Shure M64 phono/tape preamp

SeleniumFalcon

I was able to buy two non-working Shure M64 preamps off of eBay. I wanted to learn more about them and see how they sounded as a playback system for magnetic tapes. There is a three way selector switch that allows for either a "flat", "phono (RIAA)" or "tape (NAB)" equalization. There's a 30volt DC output connection for use with other Shure devices.

SeleniumFalcon

The circuitry seems pretty straight forward, the power supply is non-regulated and un-fused. The two gain stages use a single NPN and PNP transistor (TIS97) and (2N5087) per channel with 4.7uF capacitors as interstage coupling and blocking. The eBay description described them as distorted and sold "as is" and one unit definitely needs a new AC power cord. The chassis are fairly heavy gauge steel and pretty well braced internally with a single circuit board.

SeleniumFalcon

The first step is to gingerly bring one up to operating voltage and see what power supply measurements I can make.

msg

Pretty cool, Ken. Nice photos, too.

Where'd you get the schematic?

Do you plan to get one or both working and use them with your R2Rs and or turntables?

What would flat be for? You mentioned equalizations. Are these tonal type equalizations, or output gain profiles?

mhardy6647

msg wrote: »

Pretty cool, Ken. Nice photos, too.

Where'd you get the schematic?

Do you plan to get one or both working and use them with your R2Rs and or turntables?

What would flat be for? You mentioned equalizations. Are these tonal type equalizations, or output gain profiles?

I'm not @SeleniumFalcon ... but I played him in some of the less-popular Star Wars retreads reboots.

https://www.hifiengine.com/manual_library/shure/m64.shtml
It's in the datasheet.



On-topic (and FWIWI): My hifi colleagues (well, more like my hifi superiors) Kevin Kennedy & Charles King, has done a lot with phonostages/tape PB EQ in recent years.

SeleniumFalcon

As @mhardy6647 posted I found the schematic on hifiengine's site as part of the owner's manual. Shure always seemed to be a good company for making product information available. I'm not sure what my goal is at present, hopefully I can get both working and sell them, but mostly just understand them. The TIS97 transistor has a really good reputation and new old stock can be found available on eBay. It is used in the Moog synthesizers and in studio consoles and I've seen it in other circuits.
There isn't too much difference between what is needed to amplify and EQ records and what is needed to play tapes. The output from a playback head and a moving magnet/moving iron type cartridge are fairly similar. The EQs are similar, RIAA requires three frequency transition points: 2122 Hz, 500 Hz and 50 Hz . NAB EQ for 7.5IPS requires two frequency transition points: 50Hz and 3180Hz, not that far different from each other. So, the sliding switch can go between the negative feedback components that accomplish either (or none in the case of "flat") record or tape head playback.
Charles King has indicated he is working on producing a replacement playback circuit board for the Studer A80 deck. He's asked for nearby Studer owners to try out his design on their machines. The irony is that as tape transports got better at controlling the tape's movement across the heads the corresponding playback electronics got worse sounding. In my semi-humble opinion.

SeleniumFalcon

I've had some time to tinker with the M64 and have removed the main circuit board from the chassis and taken off the capacitors and resistors. I also removed the terminal posts and replaced them with some better made versions.

SeleniumFalcon

I dissembled the chassis and started taking things apart and doing some sprucing up.

mhardy6647

Your... enthusiam, work ethic, and attention to detail all astonish me (in a good way, I mean)!

SeleniumFalcon

Thank you @mhardy6647 I appreciate your comments.

SeleniumFalcon

I've received the resistors and installed them, since the lead spacing on the circuit board is small (6.1mm) I had to order smaller 1/4 watt Dale metal film versions other than my normal military specification RN series. While I was working I installed Teflon transistor sockets that allow for easy inserting and removing (nice for silicon/germanium rolling).

SeleniumFalcon

All of the capacitors have been installed.

SeleniumFalcon

I found a NOS Shure battery power supply that can be used with the M64 preamp, eliminating the AC power supply. That should reduce any hum or effects of ripple in the power supply. The goal of any well designed power supply is to come as close as possible to a battery. Granted the original Eveready 612 batteries are no longer available, but the lithium 9 volt versions should work fine.

SeleniumFalcon

It's alive!

SeleniumFalcon

I put a couple of transistors in place, hooked up the DC power supply and connected my THD meter. In the "flat" position it was within 0.8dB from 20 to 20kHz at 0.3% THD.

SeleniumFalcon

SeleniumFalcon

I bought a reverse RIAA EQ circuit from the Hagerman company, this small circuit takes any incoming audio signal and creates an equalized version that follows what a record would produce.

SeleniumFalcon

I then connected by analyzer up through the RIAA circuit and then through the M64 looking at the frequency response. The analyzer is displaying a vertical sensitivity of 2.0dB.

SeleniumFalcon

As a test I increased the value of C9 from 100uF to 1600uF since this capacitor establishes the "knee" of the high pass filter of the negative feedback. The goal is to extend the low frequency response below 60Hz.



Here is the result:

SeleniumFalcon

SeleniumFalcon

I decided I wasn't quite getting an accurate enough idea of what making a change to the value of C9 was doing so I went back to making measurements with my signal generator and AC volt meter. The graph shows three curves the top most shows the standard RIAA curve with 1.0kHz as 0dB. Looking at 20Hz, the middle curve (green) shows the circuit with the 1,600uF capacitor at C9. The lower curve red) shows the original circuit with C9 as 100uF. As you can see while the lower most frequencies (20Hz) are more near the RIAA curve the rest of the lower ones are slightly less with the increased value. I believe Shure chose the value they did to minimize problems in the infrasonic range caused by turntable problems. I think I'm going to experiment with some different value capacitors to try and find a good compromise value.

mhardy6647

Not to go all ASR/engineer on this thread... but.... you ought to be able to model the response with different C values, right? I mean, I don't know how to: maybe with SPICE or MATLAB/Simulink or something like that?

SeleniumFalcon

Good suggestion. I just wish I was a little more computer facile.

SeleniumFalcon

Now that I've finished repairing a Revox A77 I have some spare time to look at the M64 and how changing the value of C9 affects the RIAA playback response. I ordered some Elna Silmic II capacitors in 330uF and 740uF and placed two test pins in the circuit board so I could swap capacitors without having to disconnect the test rig. By using the inverse RIAA circuit I could measure the output at several frequencies and see how my voltmeter measured the deviations from 0dB. I used 1.0kHz as the hinge frequency setting the levels at 0dB. Then I went down in frequency to 20Hz and then up in frequency to 32kHz. As a sort of reference I used my NAD3020 as a comparison measuring from the preamp output to avoid the power amp section.
The top most plot is the original 100uF Shure capacitor, below that the 330uF, next the 470uF, then the 1,600uF and bottom most is the NAD response. The NAD shows the better overall response especially in the lower most frequencies, however the upper response of the Shure is more linear. However, across the broadest amount of frequencies the meter hardly moved with only 1.0dB separating them all.

SeleniumFalcon

The next step is to look at the two trasistors, the NPN TIS97 and the complimentary PNP 2N5087 on my transistor curve tracer. I want to check for gain matching and linearity most of all. I found a few sources for new old stock of these and began checking the originals and the new versions. What I found was a condition called "looping" instead of the trace being a stair step of fairly straight lines this is what it looks like:



The loops are caused by some form of hysteresis in that the device responds differently to a rising voltage than it does to a falling voltage. This hysteresis is caused by some form of storage mechanism like capacitance or inductance inside the transistor. I checked all of the NPN and PNP transistors and they all showed the same characteristics. I decided this probably wasn't a good characteristic for audio use, so I ordered some more modern versions of these transistors getting complimentary NPN/PNP pairs.

SeleniumFalcon

I decided I would construct a new enclosure for the Shure allowing the circuit board to be kept flat so different transistors could easily be fitted in the sockets. I ordered the enclosure from Mouser and drilled holes, painted the two parts and labeled the necessary inputs/outputs. I sourced the RCA connectors and the DP3T slide selector and a ground lug.
Now I can start wiring everything together.

machone

Amazing....

SeleniumFalcon

My wife figured out what my motivation is, she said, "you just like building kits, you miss Heathkit."

SeleniumFalcon

Now that I have got the Crown just about where it should be I can finish up the Shure. I've completed the construction and it seems to be working. I've connected it to the battery supply and made a few preliminary tests.

SeleniumFalcon

SeleniumFalcon

This is a photo of the battery supply.