Recreation of the suitcase preamp (schematic and layout attached)

Started by Tines&Reeds, November 11, 2016, 08:42:55 AM

Previous topic - Next topic

Tines&Reeds

Hey folks,

since I am building a custom satellite system for the pianos in my workshop, I am also interested in recreate the suitcase preamp. So I started to digitize the preamp of the Mark II suitcase. Mark II you ask? I talked to another rhodes-tech who invented a preamp by himself and he told me that the Mark II is technically the best preamp especially when it comes to the vibrato-section.

So I spent yesterday and today putting the layout into eagle which I attached to the post. The picture has a high resolution so you better download it instead of open it in your browser.

I also attached the pcb layout. The potentiometer will be connected externally.

I've got a few questions regarding the original layout which you can find here:
http://www.fenderrhodes.com/pdf/late-mark2-suitcase.pdf

1.) R14 and R26 are connected to voltage. But it's not mentioned, if it's + or -.
2.) Does anybody know how much current which voltage goes through the LEDs (power & vibrato)?
3.) There is no blocking-cap on U1. I think it's better to add one?
4.) C19 says 100uF while C20 says 100pF. Do you think that's a mistake and it both needs to be 100uF? This would make sense to me.

Any other suggestions are very appreciated! As soon as I verified the layout and prooved it's working, I'll put the layout-files online so you could etch them by yourself.

(Attachment didn't work... so I uploaded to dropbox).
https://www.dropbox.com/s/gzzuzlacgkpa32p/SuitcasePreamp.png?dl=0
https://www.dropbox.com/s/vee7flhw52twueb/SuitcasePreampPCB.png?dl=0

Thanks,
Chris
German Rhodes Tech nearby Hamburg / Bremen

http://www.tinesandreeds.com

Groove4Hire

Did you copy the slider preamp mk2 or the all pot preamp?
Jon
Rhodes-tech, www.vintagebua.no, Norway

Tines&Reeds

This is the pot-preamp. I linked to the schematic in my first post.
German Rhodes Tech nearby Hamburg / Bremen

http://www.tinesandreeds.com

pnoboy

It's been a long time since this preamp was designed, and one needs to investigate at least 3 things before proceeding too far.
First, are all the components available?  You may have noticed that this circuit uses 2 devices in which one LED shines simultaneously on 2 light-dependent resistors.  This is a clever design and I understand what the designer is doing, but I don't believe this component is available anywhere.  If it is, I'd like to know who sells them.

Second, it's worth investigating if there is an opportunity to substitute modern parts that perform better that some of the parts used back then.  One can pretty much guarantee that this is the case.

Third, this is an opportunity to potentially improve the preamp.  For example, would the tone controls be better if their frequency ranges were tweaked?  Can we make some modifications to make the preamp quieter or take less current?

Tines&Reeds

Quote from: pnoboy on November 14, 2016, 11:03:52 AM
First, are all the components available?  You may have noticed that this circuit uses 2 devices in which one LED shines simultaneously on 2 light-dependent resistors.  This is a clever design and I understand what the designer is doing, but I don't believe this component is available anywhere.  If it is, I'd like to know who sells them.
This is not a problem. Vactrols are not that hard to find.

QuoteSecond, it's worth investigating if there is an opportunity to substitute modern parts that perform better that some of the parts used back then.  One can pretty much guarantee that this is the case.
What "other parts" do you mean? Of course I'd not look for 40years old capacitors ;-) In principle the preamp is a really easy thing... just a few caps, resistors and opamps. The only thing, that also answers your last point, is, to try out other opamps. In this schematic the NE5534 is used. This is still available and still a workhorse. Anyway... now there are better op-amps available f.e. from TI OPA134, OPA604, etc. As the pinout is the same, I'd test that as soon as the preamp works with the standard NE5534.

And as far as I know the frequencies are set with some caps- and resistor-networks. So this can also be easily exchanged as soon as the preamp works.

My goal with this thread was to go sure, that I didn't overlook something in my schematic in comparison to the original one. As soon as I have the first preamp assembled and it's working, I'm open for improvements!  :)
German Rhodes Tech nearby Hamburg / Bremen

http://www.tinesandreeds.com

goldphinga

I've been doing some listening tests as I own both the black-face slider janus preamp and the later pot version. The earlier version sounds bigger and a bit looser, whilst the pot version tighter and more controlled with less noise.

Both great, hard to choose as they do sound so different.

Ben Bove

One potential improvement to the Janus preamp is the bleed-through of vibrato.  On suitcase models from '78 through at least '82 with the slider EQ preamp (not sure if this was corrected in the all-pot version), with the vibrato off, one can still hear a very faint panning of the signal - and even use the speed knob to help hear the change with the vibrato off.  Especially audible when the Rhodes is up against a wall.
Retro Rentals & Restorations
Vintage Music Gear

http://www.retrorentals.net
310-926-5799
info@retrorentals.net

FB: https://www.facebook.com/retrorentals.net/
IG: @RetroRentalsNet

Tines&Reeds

Yes! You are right. This is eliminated in the schematic I use as "template". It uses another kind of vactrols.
German Rhodes Tech nearby Hamburg / Bremen

http://www.tinesandreeds.com

goldphinga

Quote from: Ben Bove on November 14, 2016, 02:06:15 PM
One potential improvement to the Janus preamp is the bleed-through of vibrato.  On suitcase models from '78 through at least '82 with the slider EQ preamp (not sure if this was corrected in the all-pot version), with the vibrato off, one can still hear a very faint panning of the signal - and even use the speed knob to help hear the change with the vibrato off.  Especially audible when the Rhodes is up against a wall.

I had my Janus preamp modified to get rid of the bleed through. Much better now!

sean

Hey Chris,

I would assume that these questions are not as important as they were three months ago, but....

I have a Janus 5-knob Preamp (and the 1983 plastic-key Mark II it lives in).  The PC Board is marked 018017, which is part of the 018019 whole assembly.

For years, I had no power supply to test it, but in the blazing crater that is the RadioShack going-out-of-business-again sale, I bought 20 AA batteries, and the plastic carriers to mount them in.  I used the batteries to build a +15V and -15V power supply for the Janus preamp.  It works great.  (I also tested the preamp at +/-12V, and it works fine at the 20% lower voltage.)  The preamp draws less than 75mA with the Vibrato switched on (40mA from the negative supply, 35mA from the positive supply).

So here are the answers from your post from last November:

1.) R14 and R26 are connected to voltage. But it's not mentioned, if it's + or -.
R14 is connected to V+ (forward bias of the BE junction in Q1, and forward current in the LED are hints).
R26 is connected to V-.


2.) Does anybody know how much current which voltage goes through the LEDs (power & vibrato)?
The power LED draws current from the negative supply through R39.  Measured voltage drop through the resistor in my unit is 13.41V.  If we pretend that this resistor is exactly 1000 Ohms, then the current flowing through the LED would be 13mA.
The vibrato LED has the same 1K Ohm current-limiting resistor, so it will draw the same 13mA when turned on.  Remember, the LED is not on 100% of the time, and there is probably one and a half volt dropped across the switching transistor [Vce(sat) on the MPSA14 Darlington].

3.) There is no blocking-cap on U1. I think it's better to add one? 
The two op amps in U1 are not in the audio signal path, they are the oscillator and buffer amp for the vibrato.  Probably don't want to add any capacitors in the midst of the vibrato oscillator.

4.) C19 says 100uF while C20 says 100pF. Do you think that's a mistake and it both needs to be 100uF? This would make sense to me.  I agree, that has got to be an error.  My preamp has both of these caps replaced with 22uF electrolytics, and works just fine (but, of course, there is no ripple in my battery power supply). 

On my preamp, these caps were replaced thirty years ago as a result of what looks to have been a fire.  There are missing traces (mended with jumper wires), and black soot marks on the underside of the board.  My preamp has no enclosure, and it would be easy to short out the power connections on that corner of the board.

Sean


chigalihape

Hi Guys,

Even if this topic seems to be an old one, it's just my goal for now : Build this late Janus preamp section ! (only preamp, 'verb and trem on my fender amp are greats)

I've first "breadboarded" the Peterson, but this late Janus sounds better for me, more clear, more dynamic.

I used the same schematic as you : http://www.fenderrhodes.com/pdf/late-mark2-suitcase.pdf

I think there is a mistake for U2 gain control : with R8 46,4k an R13 464, gain is x101 !!!. I have tried R13 4,64k for a x11 gain. It's better, but there's still distorsion at high levels.

did you encounter this problem too ? Have you find a solution ?
I spent a lot of time looking for photos, informations for this late Janus... but nothing... 

Chris, would you share the files of your first post again, the dropbox links are dead ? I would be very grateful to you.

See you !
Patrick


'66 Sparkle top, '71 Suitcase MKI 73, ´77 Suitcase MKI 88  & '81 Stage MKII 73

sean

Patrick,

The gain control in that schematic is amusing.  Why design it like that?  The left side of the R23 potentiometer (the part between the wiper and the full clockwise position) is in the feedback voltage divider.  The right side of the R23 potentiometer creates a voltage divider with R25 that slightly diminishes the output signal that gets fed to C8 and the EQ section.   I would not re-implement this design; I don't see the benefit... well, since the wiper of R23 is tied to ground, you will never get much scratchy pot noise.  That would be important in a very high gain circuit.  Okay, fine, that's enough justification to build it this way.

I don't see the need to have gain skyrocket to 101, but I don't ever remember turning the volume on the preamp up past 12 o'clock.  The 12 o'clock position on the pot is probably only about 10% of the way through the full resistance of the pot (see charts like https://e2e.ti.com/cfs-file/__key/communityserver-blogs-components-weblogfiles/00-00-00-08-64/3348.Slide_2D00_0002.JPG).  The gain at this pot position would only be 5.43, and then the R25/R23 voltage divider removes 46.6% of that.

So, I commonly run this preamp with pretty low gain setting, and it seems to work very well like that.

If you want to raise the value of R13, you certainly won't cause any circuit problem; however, you will certainly decrease the maximum gain, and you might affect the intuitive smooth feeling of volume increase as you turn the knob.  It shouldn't be too hard to model in any of the spice-based simulators (maybe I will stop and do this someday).

I am surprised that a full-range maximum setting of gain of eleven causes any clipping.  If the input from the Rhodes harp is 600mV peak-to-peak, and +300mV peak at full gain would only be 3.3V.  The op amp shouldn't be clipping at all.

I verified that my 5-knob Janus preamp is indeed connected as shown in the schematic - except the schematic calls out a reverse audio taper for R23.  The pot installed in my preamp is a regular audio taper pot, and the clockwise rotation matches that shown on the schematic.

Hmmm... I don't know if I helped you at all, but I agree with your assessment.

Sean

sean


One more thought...  you aren't using the NE5534 are you?  I doubt you really need the external compensation and input balance.

A single NE5532AP would be just fine - one op amp for the input gain, one op amp for the baxandall EQ.

Sean

chigalihape

Hi sean,

My knowledge of electronics is very fragmented, so thank a lot for your analysis which advances my thinking!

For that late Janus, I wonder why the designers imagine schematics like this... the first Janus seems to have effectively only one double OPA for preamp.

May I ask you to take a photo of the component face of your late Janus, to compare it to the schem like a seven mistake game. The corrected schematic would be a useful resource which is missing for everyone on internet, at least for guys like us  ;D

Thanks !
'66 Sparkle top, '71 Suitcase MKI 73, ´77 Suitcase MKI 88  & '81 Stage MKII 73

HappyTinkerer

Quote from: sean on April 25, 2017, 04:49:44 PM1.) R14 and R26 are connected to voltage. But it's not mentioned, if it's + or -.
R14 is connected to V+ (forward bias of the BE junction in Q1, and forward current in the LED are hints).
R26 is connected to V-.


I would like to add another correction to  the schematic:
The schematic shows R28 ( on the non-inverting input of U4B) to be connected to the +V rail. This is not correct. R28 needs to be connected to -V in order for U4B to work as a non-inverting voltage comparator. I had found out about this because my implementation of the schematic would not work and I had to dig into the details of schematic and also simulated the circuit per ltspice.
U4A and U4B are setup as symmetrical comparators, U4A powering its LED during the negative half of the input signal from the VIBRATO INTENSITY pot center tap and U4B doing so during the positive half of the input signal. With R28 at +V this does not work.

I have thought about where to make the corrections gathered in this thread immediately available to others. Ideal would be to have a corrected schematic included at the fenderrhodes website, but I assume the original schematic is copyright protected so one cannot "reissue" it with corrections added.
On the other hand I think the schematic re-drawn in ltspice or Kicad could be "published", because only the document is protected but not the circuit design as such (this is made up of well-known sub-circuits).
Does anyone have profound knowledge about copyrights, circuit design protection and such?

HappyTinkerer