As the title says, I have a ha-330 with quite a lot of dc on the outputs. It has 215mV and 250mV. I’d like to be able to identify the problem without replacing a lot of random parts until I get it. The amp still plays music "fine" but obviously not at its best right now. Could somebody help/guide me through this ? I verified a couple things but there’s not an obvious problem that I can fin. What would be a good procedure to follow in order to find the problem ?
I checked the voltages at all the transistors following the schematic. Most of them are a little off but nothing seems completely wrong. The voltages coming from the power supply are like 1V higher than what they should. All the electrolytic caps are new and good quality.
Thanks !
I checked the voltages at all the transistors following the schematic. Most of them are a little off but nothing seems completely wrong. The voltages coming from the power supply are like 1V higher than what they should. All the electrolytic caps are new and good quality.
Thanks !
Possibly leakage in C705. Use a high quality part, preferably bipolar with 35V rating.
This offset won't harm your speakers at all.
This offset won't harm your speakers at all.
Then possibly related to the input pair, Q701/Q702. Before replacing, try removing any dust on them.
Last edited:
These voltages may be within normal design variation:
nominal Q701 and Q702 emitter current is about 2 mA into each transistor. Beta is spec'd at 160 to 500, so base current could range between 12.5uA and 4uA. Therefore, expected base voltage at Q701 developed across R703 could variy from 1.25V to 0.4V Similarly, bias voltage drop across R716 could vary from 0.275V to 0.088V. Assuming no offset between the transistor bases, and equal beta, output voltage would range between (1.25 - 0.275) = 0.975V to (0.4 - 0.088) = 0.312V.
These estimates don't match what you measure, but they're not way off and allowing for mismatched beta and offset voltage error in the input pair could bring the estimates within reason. The point is don't assume faulty parts are present. Measure drop across R707. If it's less than 1mV, cap leakage is unlikely.
nominal Q701 and Q702 emitter current is about 2 mA into each transistor. Beta is spec'd at 160 to 500, so base current could range between 12.5uA and 4uA. Therefore, expected base voltage at Q701 developed across R703 could variy from 1.25V to 0.4V Similarly, bias voltage drop across R716 could vary from 0.275V to 0.088V. Assuming no offset between the transistor bases, and equal beta, output voltage would range between (1.25 - 0.275) = 0.975V to (0.4 - 0.088) = 0.312V.
These estimates don't match what you measure, but they're not way off and allowing for mismatched beta and offset voltage error in the input pair could bring the estimates within reason. The point is don't assume faulty parts are present. Measure drop across R707. If it's less than 1mV, cap leakage is unlikely.
Are the A & B rail voltages matched in amplitude (& of course with opposite signs)?
If not matched, how close are they?
The input bases see 100k and 22k DCR, not very similar. If the beta of either or both transistors drops over time,
the base current(s) and the DC voltage between the bases will increase, along with the output DC offset.
Suggest that you parallel R703 temporarily with a 27k resistor to make it 22k. Does the offset decrease?
the base current(s) and the DC voltage between the bases will increase, along with the output DC offset.
Suggest that you parallel R703 temporarily with a 27k resistor to make it 22k. Does the offset decrease?
Last edited:
So a higher beta will decrease the voltage at the output ? Do they need to be matched together ? (Q701 and Q702)
In this case, high DC beta is more important to minimize base currents, and hence DC voltage drops across the resistors,
but matching is also good.
but matching is also good.
I suggest taking DVM measurements at bases of Q701, Q702, and junction R707- C705, and output at L701. Careful measurement should reveal a lot of what's going on. Make sure meter negative probe is consistently connected to local ground, eg. at bottom of R703.
BTW, C705 seems to have incorrect polarity installed.
BTW, C705 seems to have incorrect polarity installed.
It’s late now but i’ll take those measurements tomorrow. Regarding C705, I installed the new caps the same way as the originals were. The amp behaved exactly the same before and after the recap.
Those are the measurements I did of the original Q701/702. Left and right measure about the same, but there’s a big mismatch within each pair
Attachments
I need to find a good replacement for the 2SA836. BC557BTA looks about right but the legs aren’t in the same order…
What happens with DC offset when you turn the subsonic filter on / S701 open? No change in offset?
I agree with the posts above about the Q701 and Q702 base voltage measurements - obtain good accurate measurements with the mVmeter and post them here. Clean the probes/multimeter sockets and establish a good ground connection.
The solder residue could cause issues.... wash the PCB really well with isopropyl alcohol and scrub hard with a firm toothbrush. Then rinse it with a can of pure isopropyl alcohol.
Did you change C903 as well? For the sake of faultfinding, you could lift one side of R916 and then check the offset...
What is the offset reading with no speakers connected?
Is the DC offset stable with temp change; what is the reading when the amp is cold?
The starting offset on my amp is around 200mV...(but it does settle to less than +/-10mV in some 15-20min). It is audible (when I turn the mute on my DAC On/Off -> I can hear a very faint "plop" from the speakers.) - but not that much of an issue... really. If you can't find the root cause it might be okay to just leave it alone...?
I agree with the posts above about the Q701 and Q702 base voltage measurements - obtain good accurate measurements with the mVmeter and post them here. Clean the probes/multimeter sockets and establish a good ground connection.
The solder residue could cause issues.... wash the PCB really well with isopropyl alcohol and scrub hard with a firm toothbrush. Then rinse it with a can of pure isopropyl alcohol.
Did you change C903 as well? For the sake of faultfinding, you could lift one side of R916 and then check the offset...
What is the offset reading with no speakers connected?
Is the DC offset stable with temp change; what is the reading when the amp is cold?
The starting offset on my amp is around 200mV...(but it does settle to less than +/-10mV in some 15-20min). It is audible (when I turn the mute on my DAC On/Off -> I can hear a very faint "plop" from the speakers.) - but not that much of an issue... really. If you can't find the root cause it might be okay to just leave it alone...?
Hi Dinolobe,
I wasn't finding fault with your replacement technique; rather I was noting that if Q702 base voltage is in fact PLUS a few hundred mV, the cap design polarity is backwards--- a design error. I'm reaching a bit here, but a backward C705 might present excess leakage current which would tend to raise output voltage. Check for DC voltage across R707; any DC present will be amplified by stage gain, i.e. 22k/470 = ~47
PS. Another way of making the same point: C703 polarity seems correct. C705 polarity assignment should be the same.
Last edited:
The subsonic filter has no effect on the dc offset
C903 was replaced as well as every other electrolytic capacitors
DC offset gets to 157mV and 159mV with one side of R916 lifted
Bases voltages :
Q701L : 188mV
Q702L : 165.5mV
Q701R : 288.2mV
Q702R : 280.4mV
I was able to bring the offset down to around 80mV on the left and 165mV on the right. After reconnecting R916, now both channels are around 155mV and 160mV. Really weird 😅
Just to confirm my interpretation:
Right Channel output: 165mV with no load, 160 mV with R916 installed.
Left Channel Output: 80mV with no load, 155mV, again with R916 installed.
This is very disconcerting. Connecting R916 applies about 4mA from output to the B supply rail (-40V), yet it results in an additional output 80mV DC offset??? This suggests the DC output impedance is 80mV/4mA = 20 ohms! Further R916 is the Left sense resistor, but R115 is the analogous sense resistor for the right channel. Ideally, both channels should be immune to these sense connections. (I recommend choosing a consistent reference point for these mV resolution measurements to avoid uncertainties, eg. bottom of R703L for left channel.)
[A rant directed at Hitachi: this is an appalling excuse for a schematic. The box depicting/enclosing the right channel is a particular irritation as it conflates with the ground buss. End rant.]
I wonder if the bias current adjustment is working as specified on page 4 of manual? Would you confirm?
Thanks.