4 pin rhodes power amp gets hot on one side

[Steve Conner]

Yep, the MJ15025 would be a pretty bombproof option.

I'd increase the 2.7 ohm resistors to 3.3, and if that doesn't do it I'd decrease the 820s. If you tried to do all of the adjustment with the 820s they would end up about half the value, and dissipate a lot of power.

In an old design without much feedback, like this one, you can hear it if the bias is too cold. It sounds remarkably similar to a tube amp that's been biased too cold: kind of thin and sputtery at low volume, and harsh at high volume.

[gbono]

http://music-electronics-forum.com/t22175/

[Steve Conner]

Great thread. Now I think about it, those 2.7 ohm resistors do need to go up to something like 12 or 15. Not 3.3.

[Skiroy]

Awesome thread.

Couple questions

1. What is crossover distortion and how do you test for it?

2. When it comes to the world of audio I know capacitors quality effect sound quality,but what about resistors? Do I have to use highest quality resistors for this modification? I am very concern with sound quality and keeping its original sound.

3. How hot is too hot for they transistors to get? Maybe I am just paranoid? But one side shouldn't be cool and the other hot to touch. Maybe I will try cleaning the mica and using thermal compound. But if I had a heat gun what temp is to hot after 10 mins of use?

4.Does anyone one where I can get the 3 prong plug that goes on the rhodes power supply without paying $30.00 from vintage vibes?

[Enzo]

These?
05CL3MX Switchcraft DIN Connectors

Or these?
SL403MX Switchcraft XLR Connectors

I am working from memory, but I seem to recall the Fender used the different types of connector at different times. Maybe Rhodesplayer or someone would know better.

If you can;t hear the crossover distortion, then there isn't much to check for. One checks it with a scope and a sine wave generator. Or for that matter with a distortion analyzer.

Personally I wouldn;t worry much about resistor brands here.


I doubt the need to clean and re-silicone the transistors, but... You say they get too hot on one side. Just the transistors? Or does the heat sink get hot too? If the heat sink is real hot, then the transistors are thermally connected to it already.

If the thing heats up just sitting there at idle, then somethiing is underbiased or defective. If it only heats up while playing, we need to decide if it has a problem or if the other side has the problem and ought to be getting equally warm. Two channel, do they sound alike? Or does one side differ from the other?

[Skiroy]

No this
Vintage Fender Rhodes Suitcase Receptacle

And isit even possible to get a original matched set or delco germanium transistors. They might be $30.00 but that beats a $75.00 kit from vintage vibes for just one power amp. Plus all my resistors test okay. Im not sure how to test the one capacitor but its not shorted.

[Enzo]

Silicon transistors are about $4 each. Add a few cents worth of new resistors, and you save even more money. And it will be even more reliable. COnversion is not complex, and you can find it in numerous places. Even Fender sent out a bulletin on the conversion at one point, they recommended ditching the germanium too.

A matched set of original NOS germaniums? FOr only $30? Good luck.


For the mains power plug, consider mounting a common IEC male like on most gear these days. Then any common detachable power cord will work.

[Skiroy]

Silicon transistors are about $4 each. Add a few cents worth of new resistors, and you save even more money. And it will be even more reliable. COnversion is not complex, and you can find it in numerous places. Even Fender sent out a bulletin on the conversion at one point, they recommended ditching the germanium too.

A matched set of original NOS germaniums? FOr only $30? Good luck.


For the mains power plug, consider mounting a common IEC male like on most gear these days. Then any common detachable power cord will work.

Well I dont really know what resistors I need or need to change. Plus I dont know where to get a good quality reliable matched set of transistors(Big 2 legged ones not small 3 legged ones)? I have read various things about cheap imitations. I want the best sound for recording. If you can direct me to where to find the parts and info I would greatly appreciate it.

[gbono]

Digi-Key - MJ15023GOS-ND (Manufacturer - MJ15023G)Take a picture of the board and post it here and someone will show you where the resistors are that need to be changed. Transistors, (Si) silicon, to replace the the (Ge) ones in your circuit can be purchased at Digikey - see link above.

[Skiroy]

Digi-Key - MJ15023GOS-ND (Manufacturer - MJ15023G)Take a picture of the board and post it here and someone will show you where the resistors are that need to be changed. Transistors, (Si) silicon, to replace the the (Ge) ones in your circuit can be purchased at Digikey - see link above.

Okay but is that a good brand? Is it going to sound different than my originals? If I order 2 they wont be matched right?

And if the other power amp is Germanium I would have to convert both sides or they will sound different right?

Here is my power amp.

[R.G.]

Colin, slow down. You're trying to run before you can walk. Reading the internet for hours won't help without some basics under your belt.

1. Are these red transistors bad because of the reading and is it normal for them to get this hot?

There is no good way to tell without a lot more work. They may be getting hot for quite a few different reasons. The thermal runaway thing that was mentioned is always a lurking problem with germanium, and is in fact one reason germanium is no longer used by the electronics industry. "Hot" can be both a symptom and a cause. More facts are needed.

2. How do I tell if they are silicon or germanium

As noted, you measure the junction voltage. This involves a number of things, but it's no longer needed because those are DTG110Bs, and are germanium.

because why would the other be written in green that stay cool?

The color of the ink doesn't mean much. It probably changed from production run to production run.

And if the greens are germanium and I have to change the red ones to silicon will the Left and Right sides sound different?

Probably not. A lot depends on the rest of the circuit. It is an internet myth that an amplifier sounds different just because it has a germanium device in it. The rest of the circuit matters; the individual devices are very much subservient to the circuit as a whole.

3. The power amps on this rhodes have a adjustable pot like on the the power supply. Is this used to set bias on these as opposed to having to soldier in addition resistors? And could this be my heat problem?

It's possible. But you have not ponied up the schematic, even after being asked a few times. We don't know if what we find on the net matches what you have.

4. If this pot is for bias or even if I have to do it the old school way where am I measuring voltage while setting the bias?

Unknown, captain. We don't know what circuit/schematic it is.

What terminals,wires ect?

See above.

Germanium power transistors are rare, and expensive. Matched sets are even more expensive. And getting even matched sets of the actual thing may not guarantee you "good sound".

I did this because I have tested the transistor and got different values and sometimes it would beep while other times it didnt. So I though maybe temp had to do something with it. And it appears it does change the value and wether the DMM will indicate a short or not. So how can I know if this thing is testing okay or not?

As noted, you have (re)discovered the big temperature sensitivity of germanium. However, whether a multimeter indicates a short or not depends on the multimeter. They're not all the same, and in general they have poor discrimination of low ohms from very low ohms. How can you know if this thing is testing okay or not? You can't not with only that multimeter looking for "shorted" beeps. There are some conditions where you can know it's bad (like, all pins show shorted together, or C and E are shorted) but you can't tell if it is definitely OK.

There are some multimeter tricks for testing forward/backwards on the two diode junctions that make up each transistor, but this requires some subtlety on the meter and knowing something about how your personal meter acts. The bottom line is that you find which DC polarity your meter puts out when it's doing DC ohms testing, and what "resistance" it reads for a diode - germanium in this case. Then you put the known-positive meter lead on the base and check to see if both collector and emitter show open circuit. If they do, you test for open circuit between collector and emitter with both polarities of the meter leads. If that works, you finally put the negative lead on the base and then look for forward-diode reading on the meter from base to each of collector and emitter. That set of tests just shows "not completely dead" at the low voltages and currents of the meter. It's not conclusive for positively good in power circuit operation.

1. What is crossover distortion and how do you test for it?

It's a failure of the two output devices to hand off output voltage smoothly just as the signal "crosses over" from positive to negative and vice versa. To properly test for it, you need an oscilloscope and a distortion analyzer. But it's easy to hear as an ugly "gritch/buzz" that gets worse with smaller signals and less noticeable with big ones.

2. When it comes to the world of audio I know capacitors quality effect sound quality

You "know" that because advertisers harp on it. The plain fact is that capacitors have a very minor effect on sound quality compared to other things.

but what about resistors? Do I have to use highest quality resistors for this modification? I am very concern with sound quality and keeping its original sound.

Most resistors are very nondistorting, no matter what you hear in the advertising and hifi-tweeko forums.

The plain facts are that this design is not a subjectively high quality amplifier, in that it has lots of its own distortions. It does have a recognizable sound of its own, but that's largely a function of the driver transformer, much more than the output transistors.

3. How hot is too hot for they transistors to get?

Silicon, junction temp of 125C, germanium, 80C, or thereabouts. That's on the chips inside. You don't get to measure that directly, as you can only get at the case. Germanium power devices can die after getting over 50-60C on the case. They can go into thermal runaway and then self-heat themselves to death internally before the case and heat sink have time to heat up.

Maybe I am just paranoid?

Maybe. But just because you're paranoid doesn't mean that they are not really out to get you.

But one side shouldn't be cool and the other hot to touch.

You are correct. By the way "too hot to touch" is about equivalent to saying "over 130F surface temperature". How much over is not clear until you get blisters or hear sizzling.

Maybe I will try cleaning the mica and using thermal compound.

If it does not use thermal compound that could be a reason it's running too hot. Another reason might be loose mounting bolts/screws. Power devices need to be clamped firmly to the heat sink, and they need an insulator (the mica) as well as a way to fill the microscopic air voids on each side of the mica. That's the thermal compound.

But if I had a heat gun what temp is to hot after 10 mins of use?

If you had a bicycle, would it take too long to swim there? It does not compute. Heat guns have some limited uses in debugging and repair, but heating an already overheating power amp is not all that useful.

4.Does anyone one where I can get the 3 prong plug that goes on the rhodes power supply without paying $30.00 from vintage vibes?

You got really good advice: replace it with an IEC power entry. By the way, in doing any of this you are literally risking your life and the lives of anyone around this amp from your time forward by betting you will get the AC power wiring done safely. Think about that. You or someone you care about could get killed if you don't do it right.

And isit even possible to get a original matched set or delco germanium transistors.

Delco is long out of business. Only a few places in the world still make germanium. One is Germanium Power Devices. They are very expensive, as they specialize in making military replacement parts. Original devices may be had. It's a matter of luck. Some people find gold nuggets while on walks in the country. It's possible, but it's not really practical. You got good advice: change over to silicon PNPs.

You really, really need to go read this link:
http://music-electronics-forum.com/t22175/

Okay but is that a good brand?

Digi-Key is not a brand, it's a distributor. The MJ15xxx series was originally from Motorola back when they made transistors, and is still made by On-Semi, which Moto spun off. "Good brand" has little meaning in this context. There are other type numbers and manufacturers, but those are one good substitute and distributor.

Is it going to sound different than my originals?

Possibly. However, it's going to be very, very difficult to tell. The *design* of the original has so many oddities in it that the transistors won't be the biggest contributor to "sound" in most cases.

If I order 2 they wont be matched right?

No. You have to pay someone to match devices or buy extra parts so you can match them yourself. The big question is then what do you match FOR even if you know how. There is the other oddity that if they came out of the same wafer/lot from the original maker, they may be very close. The other oddity is that in this circuit, you can measure the DC offset at the output and largely match them by getting close to 0V as long as the +35 and -35 are balanced. The other 57 oddities are...

And if the other power amp is Germanium I would have to convert both sides or they will sound different right?

Again, maybe, but probably much less than you think. Even a touch of crossover distortion will make them sound very, very different.

You are asking an enormous task of yourself. Getting it working is going to be a big job. I nearly always tell people that having a working amplifier that's a little different at least gets you back to making sound while you go look for the theoretically perfect part that may or may not ever turn up, and may not be affordable when it does.

[Skiroy]

Thank you so much. So as far as having to have someone match the transistors I might as well just pay the money for the vintage vibes kits as they have everything I need in them and I can stop driving myself nutty for saving a couple of dollars.

[R.G.]

That may be wise. There are many ways to pay for things. Sometimes money is the cheapest way.

[gbono]

Sources for Ge transistor are far and few between - I would expect the "reliability" of Si transistors made in 2011 to eclipse Ge devices. It's your money and you you get to decide what to purchase. I usually find that Digikey sells you devices that all have the same "date code" = date when the devices were assembled. No guarantee but I haven't had any issues with this method and have even measured devices purchased this way to find they were "matched" - within 10% that is for DC HFE, BVceo, early voltage, Vbeon, etc. I really doubt vintage vibe is selling product that is matched any better than the above "process". Of course a switch to Si transistors require that you have the know how and tools to "re-design" your application. BTW this is what Rhodes/Fender did with later versions of that amplifier - so I'm assuming they felt Si devices sound okay .

[Skiroy]

So my biggest concern is I want my rhodes to sound like an Old 1975 Rhodes for recordings. If I go and change to new transistors,caps and resistors is it still going to sound like an "Old Rhodes" or is it going to sound like a 2000 Rhodes?

I mean al these original components being changed out to modern made components are really not going to mess with the sonic character of the sound we are all trying to reproduce on our favorite old records?