Tweet
Back in August 2017, one of my steady clients sent over a GK 800RB with the complaint of LF distortion. With the amp opened up on my test bench, my initial findings was the asymmetrical clipping of Q89 J113 JFET circuit. I had adequate head-room ahead of that stage, but, with the Boost control turned up and any significant LF EQ added, that stage would clip easily. Lowering the Gain control and compensating with the Woofer Master Vol could compensate, but I went ahead and tweaked the J113 circuit to get it as symmetrical as I could.
Patched the oscillator into the Effects Return jack, I trimmed the drain resistor R91 (4.7k) with an 8.2k resistor, which got the clipping symmetrical (with THIS JFET)…value lowering R91 to 2.99k, and the results looked better. As I lowered my oscillator frequency, I was still seeing LF clipping as I increased the input level. Looking at R73 & C70 (12k & 100nF), it’s LF corner is 133Hz. I increased C70 to 470nF, yielding 28.2Hz. Looking at C99 & R94 (10uF & 100 ohms), which is the RC network from Q89’s Source terminal to ground, that yielded a LF corner of 159Hz. I increased C99 to 47uF, yielding a 34Hz corner. Looking at C95 output cap from Q89’s Drain, loaded by two 50k pots and the two resistors R101 & R106, it yielded a LF corner of 45Hz. I increased C95 to 470nF, now yielding a LF corner of 21Hz.
Now, patching into the input and checking the response, I found I needed to increase C42, which was 100nF, and showing poor LF response. I placed a 10uF NP cap across it, and at C61 (220nF), I placed a 22uF NP cap across it, and a 22uF NP cap across C85, which feeds the LF Power Amp. The LF corner network in the feedback circuit for the power amp, C3 & R93 (10uF & 1k) yields 15.9Hz, so I left it as is.
Checking the results both with an oscillator, as well as bass plugged in, I wasn’t hearing any LF distortion, all now sounding solid. I fetched one of CenterStaging’s GK 800RB for comparison, and it sounded VERY distorted on the low end when pushing the input gain, LF boost, Contour and Boost control.
After sending the amp back to the client, it came back with complaints about the filter switches, and then later about noisy pots. Old amp, suffering from the usual oxidation problems that set into the switches and pots. I ended up ordering/replacing the 5 push-button switches, and, in the installation of them, placed both sets of contacts in parallel for redundancy (GK’s PCB layout didn’t bother with that). The pots…as the mechanical design of their bushings, left them as a part you couldn’t replace directly with any part from current distributor inventory. So, one by one, I removed the pots, pulled each apart, cleaned and lubed the resistive track brushing on Caig DeOxit, re-assembled, and now noise-free switches and pots.
The client has been really happy with the end results, and had me do the same on their other 800RB. The GK 800RB’s in CenterStaging’s inventory are still stock. I realize these cap changes directly affect the LF distortion character added to the tone, as a function of how much gain you dial in. Increasing the cap values to lower the corner frequencies plays with that character.
A common problem with aging GK 800RB’s is the top and bottom panel mtg hardware, as well as the side panels. Most of the # 6-32 FHMS’s had stripped out the holes in the 16ga aluminum. I’ve increased the hole size to # 8-32, using undercut # 8-32 FHMS’s, so everything was again tight fitting.
I had later looked at the schematic of a 1001RB II preamp, and saw many of the same changes in place. The resistor values around the JFET on it are approaching an order of magnitude higher….something I haven’t yet tried.
Preamp-TL604 Switch version.pdf
Power Amp schematic.pdf
Preamp Schematic Rev A.pdf
Patched the oscillator into the Effects Return jack, I trimmed the drain resistor R91 (4.7k) with an 8.2k resistor, which got the clipping symmetrical (with THIS JFET)…value lowering R91 to 2.99k, and the results looked better. As I lowered my oscillator frequency, I was still seeing LF clipping as I increased the input level. Looking at R73 & C70 (12k & 100nF), it’s LF corner is 133Hz. I increased C70 to 470nF, yielding 28.2Hz. Looking at C99 & R94 (10uF & 100 ohms), which is the RC network from Q89’s Source terminal to ground, that yielded a LF corner of 159Hz. I increased C99 to 47uF, yielding a 34Hz corner. Looking at C95 output cap from Q89’s Drain, loaded by two 50k pots and the two resistors R101 & R106, it yielded a LF corner of 45Hz. I increased C95 to 470nF, now yielding a LF corner of 21Hz.
Now, patching into the input and checking the response, I found I needed to increase C42, which was 100nF, and showing poor LF response. I placed a 10uF NP cap across it, and at C61 (220nF), I placed a 22uF NP cap across it, and a 22uF NP cap across C85, which feeds the LF Power Amp. The LF corner network in the feedback circuit for the power amp, C3 & R93 (10uF & 1k) yields 15.9Hz, so I left it as is.
Checking the results both with an oscillator, as well as bass plugged in, I wasn’t hearing any LF distortion, all now sounding solid. I fetched one of CenterStaging’s GK 800RB for comparison, and it sounded VERY distorted on the low end when pushing the input gain, LF boost, Contour and Boost control.
After sending the amp back to the client, it came back with complaints about the filter switches, and then later about noisy pots. Old amp, suffering from the usual oxidation problems that set into the switches and pots. I ended up ordering/replacing the 5 push-button switches, and, in the installation of them, placed both sets of contacts in parallel for redundancy (GK’s PCB layout didn’t bother with that). The pots…as the mechanical design of their bushings, left them as a part you couldn’t replace directly with any part from current distributor inventory. So, one by one, I removed the pots, pulled each apart, cleaned and lubed the resistive track brushing on Caig DeOxit, re-assembled, and now noise-free switches and pots.
The client has been really happy with the end results, and had me do the same on their other 800RB. The GK 800RB’s in CenterStaging’s inventory are still stock. I realize these cap changes directly affect the LF distortion character added to the tone, as a function of how much gain you dial in. Increasing the cap values to lower the corner frequencies plays with that character.
A common problem with aging GK 800RB’s is the top and bottom panel mtg hardware, as well as the side panels. Most of the # 6-32 FHMS’s had stripped out the holes in the 16ga aluminum. I’ve increased the hole size to # 8-32, using undercut # 8-32 FHMS’s, so everything was again tight fitting.
I had later looked at the schematic of a 1001RB II preamp, and saw many of the same changes in place. The resistor values around the JFET on it are approaching an order of magnitude higher….something I haven’t yet tried.
Preamp-TL604 Switch version.pdf
Power Amp schematic.pdf
Preamp Schematic Rev A.pdf
Comment