The Organ Forum

Hey guys.  Haven't posted in awhile but I'm sick and tired of people telling me to replace all these electrolytics on my vibrato amp board when there are none to be found that match up with the ones I'm replacing.  For starters I've got these two white "Mallory" caps with the following specs:

100uF/3VDC

5uF/15VDC

There is also a tall black radial capacitor:

20uF/50VDC

I would list the rest of them but I can't even find viable replacements for these three.  They just simply do not make them anymore.  If anyone can help me before I take a gallon of gas and douse this POS please shoot me a line... and no... there is no room to "double" up caps.. it's on a custom PC board on the assembly switch.  Grrrr.....  /action flicks lighter.

---

Noah Benzing

Hi Noah, I'm an Electronic hardware Designer by profession.  The capacitors you want to replace should be easy to find substitutes for unless you're trying to do an exact vintage restoration.  The 100uF 3V can be readily replaced by a 100uF 6.3V, or 10V, or even 16V.  Just make sure it's a radial type, not axial, and rated for 105C operation for highest reliability, especially in proximity to that big heatsink on the board.  Of course, mind the polarity when you replace it; I believe I can see polarizing marks on the assembly drawing in the photo.  The 5uF 15V can be replaced by a 4.7uF 16V or 25V, and the 20uF 50V can be replaced by a 22uF 63V.  Electrolytics are generally not used in tuned circuits so the minor value differences should not matter.  Just find the closest standard value to the one you're subbing (the most common mantissa values are 1.0, 2.2, 3.3, 4.7, 6.8) The voltage ratings have to be as high or higher than the originals, and I don't think you'll have trouble finding a size that fits since modern electrolytics are much more densly packed and thus have higher capacitance and voltage for a given physical size.  The only case where exact capacitance value counts is if its a delay or timer circuit where the cap is charged by a controlled current to produce a desired delay time; those generally need to be Tantalum type capacitors for low leakage, although you can use an low leakage rated electrolytic of the same capacitance value but much higher voltage rating. 

 expiano

Thank you very much expiano.  I'm relieved that I don't have to be so precise.  I've found a few good places to get some and the form factor *size* shouldn't be a problem.  Can be quite a challenge though tracking these little boogers down online.  I've had some luck with the locals so some of the parts I won't have to pay shipping on.  I'm thinking I should replace the power transistor as well... I believe it's an NPN but have been told that the main amplifier uses germanium transistors so I don't know.  The numbers that go in a circle on top of the unit are 001-021162-6641  (may not be in that order, they're evenly spaced).  And the number 274 is in the very middle.  That's about all I know about this transistor.... lol.  But at least I can start now with the electrolytics... thanks bro ;)

---

Noah Benzing

Noah,

 Glad to be of help.  If you need a hand finding the transistor type or a useable substitute, take a scan (or digital photo) of the schematic diagram that I can see part of, I presume its the vibrato amp schematic, and send to me at rykhyl@hotmail.com.  Don't put it on this site due to potential copyright issues.  You don't need to replace a transistor unless you know its blown.  They don't wear out like tubes or dry out like electrolytic caps.  Could save you some cash if the device is OK but something else (like a leaky capacitor) is causing the malfunction.  If you're talking about the big metal transistor in the TO-3 (diamond) case, it may be a germanium PNP.  If it's NPN, its more likely to be silicon, it depends on the age.  The long part number is the Hammond internal number; the 274 is not recognizable.  As I said, if I can see the schematic, I can likely tell you what it probably is just from the circuit structure.

 Regards,

expiano

Once again.. thank you for your help expiano.  I purchased three replacement capacitors for the tall standing radials on this board that I mentioned and installed them.  No luck on getting rid of the noise however.  could it be possible the transformer is bad?  How faulty are these in general?  There are two numbers on the metal housing on top of the transformer, one under the other 003-031145 and 549-6807  (sounds like a phone number.. lol).

 Also.... a very interesting thing I found was that C803 on my board is a 20uF/50V cap where they have a 5uF/15V cap on the schematic.  On my board C805 is a 5uF 15V cap. I double-checked this and they are in opposite spots!  You can clearly see the tall black 20uF 50V cap on the board where it shows that a 5uF 15V cap should be present.  Was someone at the Hammond Organ Company smoking crack or is it just me?  I'm really confused now... should I try changing them around for kicks and giggles?!?  Also... got the number from the top of the metal can transistor... and I think enough of the schematic is here in photos to see possibly what kind of transistor this might be or what kind may be the closest compatiable one available.  One other thing of interest to note here is that I'm using a T-500 Service Manual and have noticed some differences in the assembly switch wire coloring scheme.  Also R821 on my vibrato amplifier board is a fixed value resistor not a variable resistor like in the schematic I have.
Copyright violations?  The Hammond Organ Company has been dead for decades my friend.  Even Lars *Captain Foldback of Hamtech* has a treasure trove of schematics on his personal website.  I don't believe Hammond Suzuki supports any of the tonewheel generator models from 85' and below and thus have left it to us to suport each other.  To not post my findings here would defeat the entire purpose of my existance here as well... lol!  Besides... what are going to sue me for?  My 2 M-3's and L-103?  Oh we can't forget the broken T-212 I have... hehe ;)

I guess I should replace EVERY component on this board.. lol.. why not.  I think you're right though about the power transistor not being bad.  But I'd still like to find out more about it and what kind it is.... I'm curious!! :)))

---

Noah Benzing

Noah,

 Cap mixup sounds like a flub on Hammond's part unless some previous repair was attempted.  I only picked up on your posts via this current thread;  what is the problem you're experiencing?  Is it outright not working, or do you get proper level sound but just suffering distortion or noise?  C803 being 20uF will only extend the bass response of the driver stage by a factor of 4 (not boost like a bass control, but just extend the bottom end).  Since the input impedance of the driver stage is 3.3K + the transistor input impedance, the low cutoff with a 5uF was already down around 10Hz so you won't hear it.  C805 is worrisome since you have a 16V cap across a 22V supply; fix that immediately.  The vibrato recovery circuit doesn't seem to provide lots of voltage gain (150mV in, 250mV out), but the under spec cap could generate noise transients into the supply rail if it is sitting on the verge of overvoltage breakdown.  The Vibrato driver circuit is a relatively high gain, single-ended driver (14mV boosted to 1V on Q804's collector), with little or no power supply noise rejection.  If the supply source to it is noisy, that would bleed into the audio path and become audible. 

A quick and easy way to chase a noise problem is to use a pair of cheap CD player headphones and a series circuit of a 22uF capacitor (25V or more), a 220-ohm resistor, and a 10K pot to the tip of the headphone plug.  Connect the headphone GND (metal shaft of plug) to the chassis, and with the 10K at maximum, alligator-clip the + end of the capacitor to the input end of the driver amp.  Play a note and listen to the headphone, adjusting the pot until you can hear the note, and listen for noise.  If its not there, turn the pot back up and move the clip to the + end of C803.  Adjust until you can hear it clearly and check for noise.  If its not there,  turn the pot back up and move the clip to the collector of the power transistor (the heatsink since it looks directly attached).  Again, adjust the pot (or not if the signal is loud) and listen for the noise.  The low level signal at the input may be rather faint;  I've use a small audio amp for that but that's getting a bit more complicated;  earphones are pretty sensitive and you'll likely be able to hear something even as small as 10mV.   You can use the same procedure on the recovery amp, starting at the input at C807, and moving forwards towards the output signal at C809.

I've now seen the schematic since I found a T500 series schematic on the web.  The power transistor is a high voltage medium power NPN in a TO-66 metal case, originally a (long obsolete) RCA 40423;  it is readily subbed with an NTE-124.  From the assembly drawing, it looks like the T500 series used a plastic case transistor of the same ratings, although the operation voltages suggest you could get away with a 100V part.  Certainly I don't see why they used a 300V part unless the transformer kicks back some sort of transient due to momentary open circuits on the load side.  Oddly, in the T500 schematic, the base DC voltage indicated on the schematic is only 0.3V higher than the emitter DC voltage, implying a germanium transistor.  It's possible Hammond may have started off with a Germanium NPN power transistor (somewhat rare) and changed it later to a far more common silicon NPN power device.

 Good luck,

exp

Noah,

Thanks for the wmv.  If I understand correctly what you did in the experiment, you demonstrated good audio at the input to the vibrato driver, but got a badly distorted signal at the transformer?   That signal sounded like something being heavily overdriven.  Remember that the signal at the transformer is almost 100 times that at the input;  make sure you back off the gain of your signal tracer, otherwise you're overdriving the tracer and making it distort.  If you already did this, and the sound I hear is really what is coming out of the driver amp, sounds like it's operating with starved bias somewhere.  Use the T500 diagrams and with a DC voltmeter, measure the DC voltages on all of the transistors in the amp.  Voltages are marked on the schematic at each transistor node.  For instance, Q801 should have 1V on the base, 0.5V on the emitter (clue that its a silicon device), and 2V on the collector.  Q802's base is tied to the collector and should also be 2V.  THe emitter is at 1.4V (also a silicon transistor), and the collector should be at 10V, which is also tied to Q803 base.  Q803 emitter should be 9.4V, and its collector is the 22V supply.  Note that all these voltages are approximate and will vary from unit to unit and with age.  THe key point to remember is that the emmitter of an NPN silicon transistor will always be about 0.5 to 0.6V below the base voltage.   The three transistors 801,802,and 803 form a linear voltage amplifier with a gain of about 40 times, that is, a 10mV signal on the input at Q801's base will be magnified to 400mV at the emitter of Q803.  The power transistor Q804 further magnifies the signal by a factor of 3, so the total gain from input to transformer primary side is 120 times.
Now, C802 is a filter to remove audio from the DC feedback loop around Q801 and 802.  Without it, a significant amount of 801's collector signal, which is 180-degrees out of phase, would show up at the base through R802, effectively reducing the gain.  I don't think it would distort however since the DC gain and operating point is not affected.  The other cap, C803, would be a critical one to go leaky, since the emitter voltage of Q803 is quite high (9.4V).  ANy appreciable leakage and Q804 would be driven heavily "ON" into saturation.  If that were the case, the distortion would be very ugly as the transistor would be operating in something that looks like class "C", or almost like digital switching.  To check for this, measure the DC voltage on the minus side of C803; it should be at 0.9V same as the base of Q804.  ANy leakage thru C803 will cause a voltage drop across R811 and you'll see a DC voltage higher than the base DC voltage of Q804.  Replace C803, and mind the polarity; reversed electrolytics are very leaky.  The plus end should connect to Q803 emitter even though the schematic does not show this. Hopefully its marked on the assembly drawing or silkscreen print on the PCB itself.

I'll copy this note on the forum thread as well.

expiano

Noah,
 

I've copied this from our direct email discussion as it contains info useful to other Hammond owners that tinker with their boards.

Don't be hard on yourself, I just found the biggest boner from Hammond and the source of your confusion.  The T500 diagram I got off the web shows both schematic and assembly of the board in x-ray view so you can see the connections.  According to the actual copper connections, the 5uF feeding the base of the power transistor Q804, is physically identified as C805, even though schematic shows C803.  In the photo you just sent, the 5 uF should be installed where I can see the 22uF.  C803 on the PCB is across the 22V supply and should be the 22uF 50V.  C802 appears to be correct.  I suspect this won't solve the problem you're chasing but it's good to know. 
 
It might be a good excercise for you to go through the schematic and assembly diagram, tracing the copper pattern connections on the PCB to various components, in case there's other bone-headed mistakes in the drawings.  Also, remember this is a T500 drawing with a plastic power transistor, not the T200 with the metal TO-66 on your board.  There may be other differences in the design for the T500 but I suspect they changed as little as possible (I work in the industry and know how designers think).  It's common practise to make small changes to a design so a different part can just drop in when the original goes obsolete, which is what happened to the 40423 that was there originally.  I've attached the T500 file I've been using, which was acquired from the Captain-Foldback website.  

Cheers,
 
exp

Hey expiano :)  Finally got the time to go to the local electronics store and do a little bit of shopping.  I didn't find EVERYTHING I needed but alot of stuff I found.  The guy who ran the store picked out this transistor and half of the caps and resistors I ended up buying.  I've noticed some mistakes in value selection for some of the components and I'll have to take them back and swap out but he tried to be as helpful as he could and was very friendly and interested in what I am doing and for that I think Affiliated Electronics in Tulsa is the best place in Oklahoma to buy parts and gear.  *Well deserved credit there*

 Now all I have to do is hope I got the right stuff.. hehe :))  Pictured is the transistor I bought.. it says TCG-124 but on the back it says NTE (New-Tone Electronics).  But on the front of the package it says it's equivalent to an ECG-124... hm.... /action shrugs.  Also, from your perspective, what's the best way to mount this puppy?  I removed the whole assembly once but couldn't get the rivets out of the old transistor and didn't want to break them until I could figure out wtf I'm doing per say... but from what it looks like the rivets are very important since they physically connect the collector to the PCB.  Any thoughts?  Guess there's alot of hardware mounting techniques to learn.

Hope you had a wonderful Thanksgiving :)

---

Noah Benzing

Noah,

Don't worry about the transistor prefix;  it lines up with the old ECG and newer NTE numbers so long as the numeric part is the same.  You'll have to (carefully) drill out the rivets of the original and replace with screws/nuts  of appropriate size and length (I would guess 1/4-inch 4-40 machine screw).  Before you drill it out, check for continuity between the heatsink and the existing transistor.  If its connected, no problem, however if its insulated, you'll have to carefully install the transparent mica washer and use plastic isolating shoulder washers on the nuts and bolts to ensure the transistor case does not electrically contact the heatsink.   In higher power applications, you'd also have to apply some thermal grease (white non-conductive goop) to the bottom of the transistor case and to the washer, but I don't think you need to worry about that here.  Just make sure all your replaced parts have good solder joints with good solder wetting.  With older surplus parts I occasionally pick up or salvage, I scrape the trimmed metal component leads with an exacto knife (kinda like scraping a carrot but in miniature) to expose shiny metal to the solder to ensure a solid contact. 

 I had some more thoughts about this problem; you demo'ed the sound with the board installed and hooked up to load.  If there was something wrong with the load,  that could also cause the distortion I heard.  One thing to try after you do all the replacing and re installation is to disconnect the vibrato scanner load and replace the load on the driver with a suitable power resistor, say about 300 ohms or so, and check the audio again.  It should be nice and clean; if it isn't, the transformer may be your next suspect.  I'd be surprized since this is a relatively low power circuit compared to your typical audio power amp, but the transformer can't be ruled out completely.

 take care,

 exp