• 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.

High frecuency oscillation when moving volume pot

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Hi!

I'm having a low volume high pitch oscillation depending on the position of the volume pot. It is very funny, but ¿do you know where the Gremlin is hidding?. :cannotbe: Here is the schematic.
 

Attachments

  • kt88 sch 021_resize_resize.jpg
    kt88 sch 021_resize_resize.jpg
    99.9 KB · Views: 624
About the OT...

There is something I don't understand about the Hammond connexion. According to the Hammond schematic, the primary should be connect like this:

RED to B+
BLUE to Plate

But if I do like that then the amp oscilates, so I have to reverse the connections of the primary RED to Plate and BLUE to B+.

¿Maybe I should connect the NFB resistor to the GRN/YEL+GRN 4 Ohm output, and not to the YEL output?
 

Attachments

  • 1627se connection.jpg
    1627se connection.jpg
    10.1 KB · Views: 554
Positive Feedback?

the problem with oscillation is due to inversion of terminals of the secondaries. Instead of providing negative feedback, you´re providing positive feedback.

Swap the secondaries (earth turns "live", "live" turns earth) and you should be able to connect the primary according to the diagram.

I don´t know about the problem with volume setting. The ususal (1k) resistor at the input of the grid is intended to eliminate much higher (unhearable) frequencies.

Let me add that all this info is obtained from this forum. I´m just reposting it.
 
Erik, If I invert the terminals as you say, then the amp osclilates at very high volume with constant pitch (positive feeeback). What I am having now is a very low volume oscillation that only appears when the volume pot is at about 20-30% and then again when i'ts at 80-90%, changing the pitch while I move the pot.
 
It may be because of the layout of the amp rather than the circuit itself, possibly the layout is allowing some positive feedback to occur or allowing some RF pick-up.

Try to effect of a small cap, say 100-500pf in value from the grid of the KT88 to chassis or at the input, across the volume control. Alternatively, place a small cap across the feedback resistor. If these methods are unsuccessful try shielding the ECC83.
 
oscillation

Its possible something in your pre-amp maybe causing some dc to the i/p, changing the operating point of the first tube--Try a decoupling cap at the input socket, and make sure the socket isnt earthed to the chassis at this point--make a star-earth scenario to first tube, then back to main first filter cap in PSU--Best to star-earth the whole lot--Earth loops can be a pain in the xxxx!!--I have a home build 5.1 decoder that does this every now and again (Circle Surround based around AD chip), guess Ill repair it one day!

Another thing to check, are any of the high-current secondary connections from the o/p Tx, near any of the input stage components? Also check the placement of the o/p tube anode wiring for the same
A ferrite bead on the first grid may help,----You have got grid-stoppers havent you??? (150-220ohm direct to tube pin on socket, then use this as the tube grid connection ie, in series with the tube grid)

Hope this helps you, I know what these sort of faults can be like!!

:bawling:
 
I've got it !!

I have the solution :wiz: I've connected the nfb resistor to the 4 Ohm speaker output, (not to the 16 Ohm) and there's no oscillation at all. Maybe that happens because Hammond is using combinations of two separate windings for the different output impedances, so there can be some positive feedback from the induction at that point.
I have also used a star grounding and there's absolutely no hum.:cool: (except a mechanical 50 Hz vibration from the transformer that can be heard when there's complete silence. I will try to find a better one)

This is the first time I can listen to a single ended amp, and I'm impressed. The sound is very "soft"? I don't know what adjectives to use. The high frequencies of the hi-hat or snare do not sound as crispy as I was used to hear, but gentle and soft.

If one of this days I am able to activate my web page, I will show the details of the construction and all the "don't do like this" that I've already made, so other beginners like me will be able to construct this amp with no pain. I recommend it!

My next step will be to plot the curves, but the voltages I read are about 100 V lower than the ones from the original schematic.

_______________Original_____ My amp
AC transf. out:___ 300+300V___ 312+301V
5U4G_out:_____ +380V_______+288V
Choke out:______+375V_______+282V
ECC83 plate:____+360V______ +253V
KT88 plate:_____ +370V______+267V
KT88 grid2:_____+372V______+272V
KT88 cathode:___+37,5V______+21,6V

I have a 300+300V / 300 mA transformer for a stereo amp and I am not using the polypropylene condensers in parallel with the supply electrolytic caps. Any idea about why I have this voltage dropout?
 
what current?

Hi,

Pardon my saying, but I think something is seriously not right here. If your transformer has the same output as the one in the original schematic and your voltages are off by nearly 25% something is amiss. Measure the DCR of your choke and then measure the DC voltage drop across it. Calculate the total current the amp is drawing. Per the original schematic is should be ~210mA with the original voltages. Make sure your power supply and cathode bypass caps are hooked up correctly and that they are in working order.

Maybe not, but my feeling,
Michael
 
transformer

Hi,

I was talking about the power transformer, not the output transformer.

You may not have a problem with your build, after looking at the original schematic a bit more, I see a few things that could cause some voltage differences. For one, there will be increased drop across the 5U4 if the output current is doubled. Add to that the fact that the original transformer is spec'd at 300-0-300 and run at half rated current, which could result in the real AC output voltage being considerably higher than the spec'd value. Those two taken together could probably add to ~60V of the difference seen between your readings and the original's. Add a higher DCR choke and you might have most of the rest.

A different point, the 5U4 is running much closer to it's current limit now and the 100uF input cap could be too large to be conducive to a happy rectifier. The spec sheet shows a suggested value of 40uF for the 5U4GB.

A third point, changing the feedback hookup may have gotten rid of the oscillation, or it may have just moved the frequency up to something you can no longer hear. You should check with a scope if possible. Especially if your tubes are running hotter than you think they should, or your DC current draw is higher than the idle operating point would suggest it should be.

Needless to say, your tubes are operating in a significantly different area of their curves than the orginals were. This could result in significant increase in distortion, as well as a drop in output power.

Not trying to be negative on your build here, just trying to suggest some of the potential problems I see with present situation.


Michael
 
My transformer is not the suggested one. The original is a Tango U-808, but I believe the Hammond 1627SE have similar electrical characteristics; they both have a 2,5K primary impedance.

That the transformers have same primary impedance is not enough to make them interchangeable, leakage inductance and interwinding capacitance probably differs a lot between these 2 transformers and is the reason why you have oscillations in the amp, (The Tango transformer probably have much better high frequency response and therefore lower leakage inductance)

That the oscillation is dependant on the setting of the volume pot suggest that the frequency of the input stage pole at the grid is critical, try by adding a 10kohm resistor between the pot and the first grid, that will lower the pole and hopefully make the amp stable.

If you have a oscilloscope you can then check the square wave response at 10 kHz and look for ringing, if there is ringing you can try adding a small cap ~100pF maybe in parallell with the 1.5kohm feedback resistor.

I really advice you to check the square wave response as the amplifier is clearly very close to oscillation and is bound to have very small phase margin which will effect the sound in a very negative way.

Regards Hans
 
Wow! My first amp is definitely not in a bed of roses :scared:
I was talking about the power transformer, not the output transformer.
Yep, sorry Michael, I was sleepwalking at that time of the night:hypno1:.

I believe you are right, there maybe an ultrasonic oscillation, since the tubes are getting quite hot, but I don't have a oscilloscope to check the wave response. I will try to find a second hand one.

If I connect only one channel, the current is 73 mA for left channel and 72,2 mA for the right channel. When I connect both channels together, the total current drops to 123 mA.

The spec sheet shows a suggested value of 40uF for the 5U4GB
So Hiraga is running the 5U4G out of the specs :yikes:. I will try with 50 uF, since I am using 50+50 caps in parallel. Do you know where can I find the spec sheet for this rectifier?

Thank you very much for all your helpful comments :):) I will be back with the results when I have the oscilloscpope and a tone generator.
 
psu

Hi Manuel,

Do you have the program Power Supply Designer? It can be downloaded for free from http://www.duncanamps.com/ It is very easy to use and has been quite accurate for me in the past.

You can use this to model your power supply, also you should find a link to the datasheet for the 5U4 datasheet there.

I ran a quick model of your supply and your output seems a little low, but not too much. The program does give a warning that you are exceeding the forward current maximum on startup. Most likely due to the large first cap. What is the DCR and the current rating for your choke? Are you sure your power transformer is rated for 300mADC output? The per channel current drop when you connect both channels seems to indicate a further drop in output voltage with larger current demand. Some addt'l drop is to be expected through the rectifier but yours may be a bit higher than the expected. What I would do is make a list of voltage test points for the amp, such as plate, screen, cathode of KT88, plate and cathode of preamp, before choke, after choke, power transformer secondary AC. Then record the values with one channel, both channels, zero signal, full output. For me these measurements, and the current calculations you can do with them, does a lot to tell me if the circuit is operating as expected and where voltage drops are taking place, just a general feel for the amp, if you know what I mean.

I think the power transformer in the original schematic must have actually been 325-0-325. You could go to a solid state full wave rectifier if you want to raise your voltage closer to what is shown on the schematic.

Might as well get this design working well before changing horses to another design. At least then you can give a fair comparison to the two and decide which you prefer.

Michael
 
Hi Dr. Frank, Michael, Jorge, Thank you very much for your replies. I will do all this homework next weekend. It's really getting very interesting. I've never liked "easy listening" music so, why to have an easy made valve amplifier? :firefite:

Might as well get this design working well before changing horses to another design
Yes, you are right. Anyway, it will be very easy for me to fast change to another design since I am using a "rack-mount" assembly with different modules for the amp and the PS. But I will first fight with this one until everything is ok.

When I ordered the transformer I did specify 300 mA. But because it is a noisy transformer (though I did ask for it, at the factory they told me it was not necessary to encapsulate it with epoxy) I will complain and order a new transformer, maybe 325-0-325.
Is there some safe way to calculate the maximum current output of a transformer?

Thank you,

Manuel
 
¿Did I order the wrong transformer?

Hello,

I suspect that the guilty of the voltage drop is the supply transformer. This is the first time I use PSU designer (thanks for the link, Michael) and if I'm not wrong I have the following parameters:

Primary Supply Voltage: 220V
Primary resistance: 5 Ohm

Secondary off-load voltage: 312V
Secondary resistance: 143 Ohm.

With these values, PSUD gives me about the same result as the actual values that I can measure. (325V at C1, 322V at C2)

So, ¿how can I tell the manufacturer of the transformer what do I really need? I see that it's not enough to specify 2x300V/300 mA

Here is the schematic with my latest modifications and measures.
 

Attachments

  • kt88sch.jpg
    kt88sch.jpg
    74.5 KB · Views: 223
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.