Hi guys. I recently built a clone of the amplifier in my Kriesler 11-104, and it sounds great. Single ended, I used EL84s instead of the 6M5s, but otherwise pretty much an exact copy. One problem is that one channel is louder than the other, and obviously so when the balance control is turned to each extreme. I find I have to favour the weaker channel with the balance control often. The difference is not dramatic, just weaker volume (but not extreme) and somewhat less "fullness" and bass (but again not extreme). While it's certainly still listenable, the difference is bugging me and I won't be able to rest 'till I track down the cause. So my question for VR&T is...what would you advise as a troubleshooting procedure? I'm guessing that by measuring enough variables in the circuit and comparing these for each channel I could possibly track down the cause of the difference. Any thoughts? Thanks again for the help I've received in the past. Cheers.

Being a stereo circuit, start by swapping components (e.g. valves) from one channel to the other to see if the problem moves along with them.

Thanks for the response. I've done all that stuff. Swapped valves, including the preamp valve, and also input leads and speakers. It seems like it is a problem in the amp circuit itself, and not valve related.

OK, if you've done all the obvious quick checks, then it's time to use traditional diagnosis approach. Start with the power supply and make sure all voltages are present and correct. Then, using a test tone I work forward from the input stage to the output stage checking/comparing signal amplitude on each channel with my CRO.

I don't have a schematic for the 11-04. Do you?

Where did you source the parts for this amp? If you're using recycled Kriesler parts, then be aware that their output transformers can be dodgy. It may pay to swap connections to those under the chassis and see if the problem changes channel.

Agree on Kriesler OP Transformers: Give me ten of their radios, 2 might have a good one.

I would be comparing voltages & look for a decimal error in a part, wrong part, wired wrong. I normally reach for a CRO & generator in situations like this. Voltage checks first.

Thanks guys. I've got a schematic. All the parts are new. The PT and OTs are from Hammond. Unfortunately I don't have a CRO. So I'll check and compare voltages first at all points, as well as a visual inspection for dodgy solder joints etc. Also I'll be wanting to see if the values of corresponding components (capacitors and resistors) in each channel are within the limits of tolerance you would expect. What's the normal procedure for this? When do you have to remove the part from the circuit, and when is it ok to test in situ. I'm guessing the schematic will tell me this, but it's good to know what others might do. I'll get into this after the school holidays (kids and high voltages don't go together), and report back. Thanks again for the help. Cheers.

The generalisation with resistors is, that if they test high in circuit then they won't be any better out of it. So they need further attention. If a resistor is on the cathode with an electroytic across it. they have to be separated.

Plate & grid resistors when capacitive coupled, can normally be tested in circuit. Plate resistors on the det / first audio have an attrition rate, as do grid resistors on 6V6 and family. Normally one tests them as they change the wax paper caps, that way you eliminate one potential issue, before you knew you had one.

New parts can fail. Do make sure that the component values in amp "A" are the same in amp "B".

I always check new work & I have had a repaired HMV recently that would not run due to a wiring error & 2 NP caps went, one was left in & one was wrong & I tossed 5 dud resistors & replaced the OP valve with the correct one. That one never got powered, until repaired, as the assessment visual inspection, spotted the error.

What did you do with the tone control?

Sounds like a capacitor of the incorrect value was used in one channel.

Or maybe a resistor of the wrong value. Maybe a 270K resistor where a 470K should be. Voltage measurements comparing one channel to the other may point you towards such an error.

To also be considered is the colour code in respect of colour blindness & conformity. I have had a shot at Dick Smith (before it went totally feral & tried to be Harvey Norman) about several of their resistors where the colour code was not standard for particular colours & that can lead to errors. 26% of males have colour vision issues.


I have seen a couple of sets where I think that was part of its problem

Marc

To also be considered is the colour code in respect of colour blindness & conformity

To double check, I always measure resistors before using them. More and more these days the poorly-chosen body colour of resistors makes it damn hard to discern the code colours.

I still have some carbon resistors in stock from the 1970s and 1980s. They have a fawn body colour and are dead easy read, although their actual values now will be a lottery.

If you don't have a scope but do have a stable source of audio tone, may I suggest the following technique?

1. Hook audio generator up to both channels in parallel.
2. Wire a 47nF capacitor in series with your meter and switch it to AC volts.
3. Starting at the inputs, compare AC voltages between channels at each stage (pots, grids, anodes) in each amplifier.
4. You will quickly find the missing gain.

If you don't have an audio generator, you could use about 3 volts AC from a transformer. Even the amp's own heater supply would suffice.

Thanks for the replies. I think I did test components before wiring them in......but then again, maybe not well enough. Ian..thanks for the advice on measuring AC voltages. I'm new to all this (being a mere musician), and I'm particularly weak on diagnostic techniques. I'll get around to putting these things into practice in a day or two. Cheers.

Another thought. Is the volume pot a dual gang unit? I've found that they often don't track very well at low volume settings. Try reversing the wiper connections of the pot (if you are using one) and see if the discrepancy swaps.

I think I've worked it out. I used a log taper pot for the balance control (I know......I know...........dumb thing to do.........). I don't understand why I got unevenness in both extreme positions, but anyway, now I have a linear taper in there it seems to be working fine. I checked signal voltages with a 47nf capacitor as suggested by Ian above and they all matched. That's a handy technique I'm sure I'll use again. Thanks all.