• 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.

Ahuja HRC-32T Amplifier Oscillating and Distortion at Low Power

AbaddonD

Member
2015-08-20 1:49 pm
Hey guys!

I bought this Ahuja HRC-32T tube amplifier a few months back for cheap and it has this awful oscillation going. It screams with a really high amplitude.

It uses three ECC83s and a pair of EL84s in pentode mode. It has 3 inputs which 2 of them are microphone inputs and a phono input on the back. One ECC83 is used as the pre-amp for the 2 microphone inputs and half of the second tube is used as the pre-amp for the phono input. The 3rd tube is a phase splitter and a tone control tube I believe?

Since I didn't need the 2 mic inputs, I removed the first tube and it's socket to make an amp. The removal of the first tube had no effect on the oscillation.

The original electrolytics were shot, so I replaced all of them. This reduced the hum significantly.
It didn't have a grid stopper resistor, so I added them. This had no effect on the oscillation.
I have also checked all the resistors and they are fine now. When I recieved this, one 110 Ohm screen stopper resistor was open and I replaced both of the resistors with 91 Ohm ones.

The current problems are:

- Oscillation. A 20V RMS signal on the control grids of the 2 EL84s. It goes away when I remove the phono pre-amp tube or when I remove the phase splitter/tone (?) tube. Also, fiddling with the mic 1 input knob kinda messes with the oscillation and I can see the glowing screen on the one tube go dim or brighten up.

- Red screening on one of the EL84s. It moves with the tube, however, I don't think that's the cause of the oscillation since it still happens when I remove that tube (or the other one).

- Amp starts clipping or distorting at like 1W. I think this amp should be able to do 20W.

This amp once made the motor boat sound and popped for a few seconds. Haven't been able to recreate that again.

All the voltages seem OK except the grid having that weird signal going in. (Plate: 337V, screen: 336V)

I have attached a few pictures of the amp.

Thanks for all the help.

20200226_233420.jpg 20200226_232631.jpg 20200226_233427.jpg 20200226_233434.jpg 20200226_233606.jpg 20200226_233619.jpg
 

AbaddonD

Member
2015-08-20 1:49 pm
Does this amplifier use negative feedback? If so, maybe it is connected the wrong way around so the feedback gets positive. You can try to disconnect the feedback to see if the oscillation stops.

Yes. There's a wire connecting the 16 Ohm output to the cathode of the pre-amp.
Removing the wire only reduces the oscillation frequency. The amplitude decreases from 21V RMS down to 19V RMS on the grid of the EL84s.

Edit: There's also this 2.55k resistor between the 16 Ohm output and ground. Don't know what it's for.
 
Last edited:
I once modified a Jolida amp. As part of the mod I replaced the coupling caps with ones that were physically larger. To accommodate them I added solder turrents that were placed unfortunately too close to the input tube's grid. :~( Oscillated at a high frequency (screamed like a Banshee). Tried changing grid stoppers, removing feedback etc. Finally dawned on me what the problem was.

Might pay to double check layout.
 

AbaddonD

Member
2015-08-20 1:49 pm
I once modified a Jolida amp. As part of the mod I replaced the coupling caps with ones that were physically larger. To accommodate them I added solder turrents that were placed unfortunately too close to the input tube's grid. :~( Oscillated at a high frequency (screamed like a Banshee). Tried changing grid stoppers, removing feedback etc. Finally dawned on me what the problem was.

Might pay to double check layout.
I haven't added any solder turrets. I have only changed the dead electrolytics and added grid stoppers. It was oscillating before the resistors.
I can't find anything wrong. :/
 

xenofonn

Member
2013-09-20 11:48 pm
Good afternoon from Greece! First off all,must be clean the amplifier,the dust and moisture acts like resistor between parts,and the circuit acts strange.You can clean all parts with a soft cloth and wd40. After this procedure,check all capacitors and replace especially the power supply electrolytic capacitors. The tube sockets looks like rusty,and worn,the holes,must be cleaned, spraying wd40 and toothpick clean carefuly inside the base holes. A hint ,reverse the polarity of the output transformers,in the anode,and high voltage,and check ,maybe oscillation stops! The schematic is necessary to find any problems,find it before doing anything! These tubes looks like el84,the circuit maybe simple,and you can modified easily! Beware the high voltages!Good night from Greece!
 

AbaddonD

Member
2015-08-20 1:49 pm
Good afternoon from Greece! First off all,must be clean the amplifier,the dust and moisture acts like resistor between parts,and the circuit acts strange.You can clean all parts with a soft cloth and wd40. After this procedure,check all capacitors and replace especially the power supply electrolytic capacitors. The tube sockets looks like rusty,and worn,the holes,must be cleaned, spraying wd40 and toothpick clean carefuly inside the base holes. A hint ,reverse the polarity of the output transformers,in the anode,and high voltage,and check ,maybe oscillation stops! The schematic is necessary to find any problems,find it before doing anything! These tubes looks like el84,the circuit maybe simple,and you can modified easily! Beware the high voltages!Good night from Greece!

Hello!
I will clean the board. I had sprayed the sockets with WD-40 and I'm sure the pins inside are all clean.
The power supply caps are all fine; Rubycon and Nichicon caps BTW.
I have searched for days, but I couldn't find the schematic for this. :(
 

xenofonn

Member
2013-09-20 11:48 pm
Hello!
I will clean the board. I had sprayed the sockets with WD-40 and I'm sure the pins inside are all clean.
The power supply caps are all fine; Rubycon and Nichicon caps BTW.
I have searched for days, but I couldn't find the schematic for this. :(


Hello from Greece! You can the test,change the polarity of the output transformer ,connect the anode cable of output transformer in high voltage, and the high voltage cable to anode,and try. I think is single ended amp. If oscillation stops are you ok! Maybe the amplifier is modded from other person. These cirrcuits are very simple,you can find from internet,the pin out of the tubes,and you can find the proper schematic very easy! In the el84 tube i see the screen grid red,check the resistor of screen,this tube die very fast! Good luck!!
 

AbaddonD

Member
2015-08-20 1:49 pm
Hello from Greece! You can the test,change the polarity of the output transformer ,connect the anode cable of output transformer in high voltage, and the high voltage cable to anode,and try. I think is single ended amp. If oscillation stops are you ok! Maybe the amplifier is modded from other person. These cirrcuits are very simple,you can find from internet,the pin out of the tubes,and you can find the proper schematic very easy! In the el84 tube i see the screen grid red,check the resistor of screen,this tube die very fast! Good luck!!

Hi!
This is a class AB amplifier, I believe, as the tubes have their cathodes tied together and the output transformer has a winding with a center tap.

So I reversed the wires on my output transformer and everything's different. Still a bit of red screening, but the oscillation goes away when the "tone" knob is past like 1.
The red screening tube draws about 50mA idling through it's plate and the other one draws 10mA, testing both with the same winding.
With the transformer completely disconnected, the red screening tube draw around 17mA on the screen and the other one draw 10mA. These are both exceeding their max screen grid dissipation. Might try to put high value resistors on the screens, maybe like 5k?
Also, I measured my output transformer's resistance and it's 30-30 Ohms, is this normal?

Overall, it's a really weird amp. I hope I can fix it :/
 

AbaddonD

Member
2015-08-20 1:49 pm
Update: Somehow, shorting out the 4Ohm output for a second kills the oscillation and the amp does 17W at 8.2Ohms with distortion? I really don't know how that helps. (Also, the negative feedback was connected.)
The amp worked for like 2-3 hours and then started to sound funny. I shut the amp off and one of the EL84s flashed, horribly. Turned it back on and there was a fireworks show inside the tube. Can you guess which tube it was? Wrong. It wasn't the overheating and red screening one, it was the goddamn "good" one!
With a multimeter, I checked the pins and there wasn't a short between any pins, except the ones that are intentional. Seems like the screen is too close to the cathode or something because the fireworks show still happens when I disconnect the anode and it took out the screen stopper resistor with it.

Anyway... I'm very disappointed in this. I think I might need to get another tube and the only EL84 available in Iran is a Trigon brand, it's brand new, but it's around $16. Worth it? My allowance is $10 a month :| xD
I don't know... I might get it later.

Thank you all for commenting and giving me new ideas to try with it. Really appreciate it.
 

32860

Disabled Account
2009-12-06 7:13 am
It is not wise to disconnect the anode only. Doing that makes all the cathode current run through the screengrid. The maximum dissipation of the screengrid will be violated in that way.

In some Dutch electronics magazines from the 50's/60's they used screened cables for the anode leads to the EL84's in push-pull amplifiers to prevent oscillations. The outer mantel of the screened cable is than connected to the B+ so that the voltage difference between the anode leads and the outer mantel does not get too high.

Philips too often used screened cables fo the anode leads of the pentodes in their their single ended radio's. In these radio's the outer mantle of the screened cable is connected to ground.

Did you read the sticky thread "Oscillation in tube amps" written by Tubelab?
 

AbaddonD

Member
2015-08-20 1:49 pm
It is not wise to disconnect the anode only. Doing that makes all the cathode current run through the screengrid. The maximum dissipation of the screengrid will be violated in that way.

In some Dutch electronics magazines from the 50's/60's they used screened cables for the anode leads to the EL84's in push-pull amplifiers to prevent oscillations. The outer mantel of the screened cable is than connected to the B+ so that the voltage difference between the anode leads and the outer mantel does not get too high.

Philips too often used screened cables fo the anode leads of the pentodes in their their single ended radio's. In these radio's the outer mantle of the screened cable is connected to ground.

Did you read the sticky thread "Oscillation in tube amps" written by Tubelab?

When the anode was disconnected, the tube did not red screen as much as it used to when it was oscillating. The other one's screen didn't glow or anything and the calculated dissipation was less than 2W.

I might do that when I get another EL84 in about a month or two.

Do you think a coax cable with the shield being grounded is better for this situation? Cause the case is grounded.
Actually, the "tone" knob is also grounded! I got a pretty nasty shock when my left arm accidentally touched my meter probe while it was in current mode and one lead attached to the B+ and when I grabbed the knob to see if it has any effect. Gotta be a bit more careful, even though I have lots of experience with high voltages; from 300V up to 50KV.

It's a bummer that there's only 1 EL84 seller in here :/
There are more rare military grade tubes (and lots and lots of useless radio and TV tubes) in Iran than common 12AX7, 6L6s, EL34 or EL84. It's weird.
 
Hello!
I will clean the board. I had sprayed the sockets with WD-40 and I'm sure the pins inside are all clean.
The power supply caps are all fine; Rubycon and Nichicon caps BTW.
I have searched for days, but I couldn't find the schematic for this. :(


WD40 is not a contact cleaner, its a water repellant. It contains powerful solvents that may damage many materials, and greases that hang around and I'd avoid using it on electronics. Its designed for such things as engine blocks really.



The filthy state of the PCB is an obvious problem - the PCB may be conductive and need replacing even. Dust and dirt and high voltage are not good companions.
 

32860

Disabled Account
2009-12-06 7:13 am
Do you think a coax cable with the shield being grounded is better for this situation? Cause the case is grounded.

For AC signals B+ is also ground. Because the voltage at the anode of a power tube swings around the B+ voltage (so when B+ = 250Vdc, the AC signal at full power swings from about 50V to 450V), connecting the outer mantel of the shielded cable to B+ prevents the voltage difference inside the shielded cable getting high (maximum 200 V instead of 450V).
 

AbaddonD

Member
2015-08-20 1:49 pm
For AC signals B+ is also ground. Because the voltage at the anode of a power tube swings around the B+ voltage (so when B+ = 250Vdc, the AC signal at full power swings from about 50V to 450V), connecting the outer mantel of the shielded cable to B+ prevents the voltage difference inside the shielded cable getting high (maximum 200 V instead of 450V).
I see.
Well, the max anode voltage in this amp is about 360V when the tubes are cold and about 330V when operational.

Wouldn't connecting the shield to the B+ carry the B+ hum and noise to the other parts of the circuit (the pre-amp)?

I'm also thinking about adding the biggest choke I have which is a 50mH one at about 4A. I don't have anything with higher inductance, but I've used this in my PCL805 amp and it reduced the hum by a factor of 5 or something.

But first, I gotta get an EL84.