• The move to the new server is done. There are some software and database maintenance updates in process. This has us passing the hat around to help out. We appreciate any donations. Seriously, even a dollar helps. The payment page may be found here - https://www.audiokarma.org/support.html

Recapping and restoring a Harman Kardon 330C

onanysunday said:
*idle current adjust: Set function to "aux", volume all the way down, tone controls set at middle. Using DC voltmeter, measure across leads of R428, then set current to 10mV by turning VR402 pot, on other channel, measure across leads of R427 and obtain 10mV by turning VR401 pot.
Click to expand...

This looks like the same procedure for the 430, but it mentions putting an 8 Ohm 50W resistor across the speaker terminals before you do. Is that necessary? And would that be something like this?

https://www.mouser.com/ProductDetai...=sGAEpiMZZMtlubZbdhIBIOzZS54oCXX/1YZYSGnjjuk=
 
iaRIVR said:
This looks like the same procedure for the 430, but it mentions putting an 8 Ohm 50W resistor across the speaker terminals before you do. Is that necessary? And would that be something like this?

https://www.mouser.com/ProductDetail/Vishay-Dale/RH0508R000FE02?qs=sGAEpiMZZMtlubZbdhIBIOzZS54oCXX/1YZYSGnjjuk=
Click to expand...
iaRIVR said:
This looks like the same procedure for the 430, but it mentions putting an 8 Ohm 50W resistor across the speaker terminals before you do. Is that necessary? And would that be something like this?

https://www.mouser.com/ProductDetail/Vishay-Dale/RH0508R000FE02?qs=sGAEpiMZZMtlubZbdhIBIOzZS54oCXX/1YZYSGnjjuk=
Click to expand...
It's not necessary. Just keep a load on the speaker terminals by having your speaker wires hooked up as you normally would when listening to music. 8 ohm 50 watt resistors are an oddball size and almost impossible to find that large in a normal, axial design.
 
iaRIVR said:
I was reading about a restore on a 430 and they said something like:

replace preamp low signal transistors 2SC1344 with BC550C on preamp and phono

Does this part look right to replace the 2SC1344?

https://www.mouser.com/ProductDetail/512-BC550CBU
Click to expand...

Ok, so let's use the datasheets to see how well the BC550 would do in a spec-to-spec comparison:


2SC1344 Hitachi
Application: Low frequency low noise amplifier
Silicon: NPN
Package: TO-92
Pinout: ECB
Vcbo: 30V
Vceo: 30V
Vebo: 5V
Ic: 100 mA
Pc: 200 mW
hfe: 250-1200
fT: 230 MHz
nf: 1-8 dB

BC550
Application: switching and amplifier, low noise
Package: TO-92
Pinout: CBE
Vcbo: 50V
Vceo: 45V
Vebo: 5V
Ic: 100 mA
Pc: 500 mW
hfe: 110-800
fT: 300 MHz
nf: 1.2-4 dB​

They are both NPN and TO-92, so they'll do the same kind of work and fit in the same kind of space.

The voltage-handling specs of the BC550 (Vcbo, Vceo, and Vebo) exceed those of the 2SC1344, so the BC550 will operate in the same place in a circuit.

Both devices can handle the same amount of current Ic and the BC550 can dissipate more heat, so it's a good candidate so far.

The gain of the BC550, depending on the device available from your source, should be in the ballpark of the 2SC1344, and we know they both perform in the audio frequency range, so they'll do roughly the same job.

The point at which they stop amplifying, the fT, is relatively close. So, they'll both amplify the audio signal and both are subject to unintentionally amplifying stray RF frequencies, so there is the same danger of oscillation or motor-boating or overheating with the BC550 as there is with the 2SC1344.

The noise nf that the BC550 adds to the signal is less than that which the 2SC1344 adds, so the BC550 is better for audio work.

The only issue I can see is the pinout. Because your original part is ECB and the BC5nn series is all CBE, you can't just rotate the device 180 degrees when installing. You'd have to twist the leads and therefore insulate them against shorting against each other. Totally do-able, I've had to do this kind of thing and been OK, just I wouldn't recommend it if you had easy access to a device with compatible specs and a ECB or BCE pinout (so you wouldn't have to twist the leads).

So, BC550 can perform fine as a replacement, better in an important area (noise), but the pinout difference makes it a less-than-optimal choice.

Just my hobbyist $0.02 opinion.

Good luck! Let us know what you do and how it turns out!
 
dlucy said:
Ok, so let's use the datasheets to see how well the BC550 would do in a spec-to-spec comparison:


2SC1344 Hitachi
Application: Low frequency low noise amplifier
Silicon: NPN
Package: TO-92
Pinout: ECB
Vcbo: 30V
Vceo: 30V
Vebo: 5V
Ic: 100 mA
Pc: 200 mW
hfe: 250-1200
fT: 230 MHz
nf: 1-8 dB

BC550
Application: switching and amplifier, low noise
Package: TO-92
Pinout: CBE
Vcbo: 50V
Vceo: 45V
Vebo: 5V
Ic: 100 mA
Pc: 500 mW
hfe: 110-800
fT: 300 MHz
nf: 1.2-4 dB​

They are both NPN and TO-92, so they'll do the same kind of work and fit in the same kind of space.

The voltage-handling specs of the BC550 (Vcbo, Vceo, and Vebo) exceed those of the 2SC1344, so the BC550 will operate in the same place in a circuit.

Both devices can handle the same amount of current Ic and the BC550 can dissipate more heat, so it's a good candidate so far.

The gain of the BC550, depending on the device available from your source, should be in the ballpark of the 2SC1344, and we know they both perform in the audio frequency range, so they'll do roughly the same job.

The point at which they stop amplifying, the fT, is relatively close. So, they'll both amplify the audio signal and both are subject to unintentionally amplifying stray RF frequencies, so there is the same danger of oscillation or motor-boating or overheating with the BC550 as there is with the 2SC1344.

The noise nf that the BC550 adds to the signal is less than that which the 2SC1344 adds, so the BC550 is better for audio work.

The only issue I can see is the pinout. Because your original part is ECB and the BC5nn series is all CBE, you can't just rotate the device 180 degrees when installing. You'd have to twist the leads and therefore insulate them against shorting against each other. Totally do-able, I've had to do this kind of thing and been OK, just I wouldn't recommend it if you had easy access to a device with compatible specs and a ECB or BCE pinout (so you wouldn't have to twist the leads).

So, BC550 can perform fine as a replacement, better in an important area (noise), but the pinout difference makes it a less-than-optimal choice.

Just my hobbyist $0.02 opinion.

Good luck! Let us know what you do and how it turns out!
Click to expand...

Ok that is beyond helpful! Thank you for taking the time to explain the differences. It's amazing to see not just matching up spec-for-spec, but to see how each variable will effect the final sound. Maybe I'll get there someday.

I had ordered those because they seemed close and I was putting in an order of caps, I'll see if i feel comfortable trying them out and if so I'll post some pics.

If not I'll order KSC1845 to replace the 2SC1344 and with a KSA992 to replace the 2SA844.

Thanks again dlucy.
 
iaRIVR said:
I had ordered those because they seemed close and I was putting in an order of caps, I'll see if i feel comfortable trying them out and if so I'll post some pics.

If not I'll order KSC1845 to replace the 2SC1344 and with a KSA992 to replace the 2SA844.
Click to expand...

Great!

A couple of tips for when you order again:

It's good to know higher quantities get price breaks for many parts at the well known online vendors. For instance, the KSC1845 is something like $0.26 USD each when you order less than 10 and drops down to $0.06 USD when you order 100 at a time. Depending on your budget and how many pieces of gear you have to work on, the price difference between getting a few now for $3 or $4 versus getting a hundred for just $6 is something to keep in mind.

As you replace devices in a stereo circuit, you'll want to replace them in pairs to keep the operation of that circuit nearly identical in both channels.

If you get to a part of the circuit where you need to match the actual gain of the transistors in both channels of the circuit, having a large number of new replacements for that transistor is helpful. You can use a transistor tester like the Peak Atlas DC55 or DC75 or a cheap component tester to measure you the hfe of each transistor. If you buy just enough of the KSC1845, you may find one of them has a hfe of 431 while the other has a hfe of only 376. Not the end of the world, but if you had 100 KSC1845s because you did the 100 piece discount, you could measure each until you found a pair with very close hfe numbers.

All of the above is over-the-top stuff. Nice if you like, but not required for satisfactory results.

Good luck!
 
Gotcha and thanks! I can't tell from the schematic on the main board, but on the equalizer board for example there are clearly two channels. In this case if the transistors on each side are significantly different (hfe), one channel would be amplified more than the other? I attached a pic with the transistor locations.

I would have no problem buying 100 transistors, but I don't know if I want to buy a tester just yet I'm trying to limit my expenses as much as I can getting started. This is for a 430, with Klipsch KG4 speakers, so not high end but I want to make it sound the best I can. What do you think, go for it or wait until I can match them?

Finally, last question! I need a .47uF 35v +/- 20%, would I be okay using a .68uF 50v +/- 10%? Mouser is out of stock for a month on the .47uf.

Thanks, Mike

open
 

Attachments

  • Equilizer.png
    Equilizer.png
    231.2 KB · Views: 24
In an EQ circuit, the difference will most likely change the frequency response, whether for better or worse I couldn't tell you. The cap you want to replace is Tantalum. I would get rid of them due to failure modes and use a film cap in its place.

This would be a great replacement, but short pins might be a little tough depending on the lead spacing...

https://www.mouser.com/ProductDetail/WIMA/MKS2C034701C00JO00?qs=sGAEpiMZZMv1cc3ydrPrF7l45uRd9dVhXDUZ%2bDt6lMQ=

Or this with longer legs which are easier to make fit.

https://www.mouser.com/ProductDetail/KEMET/R82DC3470DQ60J?qs=sGAEpiMZZMv1cc3ydrPrF512ADDNLQZJtmppGuXf8zI=
 
Perfect, I might just order a couple of each in case the WIMA leads are too short.

Also found a transistor tester for $15 on amazon, looks like it will work fine for at least matching relative values on the transistors.
 
jheu02 said:
This would be a great replacement, but short pins might be a little tough depending on the lead spacing...

https://www.mouser.com/ProductDetail/WIMA/MKS2C034701C00JO00?qs=sGAEpiMZZMv1cc3ydrPrF7l45uRd9dVhXDUZ%2bDt6lMQ=

Or this with longer legs which are easier to make fit.

https://www.mouser.com/ProductDetail/KEMET/R82DC3470DQ60J?qs=sGAEpiMZZMv1cc3ydrPrF512ADDNLQZJtmppGuXf8zI=
Click to expand...

The key to getting long leads on WIMA MKP, MKS, FKP or FKS series caps is to order the cut tape or reel packaging. If you don't order this way, you'll get short 2mm to 5mm leads which make matching the original holes in the PCB difficult.

An example of bulk or ammo pack versus cut tape or reel:

IMG_0012.JPG
 
That's the same with the Kemet R82s...got to look at the datasheet to see which are long lead versions and which are the short pins. Guess I've never looked carefully enough at the Wima datasheet to notice that they had them. I used to use the Panasoni ECQs as they also had a great small form-factor. The Kemets seem just right to me.
 
iaRIVR said:
Has anyone replaced the bias potentiometers, VR401 and VR402? I can't get it narrowed down.

Only specs I can find are:

HK Part No 23531300
1K Ohm, Variable

I think it's on this page:

https://www.mouser.com/Bourns/Passi...e/_/N-blayd?P=1yzmq8cZ1z0zlsdZ1y97o9lZ1yzuvtw
Click to expand...

The trimpots I order are as follows:

Board Component Qty Type Subtype Value
amp board VR401,VR402 2 Resistor Trimpot 2K 652-3352H-1-202LF
tuner board VR302 1 Resistor Trimpot 250K 652-3352E-1-254
tuner board VR202 1 Resistor Trimpot 10K 652-3352E-1-103LF
tuner VR201 1 resistor trimpot 22K 652-3386F-1-253LF
tuner VR301 1 resistor trimpot 4.7K 652-3386G-1-472LF
 
Thank you again Doug!! You've been an incredible help. I was also checking out your Noisy Transistors You Should Replace list and ordered 100 of the KSC1845 for $8 that will work for 2 sets of transistors. Really appreciate all the great advice.
 
iaRIVR said:
Thank you again Doug!! You've been an incredible help. I was also checking out your Noisy Transistors You Should Replace list and ordered 100 of the KSC1845 for $8 that will work for 2 sets of transistors. Really appreciate all the great advice.
Click to expand...

I'm happy to help pay back all the advice AK'ers have given me. :)

The KSC1845's are good replacements in many applications. Having a bunch on hand to do rough gain matching or drop in for any questionable original transistor (where specs are ok) is really helpful.

Buying 10 transistors or electrolytic caps when I only need 2-4 so I get a 40-50% discount has worked well for me.
 
iaRIVR said:
Also found a transistor tester for $15 on amazon
Click to expand...

I picked up one of these (no case is kind of a bummer) and compared the hFE of several transistors with the Atlas Peak DC55.
Both identified NPN or PNP
Both identified open or shorted transistors
Both measured hfe
the Atlas has nicer mini grabbers
the Atlas has a case

The $15 amazon one was about +5hFE higher on all transistors. I'm not sure if this is a problem but i was able to pick the closest ("matching") ones with both tools.

It is much simpler than the 6 way diode test.

I just noticed doug's link to info on cheap testers. I didn't see that before I bought mine but the 5 page thread has similar info.
 
Last edited:
You must log in or register to reply here.
Back
Top Bottom