HCA 1200 ii bias and output relays

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
I opened up the amp to try and troubleshoot the output relays which are staticy at low volume and that brought up some other questions. The left channel heat sink was a lot hotter than the right, so I investigated the bias. I ended up adjusting to 10 MV from 8 on the left channel and 9 on the right channel. That doesn't explain the temp difference. Anyone have any idea why that would be? The amp sounds good and balanced to my ear- except for the random static at low volumes till the relays make contact. I also noticed that the output relays are in a tricky spot to work with, and looking online they are discontinued- they are

OMRON G4F1123TUSDC12

Has anyone had any success cleaning them? I could use some old-timer trick :)

I know they can be redone by sending the amp out, but I am out of work and on a severe budget crunch! Not to mention I enjoy fiddling!

Thanks,
Peter
 
Different bias between two inverted sets of output transistors.

So I bought used HCA-1200 MK II and it had one issue. Just after I turn it ON there was no signal in left channel. I tried to shake it slap it and suddenly paw! left channel works so I assumed some loose contact.. Well next time when i started cold - same thing no sound in left channel but this time I let it warm up and voila left channel works by it self. I did several times so loose contact became less probable... I thought maybe bias ?
Also i felt there was some difference between left and right channel so I opened it and measured bias on both channels. Then I tried to adjust it a bit and here's what I have found on left channel.

Since I don't have schematics I was just measuring blindly what is what...
And there are two sets of emitter resistors 2 x 5 each 0.33 Ohm. They are inverted on one set rail is (+) and other set of 5 has negative (-) rail.

As far as I understand bias should be identical on both rails but one rail (one on the back of the amp with resistors Q116-Q120) about 2mv higher. Difference seems to be in % of rail value as if I set bias lower difference will be less and is I'll go up with bias difference between those sets also inferences.
On the right channel is much better - difference is is much less after calibration is about 9,9mV on + rail and 10,1 on - rail.

I also noticed that left channel not working when amp is cold is most probably causes by bad contactor ob the back of the amp - if i slap it directly sound comes back to left channel...


My question is - why there is big difference between both bias sections on left channel ? What could be the reason ? Is there any chance this contact switch on the back (input board) has anything to do with it ?


cmJedFbptDm3V9hq0_1Q3FAS_0flfHBX6ee-tlsF1dIk=w1458-h1348-no

KscJQ_NpKEOge631_l-peoWold2_gsOkhndPGB7w__T2=w1469-h1348-no
 
Did you take voltage readings on all 0.33R resistors?

(9.4*5)/4=11.75, quite close to 11.9, that can indicate one of the output transistors is not working. The voltage difference on Re varying in certain percentage with the bias setting also suggests the same possibility.
 
Different bias between all transistors on left channel

So as advised I have measured all resistors and bias on all emitters.
Result is a puzzle to me. It's different between all transistors! Differences are huge.
Since they are sourced from common rails and only thing that I can see there are resistors on collectors and emitters that I measured to be ok and of same value. So if resistors are fine does it mean transistors are bad ? But not 100% bad just slightly damaged ? If yes - then which ones ?

Any suggestions ?
Here is a table with measurement results:
Pomiary%2Bxls.JPG
 
I opened up the amp to try and troubleshoot the output relays which are staticy at low volume and that brought up some other questions. The left channel heat sink was a lot hotter than the right, so I investigated the bias. I ended up adjusting to 10 MV from 8 on the left channel and 9 on the right channel. That doesn't explain the temp difference. Anyone have any idea why that would be? The amp sounds good and balanced to my ear- except for the random static at low volumes till the relays make contact. I also noticed that the output relays are in a tricky spot to work with, and looking online they are discontinued- they are

OMRON G4F1123TUSDC12

Has anyone had any success cleaning them? I could use some old-timer trick :)

I know they can be redone by sending the amp out, but I am out of work and on a severe budget crunch! Not to mention I enjoy fiddling!

Thanks,
Peter

I had similar issue with my contacts. when i measured resistance on this relay it showed that when it's closed the resistance is starting from around 50 Ohm on one channel and maybe 15 on another then slowly goes down as contacts are pushing... but it didn't went to 0 it was about one to 2 Ohm on right channel and about 4-5 Ohm on left.

I have removed output board, opened relays covers and bent a little bit contact plates so they don't touch each other but they are not fully open. After that resistance came down to below 1 Ohm on both channels but still not to 0. So i took sandpaper of 600 grade cut thin strip on on amp powered when contacts are closed i cleaned a bit both sides of contacts. After that all was working fine resistance came down do 0 (0,2 Ohm actually but my multimeter cables are 0,1 Ohm each). Now every time relay is closing immediately there is zero resistance on contacts.

Hope this helps.
 
Administrator
Joined 2007
Paid Member
Those readings look normal. I notice they increase as you work down the list which could be just the normal drift/increase in current as the amp warms. Try monitoring just one value from cold and see how it changes.

It all normal though. (you can't measure resistance with the amp on. Any voltage across the resistor totally confuses the DVM)
 
All measurements ware made on amp fully warmed up (over an hour). I used 3 multimeters at the same time. All those ware stable when I took a note of them. I even tripe checked them back and forth to be sure they are stable before putting this data in table.

Also I have monitored 3 transistors with most different values from beginning (amp cold). From very beginning they had similar difference until it got hos and stable.

Right channel is almost perfectly ok - all transistors have same value when hot and stable and they rise from cold in thecsame paste.
 
Well considering that right channel has same bias across all transistors (+/- 0,2 mV) and that on right channel there is significant difference between lowest transistor and highest (5,6mV and 11mV - that's twice the difference) I say something is not OK.

I have set left and right channel using average values of bias so both of them are about 11mV at 228V power supply and fully warm. But that leaves me with some transistors on left channel raching 18mV when other have way below 10mV.
Also I can hear difference in sound between both channels. Let channel seems to have higher pitch....

I'm sure it's not OK. Question is what could it be ? If all resistors around those transistors seems to be fine is it possible some of transistors need's replacements ?
 
Administrator
Joined 2007
Paid Member
Transistors vary enormously one to another, even from the same production run. Look up any data sheet for any device and you will see minimum and maximum values for hFE (the gain or beta). Unless those are matched between paralleled devices then you will get the effects you observe. Its not always just hFE that needs matching either.

The transistors in your amp will be 100% fine, they don't change or deteriorate, and any failure of a power device is catastrophic... they typically go short circuit.

Another factor in a stereo amp is that the heatsinks will not be identical in their ability to dissipate heat due to non symmetry of surrounding items, the location of transformers etc.

Differences between channels... if you suspect anything is amiss then you must do measurements of frequency response, transient response on squarewave testing and so on. Its easy to think there is a difference when perhaps there is not.

So you need to prove that. Nothing can alter pitch (frequency) but you could have a situation where the frequency response is not accurate. That's unlikely in a power amplifier though, more likely in a preamp where some user control or switch is damaged and giving different settings one channel to the other.
 
Don't get me wrong but I have to counter all what you said ;]

I fully understand that transistors are not the same even from one run - it's natural. But you have to notice that differences are 2x between some of them.
Also those supposed to be matched ones. And comparing n left and right channel I see that right one has all the values exactly same across all transistors (+/- 0,2mV only variation) where left channel shows huge differences (5,5mV or more)
This amp according to service manual needs to be set at 10mV bias idle +/- 2mV and manufacturer procedure requires using one point of measurement to set bias properly - that means they assume bias between all transistors should be the same.

If I would accept what you said it would mean that they (Parasound) matched transistors on one channel only and they did not care at all about other channel. Do you really think it's possible ?

As for heatsink and other factors. This amp is 100 symmetric construction. Heatsink is perfectly uniform for each channel and it's common for all transistors. If heatsink/construction would be to blame here - there would be a pattern. For example transistors closer to middle of heatsink should be experiencing higer bias. But what I het is the middle one is higher, next one is much less and one close to one edge is highest of them all where one on opposite edge is showing again lower value - and I remind you that this is symmetric build.
Also here I can point out that in symmetric build between left and right channel - only left is experiencing this phenomenon.

As for measurements why you assume I did not measure that ?
Ok to be honest in that case I have not measured the response as audible effect was clear enough. I have used same source between both channels injecting several types of noise. I also performed blind test on that to be sure what I'm hearing is correct.
If I'll find more time I will measure those responses properly but it doesn't change the fact that there is difference.

I feel that you treat me bit patronizing if not to say not seriously. That's not helping at all. I know what I'm doing and this is not some imaginary issue. As I know electronics this situation does not make much sense to me so I'm looking for some help, not undermining my opinions. If I say there is audible difference - there is one.
 
Administrator
Joined 2007
Paid Member
I feel that you treat me bit patronizing if not to say not seriously. That's not helping at all. I know what I'm doing and this is not some imaginary issue. As I know electronics this situation does not make much sense to me so I'm looking for some help, not undermining my opinions. If I say there is audible difference - there is one.

Not at all :) I'm just trying to take the middle ground without knowing your abilities or test gear available and so on. And I'm using my experience and intuition as to why the bias is unequal between the devices.

If you want to determine exactly why it behaves like this then you are going to have to do some more detailed measurements along the lines of checking each like for like output device (the npn's and pnp's) for forward current gain and vbe in relation to collector current. I think you would find there are significant differences in characteristic between them.

You could (and I wouldn't really recommend doing this) remove all the output devices from both channels and then try and match and pair them.

Two squarewave tests, one at a low frequency of say 100Hz and one at 10kHz would immediately show any problems with overall response.
 
Transistors apparently are not sharing current evenly that way. This amp might have been repaired before that power transistor replacement was involved yet matching was skipped. If you can tell the difference by listening against the "good" channel, I'd definite go ahead replace them with a matched set, I mean, what alternative there is otherwise?

As of transistor consistency in a batch, I don't know about MJL21193/4, but my tubes of 25x NJL3281/1302 have Vbe within a couple mV, and beta in range of 105-125, more than satisfactory for me. I did test them on a large heatsink, and at Ic of ~100mA.
 
These resistors have little chance if any to deteriorate. Even at an impossible continuos 1A operating current they are dissipating 0.33w each, perhaps not much of a heat to deteriorate over? Then again, this theory could be easily tested by rotating the suspected resistors.
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.