Motown Direct Amplifier-inspired Preamp?

[MidnightArrakis]

(very similar-looking (earlier?) unit with a different layout)

There's also a top of chassis pic (I've brightened it since the can cap is difficult to make out in the original) :View attachment 137423

[(very similar-looking (earlier?)] -- Your photo of this chassis more accurately follows the drawing posted earlier than the other photo I had used in my Post #433.

>> Is it possible for you to provide a list of Part Numbers of the other transformers (i.e., HAMMOND 157G, CineMag CM-????, etc.), the large electrolytic capacitors and anything else used inside the chassis? That would prove to be highly useful with my trying to recreate the mechanical design details of this chassis. NOTE: I basically -- assumed -- the chassis height as being 1.50" high. Can you confirm what the actual chassis height is??? THANKS!!!

/

 

[MidnightArrakis]

Nice one midnight. I have been using ChatGPT to find out what tubes are used. It says the first three (from left to right in your chassis) are EF86, 12AX7, 12B4A

It gets confused about the fourth tube (the rectifier). First it says it is a 6X4 (which is a small B7G tube), then it says it used a semiconductor rectifier, and lastly it thinks it is a 5AR4/GZ34 (which seems most likely).

The mains transformer (rear right of the chassis) seems to be a Hammond 309JX if my reading is right. Output transformer is a Cinemag and probably physically identical to their CM-2810 model.

Cheers

Ian

FYI:


/

 

[DaveP]

I don't understand the purpose of R5.
Cheers
Ian

R5 supplies the DC feedback, as in the V72
best
DaveP

 

[abbey road d enfer]

In the V72, the 50k resistor does not provide DC feedback, because the DC loop is open due to the coupling capacitor at the grid of the 2nd pentode.
It does provide some AC POSITIVE feedback, though, sensing the output current, which creates a negative impedance to drive the xfmr, which is a way of reducing LF distortion.
In your circuit, this "fedback" loop does not sense the transformer current, and it senses only DC, because of capacitor C5 across R6.

 

[dai h.]

View attachment 137427

>> Is it possible for you to provide a list of Part Numbers of the other transformers (i.e., HAMMOND 157G, CineMag CM-????, etc.), the large electrolytic capacitors and anything else used inside the chassis? That would prove to be highly useful with my trying to recreate the mechanical design details of this chassis. NOTE: I basically -- assumed -- the chassis height as being 1.50" high. Can you confirm what the actual chassis height is??? THANKS!!!

/

Sorry Midnight, I've got nothing more than the info and the pic in that link.

(err.. too fuzzy to make out clearly, but) possibly CM-2810 (Vacuum tube output; 4:1, 9.6K:600R)?

https://cinemag.biz/output/PDF/CM-2810.pdf

(the colors for the wires in the pdf. (BLK-ORG-RED-YEL) seem to match the ones in the pic (though maybe that's the same for many of Cinemag's XFMRs)

 

[ravachol]

I was led to believe that Electrodyne were the Motown amps?

Electrodyne solid-state and Altec passive EQ —to start– and towels on the drumheads.

 

[DaveP]

• #444

In the V72, the 50k resistor does not provide DC feedback, because the DC loop is open due to the coupling capacitor at the grid of the 2nd pentode.
It does provide some AC POSITIVE feedback, though, sensing the output current, which creates a negative impedance to drive the xfmr, which is a way of reducing LF distortion.
In your circuit, this "fedback" loop does not sense the transformer current, and it senses only DC, because of capacitor C5 across R6.

I take your point, but this is a detail that I was reluctant to discard from the V72 without trying it first on the bench. In fact the actual V72 does not have any ac feedback because it has dc feedback from the second anode to the first tube cathode. This draws an extra 2.5mA which I did not want to do. So I have actually copied the V77, not the V72 and I think as you say, the "dc" feedback could be discarded.
best
DaveP

 

[ruffrecords]

I take your point, but this is a detail that I was reluctant to discard from the V72 without trying it first on the bench. In fact the actual V72 does not have any ac feedback because it has dc feedback from the second anode to the first tube cathode. This draws an extra 2.5mA which I did not want to do. So I have actually copied the V77, not the V72 and I think as you say, the "dc" feedback could be discarded.
best
DaveP

The dc feedback in the V72 and V77 helps to improve LF stability because it ensures there is only a single zero in the loop. Without this there are two zeros in the loop and there will be a resonance at some LF frequency. To minimise its size and frequency requires careful choice of the coupling and feedback capacitors. The dc feedback in many of the V series amps is a major innovation that eliminates this. Another good amp which uses a similar technique is the Pultec MB-1 mic pre:



This design also direct couple the second triode to the cathode follower. There is no reason we could not retain the 6BX7.

Cheers

Ian

 

[DaveP]

It's not possible to use DC feedback from the anode of the second tube to the cathode of the first unless a choke load is used like the V72. The problem is the excessive voltage drop, with the choke (dcr 11k) the voltage drop is feasible, but with a 47k plate load the voltage drop is too great. The V77 does not use dc feedback, neither does the V76, so copying their schematics is possible.

I am working on the original spec, but I agree that the Pultec circuit will suffice for most readers.
best
DaveP

 

[DaveP]

This is my draft circuit for the original design.
I have removed the dc feedback.


The cathode follower with a 10k load, is able to swing 270p-p which is 95.47Vrms.
The current can swing .027A p-p which is 0.009547A rms.
The power is then 95.47x 0.009547 = 0.91W

The tube is running about 7.5mA/tube making ~15mA total
Vk-Vg is ~16V

The 10k cathode resistor burns 2.3W so a 5W component is recommended.

If I have made an error I'm sure you will let me know.
best
Dave

 

[MidnightArrakis]

Hammond 369JX pwr xfmr

[Hammond 369JX pwr xfmr] -- I have recently determined that the chassis dimension drawing shown in Post #395 must be using a different power-transformer than the HAMMOND 369JX. I say this because I have created an accurate 3D CAD-model of the HAMMOND 369JX power-transformer and its mechanical dimensions are identical to the HAMMOND 369JX datasheet. However, upon my importing the 369JX CAD-model into the chassis I have created using the dimensions shown in Post #395, the transformer mounting holes do not align anywhere near closely enough to the chassis mounting holes. But, on the other hand, it looks as though the mechanical dimensions of the chassis shown in the photo below all do align up in a variety of different ways.

Does anybody have a mechanical detail drawing of the chassis shown below?









/

 

[ruffrecords]

The chassis also looks like it could be by Hammond. It would probably be worth checking their catalogue to find one with dimensions close to that shown in the photo.

Cheers

Ian

 

[DaveP]

The poles on this circuit are 1.6Hz for C1/R3 and the feedback C2/R4 is 6Hz with 470nF but this could be increased to 2.2uF which would make it 1.3Hz.
This could be decided by testing on the bench. The V77 uses 25nF/3M to give 2.1Hz and 2uF to give 1.5Hz.

 

[ruffrecords]

The poles on this circuit are 1.6Hz for C1/R3 and the feedback C2/R4 is 6Hz with 470nF but this could be increased to 2.2uF which would make it 1.3Hz.
This could be decided by testing on the bench. The V77 uses 25nF/3M to give 2.1Hz and 2uF to give 1.5Hz.

If you have more than one pole or more than one zero at the extremes of the passband it is a good idea to separate them by a decade to simplify stablility. This effectively makes one of each pair dominant.

Cheers

Ian

 

[DaveP]

If you have more than one pole or more than one zero at the extremes of the passband it is a good idea to separate them by a decade to simplify stablility. This effectively makes one of each pair dominant.

This is true but down at 1.5Hz I doubt if there is enough gain to cause trouble, but to be safe we could use 3uF for C2/R4 (1Hz) and make R6/C5 22uF for 10Hz
best
DaveP

 

[DaveP]

For completeness I have worked out the final constants of the 6BX7 as used in the circuit

The chart given is for each section, so I had to draw the load line for 20K and then double the rp and gm.
This gives mu=4.5, rp=1.7k and gm=2.647.

We can now work out the output impedance: 5:1 opt means 25:1 z ratio
1700/25=68ohms

The output voltage for 600 ohms load would be 100v x600/600+68 = 89.8v
and for a 300 ohm load it would be 100V x 300/ 300+68 = 81.5v

This shows that this output tube can handle 300 ohm loads as well as 600 ohms as in the original spec.
My final schematic:

best
DaveP

 

[MidnightArrakis]

The chassis also looks like it could be by Hammond. It would probably be worth checking their catalogue to find one with dimensions close to that shown in the photo.

Cheers

Ian

[probably be worth checking their catalogue to find one with dimensions close to that shown in the photo] -- Will do.

/

 

[MidnightArrakis]

My final schematic:

best
DaveP

[My final schematic] -- I have probably more than likely missed something along the way through this thread, but I have been under the impression that this "Magic Motown DI Preamplifier" was a -- DI -- box. And, unless there happens to be an XLR connector mounted on the rear of this chassis, I only see a 1/4-inch jack on the front, although there is a toggle switch that is switching between -- MIC & INST -- as is shown here:



>> Shouldn't your schematic also be showing a 1/4-inch instrument input jack as well as a switch toggling between it and the XLR? Just wondering.....

>> Do you have any suggested recommendations as to what both T1 and T2 "could/might" be? With my knowing who makes them and what their Part Numbers are, I could layout a series of hole cutouts on the top of the chassis which would then be used to accommodate any one of them.

>> Is there an accompanying power-supply schematic for this circuit somewhere?

>> Are all of the resistors 1/4-Watt (other than R12)? Or, do any of the resistors need to be 1/2-Watt?

/

 

[ruffrecords]

For completeness I have worked out the final constants of the 6BX7 as used in the circuit

The chart given is for each section, so I had to draw the load line for 20K and then double the rp and gm.
This gives mu=4.5, rp=1.7k and gm=2.647.

I think we discussed earlier that when paralleled the gm doubles, rp halves and mu remains unchanged.
Cheers

Ian

 

[MidnightArrakis]

The chassis also looks like it could be by Hammond. It would probably be worth checking their catalogue to find one with dimensions close to that shown in the photo.

Cheers

Ian

[The chassis also looks like it could be by Hammond] -- I don't know whether or not if the chassis shown in the photo is a HAMMOND chassis, but I did come across this fine chassis on their website. It measures 10"W X 6"D X 2"H (254mmW X 152mmD X 51mmH) and can be viewed in all of its glory here:

https://www.hammfg.com/electronics/small-case/chassis/1444



Now.....as soon as the questions I had asked earlier can be answered, then I can begin inserting component cutouts for everything and turn this blank chassis into a "Magic Motown DI Preamplifier Chassis" and be ready to accept -- DaveP's -- finely-honed circuitry. Any comments and/or suggestions as to components placement, possible mechanical features and/or anything else are welcomed for possible inclusion. No guarantees implied.

>> LET THE FUN BEGIN!!! <<

/