LinkBack URL
About LinkBacks
Reply With QuoteIf you can't see it on a scope, did you try the internal amp with different speakers as well?Originally Posted by spy
I have a problem with a Fender Rhodes (Jordan design) amplifier, the schematic can be found here, Figure 11-17. The problem is that I have a soft distortion on the sustain of the notes, especially when I press two or more keys together. The sustain of the notes is not soft and sweet as it should be but a little bit grainy. Hope you can understand my description.
I can't see anything with an oscilloscope and generator but I can hear it when I'm playing. The problem is not on the piano as I have tested it with another amp and other speakers, too.
Is it time for el.caps or is it something else? The caps are the originals but I can see two new drivers over there. The output transistors are the RCAs but the drivers are Motorola MJ4101. I don't know what the originals were as there's nothing on the schematic.
Last edited by spy; 06-08-2012 at 04:38 PM.
If you can't see it on a scope, did you try the internal amp with different speakers as well?
Try adjusting the distance between the pickup and the tine. If it is too close it will distort.
Yes I have tested the amp with different speakers, also I have tested the piano with another amp. The problem is definitely at the internal amp.
What you describe sounds like crossover distortion to me, but I would think that you'd see that with a scope.
If the drivers have been changed, I'd probably review whatever was done to the amp before and see if there are any problems there.
Also vote for crossover distortion.
Not as easy to "see" on the scope, simply because everybody expects to see clipping or dramatic wave deformation, and this is more subtle.
The amp design is somewhat poor, to begin with , maybe the pre 1969 date explains it.
You must measure Vdc across any of the 0.27 ohm resistors , no speaker, everything on "0".
I bet now you have 0 mV.
Add a 100 ohm preset in series with the 1N816 bias diodes , set it on 0 ohms and slowly raise its value until you measure 5mV across any 0.27 ohm resistor.
Be careful and try to mount it somewhat firm (you can cut a track, add a couple holes and solder it there) because if it opens your amp blows overbiased.
If you dare to mod it a little I may add the 2 or 3 parts it needs to be more "normal", such as a Zobel, some bootstrapping and hum filtering.
Juan Manuel Fahey
I do dare! Let's solve the problem with the crossover distortion and then continue with the mods. I'll measure Vdc and if it's 0mV I'll add the preset, I'll let you know! Thanks!
Juan, you're right, across each .27R I have 0mV. I added the trimmer and I set it where I measure 5mV on one resistor and 2.8mV at the other. Is this ok or do I have to set it higher?
As the amp is still on my workbech, please let me know how to add these 2-3 parts to make it more normal.
OK.
Referring to the schematic you posted:
1) split the 150k resistor going from Qo/Q9 emitters to -V rail into 2 x 75K ones , you don't need new holes, just put a leg from each resistor into one of the original holes and twist (and solder) the free legs together.
Then connect a 47uFx50V capacitor from that centerpoint to ground.
+ to ground, - to centerpoint.
With this you lower hum *a lot* when clipping, which makes things worse because it intermodulates the signal like a horrible 50 Hz tremolo.
The value is not critical; instead of 75K you can use 68K or even 47K , which usually are more common, or a 100K in series with 47K.
2) similar to (1), split the 4K7 resistor from Q12 base to -V into 2 x 2K2 or 2K7 , same as before.
Now connect another 47uFx50V from the centerpoint (cap minus leg) to the speaker out rail (cap positive).
That's the track which touches Q15 and Q16 collectors, the 22 ohm resistor, the 10 ohm one, etc.
This creates a bootstrapped load for Q11, linearizes the amplifier and, nost important, allows full drive on the negative side, which otherwise was impossible.
3) add a Zobel network.
Classic values: solder a 10 ohm 1/2W (or 1/4W) resistor in series with a 0.1uFx50 or 100V cap.
One free end goes to the speaker out rail; the other to ground.
Preferrably as direct as possible to "real" ground which is the centerpoint of the big filter caps and as far as possible from the signal input ground.
In a nutshell: your amp will have less hum, clip cleaner, and have a little more power, plus being more stable.
Not bad.
Juan Manuel Fahey
Yeah, but will it still get the sustain I need?
Education is what you're left with after you have forgotten what you have learned.
Thanks Juan for the detailed mods!
Enzo, what do you mean with the question? That sustain will decrease?
EDIT: I don't have good news as the problem with the "distorted" sustain is still there. The bias mod didn't work.
Last edited by spy; 06-13-2012 at 09:03 AM.
Was just thinking the power amp is also the pre-amp power supply.
Is it possible the 1 watt 1k5 dropping resistor has gone a bit high in it's old age?
Maybe the pre-amp is a bit starved ?
That (I presume 6.4uF 40v) filter cap on the amp input & supply TRS jack too could be a bit tired.
Just trying to throw a different slant on the diagnosis !
I have tested the amp with a guitar preamp (with preamp's power supply disconnected from TRS jack) and I have the same results with guitar chords, the notes do not sound separated, there's a slight distortion between them.
Last edited by spy; 06-13-2012 at 11:11 AM.
Well, all you did is good, so no regrets on that.
But the defect *may* be not electronic but mechanical.
The *other* defect which sounds very much like crossover distortion is a dirty/scratchy/misaligned voice coil.
After all, that speaker is, what, 40 years old?
Plug your Rhodes amp into a new speaker, any guitar type one will do.
Juan Manuel Fahey
well that rules out my suggestion .. I'm still trying to consider some simple stupid thing .. well now the experts have left the building
you are still confronted with this...Power supply caps ~dirty connections~ speakers most of these things I'm sure you would have already covered
I had a later model recently which was straight forward- din socket disintegrated at keyboard shorted out regulators one of which exploded
easy fix by comparison ..couldn't help noticing that vintage electrical smell while marveling at the fact that these pieces of vintage gear
despite their shortcomings still sound fantastic in the hands of a good player and are still going !
Their character and sonic signature still in demand.
I have tried an other speaker, the problem is still there. Also, the original speakers have been tested with another Rhodes amplifier and they sound great.
Let me know if you have anything else in mind. Power supply caps are still the originals, I think it's time to replace them!
When in doubt, change them out.
I'd go back to the new drivers. If someone had replaced them before, there may be an issue there that was never solved or was created by the new parts. What voltage reading do you get from base to emitter on the two driver transistors?
Spy, my "sustain" comment was merely a joke. It had absolutely no technical value, and nothing to add to your situation, sorry. It is a sort of shop joke here, refering to a stereotype young kid coming into the repair shop complaining he can't get enough sustain from his amp. Up there on our list with the kid who got a Strat Pack three months ago and proclaims "Music is my life."
Education is what you're left with after you have forgotten what you have learned.
Not forgetting the guy who asks " I have XXXXXX problem (which may be *anything*, including bad breath) ... it's the transformer, isn't it?... "
Juan Manuel Fahey
Enzo, ok! Now I got it!
52 Bill, you may be right but I don't know what were the originals.
Does anybody know?
I did the measurements and Vbe on both transistors is .54-.57V, they are MJ4101.
I don't know what the original generic part numbers were either. Your Vbe voltage readings are in the ballpark though.
Are these transistors in the TO-66 cases, like mini power transistors?
Yes, you're right, MJ4101 are in the TO-66 cases! Let me know what do you suggest!
With all that you have already done, there's not much left to suggest.
I remember once working on an old stereo amp that did something similar. Similar case, as it had been repaired once before I got it. Pulled out what little hair I had left on my head until I went back and checked what was done before it came to me. While there was nothing that I could see that was wrong with the earlier repair, I decided to replace the newly replaced transistors. That fixed the problem. The SK series replacement transistors that were used apparently didn't have enough gain to work in the circuit as designed.
Perhaps if you go back and try replacing the drivers with a different type it may clear up your problem output. TO-220 transistors were designed to fit the TO-66 pin and screw spacing, so you could try a TIP series part in there. Just watch the voltage and current specs of the replacements.
Fully agree with 52Bill and add: all the schematic says about the drivers is "1A 60V"; apparently those 2N4101 fit the bill.
Exact match (per schematic) would be TIP29A but I would splurge and use TIP31C , specially because it's the "preferred part".
You can , as noted above, mount TO220 instead of the TO66 case ones, by carefully bending B/E legs down into original PCB holes. Use the proper rectangular mica, of course.
As of the Collector, how is it connected?
Electrical contact through the screws? Insulated screws? Post a couple pictures.
If contact through screws, you can cut the center collector leg; if not, you may bend it upwards, cit it leaving a 1/8" stub and solder some wire to it, if necessary.
I am replacing TO3 metallic 2N3055 in my old amps with TO218 or TO247 transistors with similar small mods; so I know it works.
Only problem is that *sometimes* the metallic case is used as a "bridge" between 2 pads, so if you use a 1 screw TO220 later the amp does not work.
Easy to solve with a small piece of wire ... if you are aware of the possibility.
Good luck.
Juan Manuel Fahey
Juan and others there is an unrelated You Tube clip that will give you a visual idea of whats going on.
He had a problem with the differential pair having it's earth via a panel mounted jack.
Fender Rhodes Jordan Amplifier Tech Note - YouTube
Just replaced the drivers with a pair of TIP31C, thanks Juan for the suggestion, and I'm feeling that the problem is gone. I haven't heard the amp with the piano but with a guitar preamp connected and a guitar plugged in, the notes are clean and separated as they should be! To be 100% sure, I'll test it with the Rhodes and I'll let you know! Hope everything is fine and don't come back disappointed.
Edit: Forgot to mention that on new drivers I have Vbe around 0.63V on both of them!
Last edited by spy; 06-26-2012 at 10:52 AM.
That's great to hear, I hope it works right with the piano as well.
Did you leave the earlier bias modifications in place as well?
So, it was the drivers! It works great with the piano and the amp is clean. Thanks for your help guys!
52 Bill, yes I left the bias mod in place, it's a great mod. I rebiased the amp after the drivers replacement.
So, it was the drivers! It works great with the piano and the amp is clean. Thanks for your help guys!
52 Bill, yes I left the bias mod in place, it's a great mod. I rebiased the amp after the drivers replacement.
There are currently 1 users browsing this thread. (0 members and 1 guests)
Posting Permissions
Bookmarks