Looking for design consultation MCI-JH24

[Westrek]

I working on a modification to the MCI-24 tape machine bias/erase card that will allow use of the Nortronics Ferrite/Mu-metal head in a 16-track format.
When using this head with the stock card (which was designed for the Woelke erase head and impossible to fine) the erase voltage cannot be adjusted high enough to achieve reasonable erasure.
It's my conclusion that the drive amplifier simply doesn't have enough power output.

I'm looking for an design engineer to consult with me on this project. I've done lots of experiments and have data, schematics, head specs, ect. to supply.

Thanks,

Mike Miller
Westend Recording Studios
Kansas City

[nickb]

There are many engineers on the site. Can I suggest a good place to start would be to post the schematics so everyone can see what we are dealing with?

[Enzo]

We do have a collection of engineers here, and they may have some insights. I would also recommend you bounce this off a forum of professional recording engineer types.

[Westrek]

Thanks. I will post the schematic and the head specs tomorrow.

[Westrek]

Thanks Enzo. I've been working on this for awhile. This is a difficult area because it rarely needs attention from an engineering standpoint. Most are more adept at upgrading and modifying audio path circuits. I will post what I have tomorrow.

Mike Miller

[nickb]

Mike, are you the same guy as the moderator over at the JH-16/JH24 forum of the mcirecording.com site?

[Westrek]

Yes

[DrGonz78]

It is funny that I accidently just found this website...
Welcome to CrazyJoe.org, the OFFICIAL website of Dr. "Crazy Joe" Tritschler and your EXCLUSIVE source for Crazy Joe LP's, CD's, downloads, and more! From Tritschler Precision Engineering.

Not sure if he is willing to help, but I think he is the right kind of guy you are looking to meet. Just a thought...

[Westrek]

Here's the bias/erase card schematic: http://westendstudio.com/wp-content/...-schematic.pdf

The record block diagram: http://westendstudio.com/wp-content/...rd-Block-3.pdf

Bias/Erase circuit description: http://westendstudio.com/wp-content/.../MCI-Audio.pdf

The original Woelke erase head specs:

Inductance: 1.4 mHy
DCR: 2.8 Ohms
Erase Freq: 100K
Erase Current: 65ma
Max Erase Current: 80ma
Q-Factor: 2.66

The replacement Nortronics 9997 erase head specs:

Inductance: 1.3mHy
DCR: 17 Ohms
Erase Freq: 120K
Erase Voltage: RMS=60v
Erase current: 67ma
Impedance at erase freq: 900 ohms.
Q-Factor: .77

I believe that the Nortronics Q is very narrow and since the erase freq is 105K, the impedance is too low for the existing erase amp to drive it to the 65v necessary for good erasure. 35 to 40v rms is all the stock card will output. The waveform looks fine, the amp appears to just run out of soup.
I've tried changing the series capacitance after T2, but to no avail.
I've been able to push the erase amp harder (increasing the output by about 10 volts) by lowering the value of R41 to about 10 ohms and paralleling an additional VN66AF to Q13 and 14. It draws lot of additional current this way and will most likely melt down quickly.

I appreciate any looks and advice.

Mike Miller

[km6xz]

Actually Q factor is just the opposite of what you assume. The original head has a sharper/steeper slope at surrounding the resonant frequency so for any given power of signal applied the amplitude will be higher at resonance with the Q of 2.66 than the same oscillator driving a head with Q of .77 if all else is held constant.
It sounds as if you will need more power (but be limited by the 65ma) if you do not get the oscillator right at the center frequency of the head inductance and tank capacitor. Have you swept the drive signal over the range to find the precise resonate frequency? The low Q head is reasonable for a replacement in a variety of machines because it has a broad but lower peak, making it functional with circuits that deviate from the design frequency which was fine in the early days when flux densities were low and it was easier to erase tape. But you need high flux density to erase later tapes that ran over 350nw/m^2

The low Q of the Nortronics head is due to the high DC resistance compared to the original. Do you have a way of measuring flux, say with a loop, or Hall Effect transducer, instead of a scope that will load the tuned circuit and give ambiguous results? The scope will give you oscillator amplitude but not at the head that, due to being a resonant circuit will be very high Z at center frequency and much higher amplitude than the oscillator output.

[Westrek]

Thanks for your kind reply.

I think if I could change the erase frequency from 105K to 120K, it would probably work fine.

I don't have a way of measuring flux, however.

The old discrete electronics of the prior MCI machines used 120K for both the bias and erase and did use the Nortronics erase head.

If the master oscillator could be changed to 240K and then the erase divided down to 120K....might work. Unless it would create problems on the bias side.

On the Nortronics spec sheet it states this:

"High efficiency erase heads with ferrite/Hy Mu cores for use at operating frequencies up to 250k.
Operating voltage should be set proportional to frequency; i.e. at 150k it is 150% of 100k voltage.
Do not exceed specified voltage by more than 20% to avoid core saturation"

Here's how they did it on the older machines that used the head in question:

http://westendstudio.com/wp-content/...4/07/JH-16.pdf

[km6xz]

I never had a MCI, we had a bunch of Studer, Ampex and a couple Stevens. Did the old deck use the same Nortronics model? That Q seems pretty low for a spec sourced head, more of a general purpose replacement head.
Does C40 peak somewhere near the center of its adjustment range? Moving it up in frequency will require changing all the reactive components on each card. The voltage developed across it at resonance will be based on the Q and the inductive reactance. The higher the X-L the higher peak voltage at resonance, with the Q determining the power losses. Hi Q, at resonance, will have very little loss. But there is no free lunch going up increases head magnetic losses.
I think the first step would be to find out what the head and capacitor(in the tank circuit plus the head winding capacitance) frequency for resonance is right now.

The symbols for T3 and T4 indicate tuned core transformers, do these have slugs or are they cupcore type? If cupcore you can get the range by rewinding them but if they are slugs, the slug material would need to change to double the resonate frequency.
You do not need a flux meter for tuning since all you need is a sample using a non-resonant loop that will lightly couple, as to not load the resonant circuit. You only need that for seeing a peak. If tightly coupled like a scope probe directly on the erase hot lead will distort the peak....flatten it out and shift the center point. So just wind some turns on a flat piece of vector board, maybe an inch across. 10 turns of #28 wire would be fine. glue the flat coil board to a handle, like a Popsicle stick, connect some small coaxial cable, RG-174 would be small enough and connect that to your scope or AC meter with a high frequency bandwidth. You can measure relative amplitude across the face of the head and monitor frequency without detuning the tank circuit that happens with a scope probe connected directly.

[nickb]

Looks like TP2 is designed for using a meter to check for a voltage peak w/o messing up the resonance. It gives a DC output equal to 1/100 of the peak voltage across the erase head.

[Westrek]

The old decks used an all ferrite head. John French of JRF Magnetics measured the Q of a Nortronics all ferrite head and that's where I go my Q spec. Not from the actual head that I'm trying to use. He actually had the head in question in his shop at one time, but at that point in the project, we never thought to measure it.

T3 and T4 are slug type transformers. When adjusting C40, the amplitude will continue to rise till the trimmer is tight and the waveform still looks good. I am using a 10x probe which does appear to cut down on the loading effect.

Thanks so much for your help!

Mike Miller

[Westrek]

Yes, TP-2 makes it easy to adjust the erase output with just a dc voltmeter. Pretty clever.