Ampex 600/601 Tube mic & Line amp

GroupDIY Audio Forum

Help Support GroupDIY Audio Forum:

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

thomasdf

Well-known member
GDIY Supporter
Joined
Aug 10, 2011
Messages
966
Location
Paris, France
Hello
I recently bought a nice Ampex 600 chassis, I gutted the electronics and I intend to rebuild it fresh from the ground up with some tweaks.  I want to be able to use it as a mono mic preamp, or stereo line amplifier. I also want to keep the main power transformer and the 5,5 henry choke.

I have bought some iron too, a pair of Altec 15335A for the line inputs, and a pair of Altec 15095A for the line outputs, and I still have to choose the mic input transformer. I have a pair of Gates mic transformers coming from a Yard tube console, maybe I'll use one of these, maybe another one. UTC A10 & 11 are a bit expensive for my budget...

I made a first version of the schematics, merging the schematics from the 600 and 601 models and doing some tweaks.
What is your opinion on that?
My main doubts are the PSU section (untouched for now on), I now it won't behave the same way without the same number of tubes and different tube power consumption etc. But I'd like to keep it tube rectified. And the Right side line input (I don't know if it will behave like the left side)

Any help is welcome :)

Linked is the schematics I have for now on...

Thanks a whole lot

Thomas
 

Attachments

  • AmpexHybridV2.jpg
    AmpexHybridV2.jpg
    519 KB · Views: 184
You do not need the 270K in series with the second (lower) line input. It is there in the top mic/line input stage because it forms a simple passive mixer with the line input. It is poor engineering practice to make the dc path from a grid to ground via the wiper of a pot so I would recommend you a  1M resistor from grid to 0V in both stages.

The HT smoothing coule be improved. In those days high value high voltage electrolytics did not exist. I recommend you increase C103C and C103D to 220uF.

Cheers

Ian
 
Copy that!
I knew some tube gurus would have awesome feedback to give me :)
Thanks a lot!

I'll re do the schematics with the modifications.
However, I still have a question. If I do not use that 270R resistor on the R line input, will I have the same level in the L and R channels? I will use rotary switches + resistors used as pots for input and outputs attenuators, and wanted to have a precise LR matching in order to use on my mixbus if necessary... (I'll use metal film resistors btw)
 
thomasdf said:
However, I still have a question. If I do not use that 270R resistor on the R line input, will I have the same level in the L and R channels? I will use rotary switches + resistors used as pots for input and outputs attenuators, and wanted to have a precise LR matching in order to use on my mixbus if necessary... (I'll use metal film resistors btw)

No it will not match with the 270K in or with it out. On the mic/line input, the pair of 270K resistors make a simple passive mixer with a 6dB insertion loss. In the single line input there is no loss so the sensitivity will be 6dB higher. Also, if you are using 10K:10K input transformers for the line inputs then you should change the pots from 250K to 10K. This will provide a better match and improve the noise.

Cheers

Ian
 
> 6dB insertion loss.

Include the 12AY7 plate node. More like 5dB, if we care. A variable 270K-400K network where the mike output is missing could be trimmed for exactness.
 
ruffrecords said:
Also, if you are using 10K:10K input transformers for the line inputs then you should change the pots from 250K to 10K. This will provide a better match and improve the noise
The input transformers are Altec 15335A bridging transformers 15K:15K. Should I build a custom 15K attenuator scale, or will 10K be good enough?

PRR said:
A variable 270K-400K network where the mike output is missing could be trimmed for exactness.

Edit:
Attached is the updated version of the schematics!

Will do!
 

Attachments

  • AmpexHybridV2B.jpg
    AmpexHybridV2B.jpg
    544.1 KB · Views: 86
I am in the process of buying the missing parts... I will keep you updated!
I still have a question though: what's the simplest and cleanest way of adding a DI input to this? I guess, it is straight into the 5879? Should I use some kind of capacitor to protect the DI input from HT in case something goes wrong with the tube, or whatever?
Thanks!!
 
I have been busy and I have many projects on the bench at the same time, but I took time to plan things a bit in the enclosure and figure out a good placement for tube socket, transformers, etc. I should start the soldering pretty soon :)

thomasdf said:
I still have a question though: what's the simplest and cleanest way of adding a DI input to this? I guess, it is straight into the 5879? Should I use some kind of capacitor to protect the DI input from HT in case something goes wrong with the tube, or whatever?
I didn't have time to figure this out, any help is welcome as I haven't done this before.

I have some other transformer related questions:
Line input transformers:
I bought 15k/15k transformers for the line inputs as I thought I would feed the unit mostly things from a D/A converter. But if I want to use it on a mixbus after an analog console output, or a GSSL or something like that, or even in a signal chain after some other hardware unit, wouldn't it be better to use a 600:10K transformer instead?  Let's say, another pair of 15095A's?

Mic input transformer:
I have tiny Gates the Yard mic input transformers that I want to try here, I've read mixed things about them and I want to hear them for myself. But in case, wouldn't another Altec 15095A be nice as a mic input transformer? It has the ability to have a 150 or 600 ohms primary.. Has anybody ever tried it as a mic input xfmr?

Thanks a lot!
 
thomasdf said:
The input transformers are Altec 15335A bridging transformers 15K:15K. Should I build a custom 15K attenuator scale, or will 10K be good enough?
Transformers are not finnicky; they tolerate some mismatch. But, as these are spec'd as 15k, I would use a potentiometer of higher value, at least 22k, even as high as 100k. Using a lower load than nominal often results in loss of HF (and level).
 
thomasdf said:
Line input transformers:
wouldn't it be better to use a 600:10K transformer instead?  Let's say, another pair of 15095A's?
I believe you mean using them in a 10k:600 configuration... Remember that it would result in a 12dB  lower output level. But also probably lower distortion and higher headroom.

Mic input transformer:
wouldn't another Altec 15095A be nice as a mic input transformer? It has the ability to have a 150 or 600 ohms primary..
In fact, the 15095A is designed as an input xfmr, so there is no reason it wouldn't be adequate in this position.
According to Altec litt, it is not specifically recommended for outputs, although most of the specs indicate that it should be adequate. A transformer specifically designed for outputs may have somewhat lower DC resistance and slightly different magnetic properties, but I would think this one should work quite well in practice.
 
I chose it for the output because it seems to sound really good in similar converted Ampex 600... I'll tell you more about it :)
Thanks for your input, I'll three other ones in case the mic & line inputs do not suit me.
 
I would differ with Abbey Road on the 1595 transformer. It was designed as an output transformer & was used in Altecs' 1567 tube mixer. It was driven by a 5CG7 dual triode driver tube & has an impedance of 13k on the primary.

As you have plenty of gain in the chain I recommend leaving out the 40mf cathode bypass capacitor 101 a,  b & 105. The circuit has plenty of gain & does not need the extra gain that these capacitors provide. The result will be an improvement in signal to noise ratio.

Your circuit is quite similar to the circuit of the 6 input mike mixer I designed & used in the 60's, for on location master taping. The main difference was that I did not use a pentode as the mike preamp stage. Rather the Mullard industrial 12AX7/7025 were quieter for mike preamplifier.
 
the 601 can exhibit an EIN of better than -128 with the cathode bypass capacitor 101a in place.
noise performance is degraded if removed.  consider the hum from the AC heaters.
the 351's preamp cathode is not bypassed and has a decent EIN, but is improved with an appropriate capacitor.
as the 351's heaters are DC, the noise increase when not bypassed is less severe.
 
Bill Wilson said:
I would differ with Abbey Road on the 1595 transformer. It was designed as an output transformer & was used in Altecs' 1567 tube mixer. It was driven by a 5CG7 dual triode driver tube & has an impedance of 13k on the primary.
15095 is not the same as 15095A
15095 was specifically presented as an output xfmr, with the 15k winding as primary; its cold leg was internally grounded. The 15095A was presented as an input xfmr, with a 600 ohm primary. Both windings were floating.
Indeed, xfmrs can be used in either direction.
Altec's choice to present them differently may have come from marketing reasons but their engineers may have modified slightly the design in order to make the "A" more suitable to the preferred application.
 
gridcurrent said:
the 601 can exhibit an EIN of better than -128 with the cathode bypass capacitor 101a in place.
noise performance is degraded if removed.  consider the hum from the AC heaters.
I don't get that. I would think that heater hum is due to capacitive coupling to the grid.


the 351's preamp cathode is not bypassed and has a decent EIN, but is improved with an appropriate capacitor.
as the 351's heaters are DC, the noise increase when not bypassed is less severe.
Had the Ampex engineers been concerned with noise on the 601, they would have used a triode instead of a pentode with its partition noise. That's what they did on the 351.
 
heater induced hum is discussed in RDH4, page 1197, section (2).

The "progression" of Ampex circuitry is thought provoking:
models 200, 400, and early 300's were pentode based while the 3rd iteration of the 300 and the 351 used un-bypassed triodes. 
354/PR-10 models employed a combination of tube types, though no dedicated on-board mic preamps.

Partition noise aside, a pentode based preamplifier can be extremely quiet, whether an Ampex 601 or Langevin 5116.

 
gridcurrent said:
heater induced hum is discussed in RDH4, page 1197, section (2).
In a circuit where the grid circuit impedance is about 50 kohm, electrostatic coupling between heater lines and cathode circuit is a non-issue. In RF applications, where the grid circuit impedance may be very small, heater-to-cathode coupling may become the main hum contribution.
 
Audio Cyclopedia, 1st edition, page 319,
"Unbypassed cathode circuits are more susceptible to hum pickup from AC heater circuits."
 

Latest posts

Back
Top