads around and sure enough, the front of the mic is more sensitive.

have I got something wrong here? Is the diaphragm facing out from the rear of the mic supposed to be the grounded one? Or not?

Thanks

 

[hazmatstrat]

Thanks kindly for the info regarding the use of an 'eq cap' from first stage plate to ground, particularly in a ccda config.

What you say makes a lot of sense.

I've just completed changing the capsule to a sino qc'd capsule from beezneez. It all works and I'm just starting to listen and compare now, so I'll have a better idea soon.

I ended up with a 1uF output cap as from my measurements, the ccda response is quite flat with that value.
I have the 330pf in there for now and will do the listening.

One thing I did notice was, I thought I replaced the capsule correctly, with the 'front diaphragm connecting to the variable polarisation voltage and the rear connecting to ground.

I did notice in cardioid, the back of the mic seems significantly higher sensitivity than the front.

So I swapped the diaphragm leads around and sure enough, the front of the mic is more sensitive.

have I got something wrong here? Is the diaphragm facing out from the rear of the mic supposed to be the grounded one? Or not?

Thanks

\

The front goes to ground and the back to the pattern voltage.

JJ

 

[alexc]

8) 8) 8)

Thanks so much!

That explains it.

In that case, all is good.

Just burning in that EH 12AY7/6072 tube for a day or two before listening and further tests.


Cheers

 

[alexc]

Did some intital tests and the early results are very favourable indeed

The first thing is the mic is now very, very QUIET!!! 

The stock tnc acm1200  was like putting a huge conch shell to your ear.  You can hear the ocean. LOUD!

Now quiet as a mouse.

No problems with gain - it is now a lot more sensitive.

The stock capsule would be well suited to being a drum midi transducer.

More tests to come but for now it appears that the mic is sensitive, full bodied response and quiet.
A little brighter still than my comparison mic, a cheap but good sounding mxl v67 fet (very deep and chesty) and surely much more hifi.

Awesome

 

[monkeyxx]

Could someone please tell me where in the circuit to solder the 100pF high end smoothing cap?  I can't figure this out anywhere, all the ELA M schematics I can find are so tiny I can't read them.  Should be a 500 V?

My CEK-12 "C-12" Apex 460s are too bright for my taste, right now.

 

[HellfireStudios]

You want to hook the smoothing cap between C8 (NOT the transformer side) and ground. Also, it states in the 251 schematic that cap is rated for 500 volts. So, yes.

-James-