• WARNING: Tube/Valve amplifiers use potentially LETHAL HIGH VOLTAGES.
    Building, troubleshooting and testing of these amplifiers should only be
    performed by someone who is thoroughly familiar with
    the safety precautions around high voltages.

Hybrid amplifier help...

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Hi,
I have been experimenting with a hybrid amplifier design, After tragically loosing all the magic smoke a couple of times i decided to switch the positive and negative rails and now the magic smoke escapes no more (still think the positive and negative are the wrong way round now though but it works.. ) and i have output! not exactly good quality.. not even sure it could be classed as music.. but it is there.. and loud... it sounds like the amp is clipping or the speaker is broken but the speaker is fine.

Schemnatics of the original are atached below, my amplifier is slightly different as i just used things i had lying around that were already built, i will try and get updated schematics posted asap!
Basically The SS output module i am useing apeared to only loose its magic smoke from one resistor BUT i am not 100% sure that the driver transistors arent damaged so i will test them next after i find out how..lol

The tube stage..
basically the 1K resistor is a 120 or 150 Ohm in mine because i simply used a circuit i had already built for something else
How does this affect it?
Could this be my problem?

I'm thinking mabey it is stoping the amp from geting enough voltage or something because i have to have the source cranked right up in order to get anythign that remotley resembles music out of it..
Thanks in advance,
Owen
 

Attachments

  • 6922.gif
    6922.gif
    6.3 KB · Views: 822
Administrator
Joined 2004
Paid Member
Hi DoomPixie,
Your second diagram shows your Vbe multiplier is shorted out. Error?

The diode in the emitter of the Vbe multiplier should be removed (short). The input impedance is too low to the output stage. I would use newer output devices that are more linear. MJ15022 / MJ15023 would work well here. The newer MJ21193 / MJ21194 would do well also. I don't think those drivers can deliver the peak currents that might be required, they could still be used for the first transistors in a triple darlington config, or another setup as you choose. The 270K resistor should be 270 ohms?

Many hybrid designs show the fault of a low output stage impedance, increasing this will bring better results. Feedback will improve the sound (from experience). You don't need much, try 10 dB as a start.

-Chris
 
the output cap is doing ok.. the 270k is infact 270ohm.. my mistake..
the output stage isnt the same.. there is no diode itn eh output stage on mine.. schematics of the output stage are enclosed.. i changed the 2SA984 for a BC327.. think it was a 327.. it works so much better after changeing that transistor.. the midrange is soo sweet on the amp, there is just something really wrong somewhere and the bass and trebble are all messed up.. sounds almost like a bad connection in a lead does when you shake the lead around now.. on and off.. except the midrange which is perfect..
anywah, here is the schematic for the output stage. :) the 330 ohm in that pic is 270ohm.. had to change it after if released its magic smoek and didnt have any 330ohm resistors :(
Owen
 

Attachments

  • ampboardval.jpg
    ampboardval.jpg
    33.7 KB · Views: 609
anatech said:
Nah, eliminate that caps and reference the output stage to ground.

In the case of direct connection to the load , it is wise to use some zeners in the feeding of the voltage rails for the 15k /12k resistors , feeding the VBE , if not , the offset voltage will vary with the rails voltage , that are also dependent of the mains voltage fluctuations ...
 
Administrator
Joined 2004
Paid Member
Hi Owen,
I suspect the tube stage is going crosseyed :xeye: trying to drive the output stage. Things will improve if you can increase the impedance seen by your tube stage.

Tube_Dude,
I have build a few, if you reference the output stage to ground it will behave quite well. Not the same circuit, but this one has problems to solve anyway. If you redo the input to this circuit, you can lose the dependance on the rail voltages. CCS drive comes to mind for more than one reason.

-Chris
 
please slap me!
i never noticed that 15K resistor before....
I have been puzzeling over why the thing sounds better when i disconect the 12K resistor from the supply for ages! thats probably all thats wrong..lol... and theres me changeing resistors and transistors and stuff trying to work it out and gradually getting better and better sound out of it... i'll go and add that now and see if it works. lol
Owen
 
ok, i added that resistor and it got rid of the annoying problem with dc offset.. :) i was thinking something along those lines myself, the amplifier is behaveing very well at max volume now. but when i turn it down it sounds liek the tube isnt giving enough current for the SS section to switch properly.. How would i go abotu fixing this? The driver transistors are now BC327-25 and BC337-25.. changeing to these gave a pretty big improvement over the originals which i tried to measure useing hfe checker on my multimeter but the legs were too short so i just changed them for some transistors i had on hand..
So yeh, how would i go about increasing the impedance? :S also will that resistor in the tube stage make a diffrence? as i said it is meant to be a 1K but i am useing a 150Ohm still?
Owen
 
Administrator
Joined 2004
Paid Member
Hi Owen,
it is meant to be a 1K but i am useing a 150Ohm
Sets the current of the cascode stage. Watch your H-K voltage on the top tube.

Your BC3X7-25 transistors. Are they 25V? If so, try something like A970 and C2240, put another transistor stage in between that can deliver a few amps to drive the outputs. This will help raise the input impedance. To generate your bias, use a current source and current sink to isolate from the supply and make the circuit relatively independant on supply voltage. Use a relay to isolate the output stage from the large on-off voltage swings.

Just a few hints, get designing.

-Chris
 
I will change the resistor in the tube stage i think, probably easier.
the BC3X7-25 are 50V as far as i'm aware.. they come in -16 and -25 flavours amongst others, the larger the number the larger the Hfe acording to the datasheet. the problem with adding another transistor stage is that it is currently all neatly mounted on a PCB that mounts to the heatsink, which basically means i can have everythign else wired p2p. Would adding an extra ss stage increase the impedence and help with the problems i'm haveign with the amp? would adding a second tube stage help the situation?
i get what you mean about the current source and sink though, you mean inplace of the 12k /15k resistors? i can do that fairly easily. have plenty of BD139 and BD140 that would be up the the task.. i don't see how this will help with the impedance issues though? I want to get the amp running really before i start modifying it too much so that i have some incentive to carry on the project, i get bored easily with things when they wont work properly..lol..
OWen
 
Administrator
Joined 2004
Paid Member
Hi Owen,
Your supply voltages are +- 50V. You need 120V devices minimum allowing a tiny bit of safety. Another transistor stage will help to increase the input impedance of the output stage, the CCS's work to that end as well. It's all important and inter-related.
did you mean add an extra transistor stage between the tube stage and ss stage to increase the current from the tube stage?
No, I do want another stage added (see above). This will reduce the current required from the tube stage and decrease distortion. Leave the tubes alone for now.

-Chris
 
Sorry to be blunt, but that schematic has so many issues it's not even funny, and your haphazard changes to the component values are making it signifficantly worse.
BC337 and 327 are 50V BJTs in a place where up to 100V can happen across them. The bias setting is connected in such a way that if the pot ever even develops a smidgen of scratchyness, the output stage will blow up as the bias voltage will become about 100V or so. The input impedance is about 7.5k which is WAy too low for a 6922 SRPP. Then you also use a 6 times lower resistor in part of that SRPP and say it's better to change that back to the way it should be, than put proper driver BJTs, the BD139/140 being far closer to that than the BC337/327 - and youi have BD140/139 at hand. I am not even going to attempt a comment on the 'i think the power supply is the wrong way around' statement, my vocabulary fails me.
Basic laws of electricity are not subject to subjectivism (pun intended). So, what worth is it that it sounds good when only the grace of some unknown god is keeping it from blowing up? Or, are you trying to make an even less reliable (I am barely restraining myself from describing the sound!) amp than a Counterpoint SA-100?
 
Administrator
Joined 2004
Paid Member
Hi ilimzn,
Yes, correct on all counts. I was trying to get Owen started in the right direction (away from Counterpoint for one). I figured the overall design should be corrected first, then the minor details like bias after. BTW, Counterpoints version is far more scary than this even (SA-100)

Go ahead and describe the Counterpoint sound. I'm sure I'll agree with you at least 100%. :yes:

Anyhow, ilimzn, if you read my suggestions above, what do you think of the direction? Keep in mind we are stuck with tube front end and BJT current stages.

-Chris
 
no.. my supplies are +/-25V on the output stage.. same voltage the output stage was running on int he original amp. according to the spec's i found on the net the transistors i replaced were originaly 50V devices aswell so that isnt an issue? or am i missing something?

The way i originally had the power rails is the way that they are in the schematic as i traced it out fromt he output PCB's and they kept burnign out so i swapped the supply rails and now it works. My guess there is that i made a mistake traceing out the schematic and got the pin numbers on the PCB the wrong way around.

I will change the resistor in the tube stage tonight probably as it won't take long, the reason the resistor is 150Ohm and not 1K is that the tube stage is actually a prototype for an SRPP linestage i built and that circuit had a 150ohm resistor in it, that was the only diffrence between the linestage i have built and the tube stage in this amplifier that was why i decided to try this "experiment" out and see if it would work.. because i had a prototype of the tube stage already built from my previous project and i had these cool output boards i wasnt currently doing anythign with.

So I have to add another stage between the tube and the SS stuff so that it dosnt need to draw so much current from the tube stage?
This is the first time i have palyed around with a hybrid amp, just thought it would be quite fun to experiment with **Was bored**

Many thanks,
Owen
 
Administrator
Joined 2004
Paid Member
Hi Owen,
Ignore the tube stage for now, it works. Just make sure you are not stressing the heater cathode ratings.

I feel you need more headroom with your transistors even at +-25V. So let's concentrate on the current amplifier stage. Add driver transistors and compensate the circuit design for these. Those small transistors will never supply enough current for the output transistors. ie you are going to blow them.

This is an area where ilimzn can provide a great deal of expert advice.

-Chris
 
OK, sorry for the reaction, it's one of those days :(
For starters, PLEASE give us an ACCURATE schemaitic diagram - so far most currections sugegsted ended up being to things that were not really there.
Regarding Anatech's sugegstions, I do agree with them, and we had SA-100 discussions before :)
There is another thread on a hybrid just below this one, it may prove worthwhile to have a look there too!
 
i get you now. :) basically scrap the driver transistors and put some thing beefier in their place and then ad an extra stage before the driver stage to drive the new beefier driver stage so that it can supply enough current for the output transistors.

I'll leave the tube stage for now, i'm useing one of my old "experiment only" 6922's in there anyway so i'm not really that bothered if it dies.

I'll go have a play around and see what i can come up with.
Many thanks,
Owen
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.